X-Git-Url: http://nsz.repo.hu/git/?a=blobdiff_plain;f=ldso%2Fdynlink.c;h=7ac0bf7552bc2aac596ec6ffb24f23217714202a;hb=d0209a2a334ff714063b70dc69c1fea131304836;hp=025ed1b0c81eae20d5b01593393c17931b23d944;hpb=ba18c1ecc6a18203ad8496791154af86f706f632;p=musl

diff --git a/ldso/dynlink.c b/ldso/dynlink.c
index 025ed1b0..7ac0bf75 100644
--- a/ldso/dynlink.c
+++ b/ldso/dynlink.c
@@ -1,4 +1,5 @@
 #define _GNU_SOURCE
+#define SYSCALL_NO_TLS 1
 #include <stdio.h>
 #include <stdlib.h>
 #include <stdarg.h>
@@ -29,6 +30,9 @@ static void error(const char *, ...);
 #define MAXP2(a,b) (-(-(a)&-(b)))
 #define ALIGN(x,y) ((x)+(y)-1 & -(y))
 
+#define container_of(p,t,m) ((t*)((char *)(p)-offsetof(t,m)))
+#define countof(a) ((sizeof (a))/(sizeof (a)[0]))
+
 struct debug {
 	int ver;
 	void *head;
@@ -69,14 +73,19 @@ struct dso {
 	char relocated;
 	char constructed;
 	char kernel_mapped;
+	char mark;
+	char bfs_built;
+	char runtime_loaded;
 	struct dso **deps, *needed_by;
+	size_t ndeps_direct;
+	size_t next_dep;
+	int ctor_visitor;
 	char *rpath_orig, *rpath;
 	struct tls_module tls;
 	size_t tls_id;
 	size_t relro_start, relro_end;
 	uintptr_t *new_dtv;
 	unsigned char *new_tls;
-	volatile int new_dtv_idx, new_tls_idx;
 	struct td_index *td_index;
 	struct dso *fini_next;
 	char *shortname;
@@ -116,16 +125,21 @@ static int runtime;
 static int ldd_mode;
 static int ldso_fail;
 static int noload;
+static int shutting_down;
 static jmp_buf *rtld_fail;
 static pthread_rwlock_t lock;
 static struct debug debug;
 static struct tls_module *tls_tail;
 static size_t tls_cnt, tls_offset, tls_align = MIN_TLS_ALIGN;
 static size_t static_tls_cnt;
-static pthread_mutex_t init_fini_lock = { ._m_type = PTHREAD_MUTEX_RECURSIVE };
+static pthread_mutex_t init_fini_lock;
+static pthread_cond_t ctor_cond;
+static struct dso *builtin_deps[2];
+static struct dso *const no_deps[1];
+static struct dso *builtin_ctor_queue[4];
+static struct dso **main_ctor_queue;
 static struct fdpic_loadmap *app_loadmap;
 static struct fdpic_dummy_loadmap app_dummy_loadmap;
-static struct dso *const nodeps_dummy;
 
 struct debug *_dl_debug_addr = &debug;
 
@@ -269,12 +283,16 @@ static Sym *gnu_lookup_filtered(uint32_t h1, uint32_t *hashtab, struct dso *dso,
 #define ARCH_SYM_REJECT_UND(s) 0
 #endif
 
-static struct symdef find_sym(struct dso *dso, const char *s, int need_def)
+#if defined(__GNUC__)
+__attribute__((always_inline))
+#endif
+static inline struct symdef find_sym2(struct dso *dso, const char *s, int need_def, int use_deps)
 {
 	uint32_t h = 0, gh = gnu_hash(s), gho = gh / (8*sizeof(size_t)), *ght;
 	size_t ghm = 1ul << gh % (8*sizeof(size_t));
 	struct symdef def = {0};
-	for (; dso; dso=dso->syms_next) {
+	struct dso **deps = use_deps ? dso->deps : 0;
+	for (; dso; dso=use_deps ? *deps++ : dso->syms_next) {
 		Sym *sym;
 		if ((ght = dso->ghashtab)) {
 			sym = gnu_lookup_filtered(gh, ght, dso, s, gho, ghm);
@@ -299,6 +317,11 @@ static struct symdef find_sym(struct dso *dso, const char *s, int need_def)
 	return def;
 }
 
+static struct symdef find_sym(struct dso *dso, const char *s, int need_def)
+{
+	return find_sym2(dso, s, need_def, 0);
+}
+
 static void do_relocs(struct dso *dso, size_t *rel, size_t rel_size, size_t stride)
 {
 	unsigned char *base = dso->base;
@@ -349,7 +372,7 @@ static void do_relocs(struct dso *dso, size_t *rel, size_t rel_size, size_t stri
 			sym = syms + sym_index;
 			name = strings + sym->st_name;
 			ctx = type==REL_COPY ? head->syms_next : head;
-			def = (sym->st_info&0xf) == STT_SECTION
+			def = (sym->st_info>>4) == STB_LOCAL
 				? (struct symdef){ .dso = dso, .sym = sym }
 				: find_sym(ctx, name, type==REL_PLT);
 			if (!def.sym && (sym->st_shndx != SHN_UNDEF
@@ -376,7 +399,7 @@ static void do_relocs(struct dso *dso, size_t *rel, size_t rel_size, size_t stri
 		tls_val = def.sym ? def.sym->st_value : 0;
 
 		if ((type == REL_TPOFF || type == REL_TPOFF_NEG)
-		    && runtime && def.dso->tls_id > static_tls_cnt) {
+		    && def.dso->tls_id > static_tls_cnt) {
 			error("Error relocating %s: %s: initial-exec TLS "
 				"resolves to dynamic definition in %s",
 				dso->name, name, def.dso->name);
@@ -393,6 +416,9 @@ static void do_relocs(struct dso *dso, size_t *rel, size_t rel_size, size_t stri
 		case REL_PLT:
 			*reloc_addr = sym_val + addend;
 			break;
+		case REL_USYMBOLIC:
+			memcpy(reloc_addr, &(size_t){sym_val + addend}, sizeof(size_t));
+			break;
 		case REL_RELATIVE:
 			*reloc_addr = (size_t)base + addend;
 			break;
@@ -436,7 +462,7 @@ static void do_relocs(struct dso *dso, size_t *rel, size_t rel_size, size_t stri
 #endif
 		case REL_TLSDESC:
 			if (stride<3) addend = reloc_addr[1];
-			if (runtime && def.dso->tls_id > static_tls_cnt) {
+			if (def.dso->tls_id > static_tls_cnt) {
 				struct td_index *new = malloc(sizeof *new);
 				if (!new) {
 					error(
@@ -906,7 +932,7 @@ static void *dl_mmap(size_t n)
 #else
 	p = (void *)__syscall(SYS_mmap, 0, n, prot, flags, -1, 0);
 #endif
-	return p == MAP_FAILED ? 0 : p;
+	return (unsigned long)p > -4096UL ? 0 : p;
 }
 
 static void makefuncdescs(struct dso *p)
@@ -1103,6 +1129,7 @@ static struct dso *load_library(const char *name, struct dso *needed_by)
 	p->ino = st.st_ino;
 	p->needed_by = needed_by;
 	p->name = p->buf;
+	p->runtime_loaded = runtime;
 	strcpy(p->name, pathname);
 	/* Add a shortname only if name arg was not an explicit pathname. */
 	if (pathname != name) p->shortname = strrchr(p->name, '/')+1;
@@ -1110,9 +1137,9 @@ static struct dso *load_library(const char *name, struct dso *needed_by)
 		p->tls_id = ++tls_cnt;
 		tls_align = MAXP2(tls_align, p->tls.align);
 #ifdef TLS_ABOVE_TP
-		p->tls.offset = tls_offset + ( (tls_align-1) &
-			-(tls_offset + (uintptr_t)p->tls.image) );
-		tls_offset += p->tls.size;
+		p->tls.offset = tls_offset + ( (p->tls.align-1) &
+			(-tls_offset + (uintptr_t)p->tls.image) );
+		tls_offset = p->tls.offset + p->tls.size;
 #else
 		tls_offset += p->tls.size + p->tls.align - 1;
 		tls_offset -= (tls_offset + (uintptr_t)p->tls.image)
@@ -1138,30 +1165,99 @@ static struct dso *load_library(const char *name, struct dso *needed_by)
 	return p;
 }
 
+static void load_direct_deps(struct dso *p)
+{
+	size_t i, cnt=0;
+
+	if (p->deps) return;
+	/* For head, all preloads are direct pseudo-dependencies.
+	 * Count and include them now to avoid realloc later. */
+	if (p==head) for (struct dso *q=p->next; q; q=q->next)
+		cnt++;
+	for (i=0; p->dynv[i]; i+=2)
+		if (p->dynv[i] == DT_NEEDED) cnt++;
+	/* Use builtin buffer for apps with no external deps, to
+	 * preserve property of no runtime failure paths. */
+	p->deps = (p==head && cnt<2) ? builtin_deps :
+		calloc(cnt+1, sizeof *p->deps);
+	if (!p->deps) {
+		error("Error loading dependencies for %s", p->name);
+		if (runtime) longjmp(*rtld_fail, 1);
+	}
+	cnt=0;
+	if (p==head) for (struct dso *q=p->next; q; q=q->next)
+		p->deps[cnt++] = q;
+	for (i=0; p->dynv[i]; i+=2) {
+		if (p->dynv[i] != DT_NEEDED) continue;
+		struct dso *dep = load_library(p->strings + p->dynv[i+1], p);
+		if (!dep) {
+			error("Error loading shared library %s: %m (needed by %s)",
+				p->strings + p->dynv[i+1], p->name);
+			if (runtime) longjmp(*rtld_fail, 1);
+			continue;
+		}
+		p->deps[cnt++] = dep;
+	}
+	p->deps[cnt] = 0;
+	p->ndeps_direct = cnt;
+}
+
 static void load_deps(struct dso *p)
 {
-	size_t i, ndeps=0;
-	struct dso ***deps = &p->deps, **tmp, *dep;
-	for (; p; p=p->next) {
-		for (i=0; p->dynv[i]; i+=2) {
-			if (p->dynv[i] != DT_NEEDED) continue;
-			dep = load_library(p->strings + p->dynv[i+1], p);
-			if (!dep) {
-				error("Error loading shared library %s: %m (needed by %s)",
-					p->strings + p->dynv[i+1], p->name);
-				if (runtime) longjmp(*rtld_fail, 1);
-				continue;
-			}
-			if (runtime) {
-				tmp = realloc(*deps, sizeof(*tmp)*(ndeps+2));
-				if (!tmp) longjmp(*rtld_fail, 1);
-				tmp[ndeps++] = dep;
-				tmp[ndeps] = 0;
-				*deps = tmp;
-			}
+	if (p->deps) return;
+	for (; p; p=p->next)
+		load_direct_deps(p);
+}
+
+static void extend_bfs_deps(struct dso *p)
+{
+	size_t i, j, cnt, ndeps_all;
+	struct dso **tmp;
+
+	/* Can't use realloc if the original p->deps was allocated at
+	 * program entry and malloc has been replaced, or if it's
+	 * the builtin non-allocated trivial main program deps array. */
+	int no_realloc = (__malloc_replaced && !p->runtime_loaded)
+		|| p->deps == builtin_deps;
+
+	if (p->bfs_built) return;
+	ndeps_all = p->ndeps_direct;
+
+	/* Mark existing (direct) deps so they won't be duplicated. */
+	for (i=0; p->deps[i]; i++)
+		p->deps[i]->mark = 1;
+
+	/* For each dependency already in the list, copy its list of direct
+	 * dependencies to the list, excluding any items already in the
+	 * list. Note that the list this loop iterates over will grow during
+	 * the loop, but since duplicates are excluded, growth is bounded. */
+	for (i=0; p->deps[i]; i++) {
+		struct dso *dep = p->deps[i];
+		for (j=cnt=0; j<dep->ndeps_direct; j++)
+			if (!dep->deps[j]->mark) cnt++;
+		tmp = no_realloc ? 
+			malloc(sizeof(*tmp) * (ndeps_all+cnt+1)) :
+			realloc(p->deps, sizeof(*tmp) * (ndeps_all+cnt+1));
+		if (!tmp) {
+			error("Error recording dependencies for %s", p->name);
+			if (runtime) longjmp(*rtld_fail, 1);
+			continue;
 		}
+		if (no_realloc) {
+			memcpy(tmp, p->deps, sizeof(*tmp) * (ndeps_all+1));
+			no_realloc = 0;
+		}
+		p->deps = tmp;
+		for (j=0; j<dep->ndeps_direct; j++) {
+			if (dep->deps[j]->mark) continue;
+			dep->deps[j]->mark = 1;
+			p->deps[ndeps_all++] = dep->deps[j];
+		}
+		p->deps[ndeps_all] = 0;
 	}
-	if (!*deps) *deps = (struct dso **)&nodeps_dummy;
+	p->bfs_built = 1;
+	for (p=head; p; p=p->next)
+		p->mark = 0;
 }
 
 static void load_preload(char *s)
@@ -1277,7 +1373,18 @@ void __libc_exit_fini()
 {
 	struct dso *p;
 	size_t dyn[DYN_CNT];
+	int self = __pthread_self()->tid;
+
+	/* Take both locks before setting shutting_down, so that
+	 * either lock is sufficient to read its value. The lock
+	 * order matches that in dlopen to avoid deadlock. */
+	pthread_rwlock_wrlock(&lock);
+	pthread_mutex_lock(&init_fini_lock);
+	shutting_down = 1;
+	pthread_rwlock_unlock(&lock);
 	for (p=fini_head; p; p=p->fini_next) {
+		while (p->ctor_visitor && p->ctor_visitor!=self)
+			pthread_cond_wait(&ctor_cond, &init_fini_lock);
 		if (!p->constructed) continue;
 		decode_vec(p->dynv, dyn, DYN_CNT);
 		if (dyn[0] & (1<<DT_FINI_ARRAY)) {
@@ -1292,22 +1399,90 @@ void __libc_exit_fini()
 	}
 }
 
-static void do_init_fini(struct dso *p)
+static struct dso **queue_ctors(struct dso *dso)
 {
-	size_t dyn[DYN_CNT];
-	int need_locking = libc.threads_minus_1;
-	/* Allow recursive calls that arise when a library calls
-	 * dlopen from one of its constructors, but block any
-	 * other threads until all ctors have finished. */
-	if (need_locking) pthread_mutex_lock(&init_fini_lock);
-	for (; p; p=p->prev) {
-		if (p->constructed) continue;
-		p->constructed = 1;
+	size_t cnt, qpos, spos, i;
+	struct dso *p, **queue, **stack;
+
+	if (ldd_mode) return 0;
+
+	/* Bound on queue size is the total number of indirect deps.
+	 * If a bfs deps list was built, we can use it. Otherwise,
+	 * bound by the total number of DSOs, which is always safe and
+	 * is reasonable we use it (for main app at startup). */
+	if (dso->bfs_built) {
+		for (cnt=0; dso->deps[cnt]; cnt++)
+			dso->deps[cnt]->mark = 0;
+		cnt++; /* self, not included in deps */
+	} else {
+		for (cnt=0, p=head; p; cnt++, p=p->next)
+			p->mark = 0;
+	}
+	cnt++; /* termination slot */
+	if (dso==head && cnt <= countof(builtin_ctor_queue))
+		queue = builtin_ctor_queue;
+	else
+		queue = calloc(cnt, sizeof *queue);
+
+	if (!queue) {
+		error("Error allocating constructor queue: %m\n");
+		if (runtime) longjmp(*rtld_fail, 1);
+		return 0;
+	}
+
+	/* Opposite ends of the allocated buffer serve as an output queue
+	 * and a working stack. Setup initial stack with just the argument
+	 * dso and initial queue empty... */
+	stack = queue;
+	qpos = 0;
+	spos = cnt;
+	stack[--spos] = dso;
+	dso->next_dep = 0;
+	dso->mark = 1;
+
+	/* Then perform pseudo-DFS sort, but ignoring circular deps. */
+	while (spos<cnt) {
+		p = stack[spos++];
+		while (p->next_dep < p->ndeps_direct) {
+			if (p->deps[p->next_dep]->mark) {
+				p->next_dep++;
+			} else {
+				stack[--spos] = p;
+				p = p->deps[p->next_dep];
+				p->next_dep = 0;
+				p->mark = 1;
+			}
+		}
+		queue[qpos++] = p;
+	}
+	queue[qpos] = 0;
+	for (i=0; i<qpos; i++) queue[i]->mark = 0;
+
+	return queue;
+}
+
+static void do_init_fini(struct dso **queue)
+{
+	struct dso *p;
+	size_t dyn[DYN_CNT], i;
+	int self = __pthread_self()->tid;
+
+	pthread_mutex_lock(&init_fini_lock);
+	for (i=0; (p=queue[i]); i++) {
+		while ((p->ctor_visitor && p->ctor_visitor!=self) || shutting_down)
+			pthread_cond_wait(&ctor_cond, &init_fini_lock);
+		if (p->ctor_visitor || p->constructed)
+			continue;
+		p->ctor_visitor = self;
+		
 		decode_vec(p->dynv, dyn, DYN_CNT);
 		if (dyn[0] & ((1<<DT_FINI) | (1<<DT_FINI_ARRAY))) {
 			p->fini_next = fini_head;
 			fini_head = p;
 		}
+
+		pthread_mutex_unlock(&init_fini_lock);
+
 #ifndef NO_LEGACY_INITFINI
 		if ((dyn[0] & (1<<DT_INIT)) && dyn[DT_INIT])
 			fpaddr(p, dyn[DT_INIT])();
@@ -1317,17 +1492,21 @@ static void do_init_fini(struct dso *p)
 			size_t *fn = laddr(p, dyn[DT_INIT_ARRAY]);
 			while (n--) ((void (*)(void))*fn++)();
 		}
-		if (!need_locking && libc.threads_minus_1) {
-			need_locking = 1;
-			pthread_mutex_lock(&init_fini_lock);
-		}
+
+		pthread_mutex_lock(&init_fini_lock);
+		p->ctor_visitor = 0;
+		p->constructed = 1;
+		pthread_cond_broadcast(&ctor_cond);
 	}
-	if (need_locking) pthread_mutex_unlock(&init_fini_lock);
+	pthread_mutex_unlock(&init_fini_lock);
 }
 
 void __libc_start_init(void)
 {
-	do_init_fini(tail);
+	do_init_fini(main_ctor_queue);
+	if (!__malloc_replaced && main_ctor_queue != builtin_ctor_queue)
+		free(main_ctor_queue);
+	main_ctor_queue = 0;
 }
 
 static void dl_debug_state(void)
@@ -1356,7 +1535,8 @@ static void install_new_tls(void)
 {
 	sigset_t set;
 	pthread_t self = __pthread_self(), td;
-	uintptr_t (*newdtv)[tls_cnt+1] = (void *)tail->new_dtv;
+	struct dso *dtv_provider = container_of(tls_tail, struct dso, tls);
+	uintptr_t (*newdtv)[tls_cnt+1] = (void *)dtv_provider->new_dtv;
 	struct dso *p;
 	size_t i, j;
 	size_t old_cnt = self->dtv[0];
@@ -1371,7 +1551,7 @@ static void install_new_tls(void)
 	}
 	/* Install new dtls into the enlarged, uninstalled dtv copies. */
 	for (p=head; ; p=p->next) {
-		if (!p->tls_id || self->dtv[p->tls_id]) continue;
+		if (p->tls_id <= old_cnt) continue;
 		unsigned char *mem = p->new_tls;
 		for (j=0; j<i; j++) {
 			unsigned char *new = mem;
@@ -1476,7 +1656,7 @@ hidden void __dls2(unsigned char *base, size_t *sp)
  * so that loads of the thread pointer and &errno can be pure/const and
  * thereby hoistable. */
 
-_Noreturn void __dls2b(size_t *sp)
+void __dls2b(size_t *sp)
 {
 	/* Setup early thread pointer in builtin_tls for ldso/libc itself to
 	 * use during dynamic linking. If possible it will also serve as the
@@ -1497,7 +1677,7 @@ _Noreturn void __dls2b(size_t *sp)
  * process dependencies and relocations for the main application and
  * transfer control to its entry point. */
 
-_Noreturn void __dls3(size_t *sp)
+void __dls3(size_t *sp)
 {
 	static struct dso app, vdso;
 	size_t aux[AUX_CNT], *auxv;
@@ -1517,6 +1697,8 @@ _Noreturn void __dls3(size_t *sp)
 	libc.auxv = auxv = (void *)(argv+i+1);
 	decode_vec(auxv, aux, AUX_CNT);
 	__hwcap = aux[AT_HWCAP];
+	search_vec(auxv, &__sysinfo, AT_SYSINFO);
+	__pthread_self()->sysinfo = __sysinfo;
 	libc.page_size = aux[AT_PAGESZ];
 	libc.secure = ((aux[0]&0x7800)!=0x7800 || aux[AT_UID]!=aux[AT_EUID]
 		|| aux[AT_GID]!=aux[AT_EGID] || aux[AT_SECURE]);
@@ -1626,10 +1808,9 @@ _Noreturn void __dls3(size_t *sp)
 		app.tls_id = tls_cnt = 1;
 #ifdef TLS_ABOVE_TP
 		app.tls.offset = GAP_ABOVE_TP;
-		app.tls.offset += -GAP_ABOVE_TP & (app.tls.align-1);
-		tls_offset = app.tls.offset + app.tls.size
-			+ ( -((uintptr_t)app.tls.image + app.tls.size)
-			& (app.tls.align-1) );
+		app.tls.offset += (-GAP_ABOVE_TP + (uintptr_t)app.tls.image)
+			& (app.tls.align-1);
+		tls_offset = app.tls.offset + app.tls.size;
 #else
 		tls_offset = app.tls.offset = app.tls.size
 			+ ( -((uintptr_t)app.tls.image + app.tls.size)
@@ -1659,6 +1840,7 @@ _Noreturn void __dls3(size_t *sp)
 	reclaim_gaps(&ldso);
 
 	/* Load preload/needed libraries, add symbols to global namespace. */
+	ldso.deps = (struct dso **)no_deps;
 	if (env_preload) load_preload(env_preload);
  	load_deps(&app);
 	for (struct dso *p=head; p; p=p->next)
@@ -1680,6 +1862,7 @@ _Noreturn void __dls3(size_t *sp)
 		vdso.name = "";
 		vdso.shortname = "linux-gate.so.1";
 		vdso.relocated = 1;
+		vdso.deps = (struct dso **)no_deps;
 		decode_dyn(&vdso);
 		vdso.prev = tail;
 		tail->next = &vdso;
@@ -1695,19 +1878,36 @@ _Noreturn void __dls3(size_t *sp)
 		}
 	}
 
-	/* The main program must be relocated LAST since it may contin
-	 * copy relocations which depend on libraries' relocations. */
-	reloc_all(app.next);
-	reloc_all(&app);
+	/* This must be done before final relocations, since it calls
+	 * malloc, which may be provided by the application. Calling any
+	 * application code prior to the jump to its entry point is not
+	 * valid in our model and does not work with FDPIC, where there
+	 * are additional relocation-like fixups that only the entry point
+	 * code can see to perform. */
+	main_ctor_queue = queue_ctors(&app);
 
+	/* Initial TLS must also be allocated before final relocations
+	 * might result in calloc being a call to application code. */
 	update_tls_size();
+	void *initial_tls = builtin_tls;
 	if (libc.tls_size > sizeof builtin_tls || tls_align > MIN_TLS_ALIGN) {
-		void *initial_tls = calloc(libc.tls_size, 1);
+		initial_tls = calloc(libc.tls_size, 1);
 		if (!initial_tls) {
 			dprintf(2, "%s: Error getting %zu bytes thread-local storage: %m\n",
 				argv[0], libc.tls_size);
 			_exit(127);
 		}
+	}
+	static_tls_cnt = tls_cnt;
+
+	/* The main program must be relocated LAST since it may contain
+	 * copy relocations which depend on libraries' relocations. */
+	reloc_all(app.next);
+	reloc_all(&app);
+
+	/* Actual copying to new TLS needs to happen after relocations,
+	 * since the TLS images might have contained relocated addresses. */
+	if (initial_tls != builtin_tls) {
 		if (__init_tp(__copy_tls(initial_tls)) < 0) {
 			a_crash();
 		}
@@ -1721,7 +1921,6 @@ _Noreturn void __dls3(size_t *sp)
 		if (__copy_tls((void*)builtin_tls) != self) a_crash();
 		libc.tls_size = tmp_tls_size;
 	}
-	static_tls_cnt = tls_cnt;
 
 	if (ldso_fail) _exit(127);
 	if (ldd_mode) _exit(0);
@@ -1782,6 +1981,7 @@ void *dlopen(const char *file, int mode)
 	size_t i;
 	int cs;
 	jmp_buf jb;
+	struct dso **volatile ctor_queue = 0;
 
 	if (!file) return head;
 
@@ -1790,6 +1990,10 @@ void *dlopen(const char *file, int mode)
 	__inhibit_ptc();
 
 	p = 0;
+	if (shutting_down) {
+		error("Cannot dlopen while program is exiting.");
+		goto end;
+	}
 	orig_tls_tail = tls_tail;
 	orig_tls_cnt = tls_cnt;
 	orig_tls_offset = tls_offset;
@@ -1813,11 +2017,12 @@ void *dlopen(const char *file, int mode)
 			free(p->funcdescs);
 			if (p->rpath != p->rpath_orig)
 				free(p->rpath);
-			if (p->deps != &nodeps_dummy)
-				free(p->deps);
+			free(p->deps);
 			unmap_library(p);
 			free(p);
 		}
+		free(ctor_queue);
+		ctor_queue = 0;
 		if (!orig_tls_tail) libc.tls_head = 0;
 		tls_tail = orig_tls_tail;
 		if (tls_tail) tls_tail->next = 0;
@@ -1840,24 +2045,25 @@ void *dlopen(const char *file, int mode)
 	}
 
 	/* First load handling */
-	int first_load = !p->deps;
-	if (first_load) {
-		load_deps(p);
-		if (!p->relocated && (mode & RTLD_LAZY)) {
-			prepare_lazy(p);
-			for (i=0; p->deps[i]; i++)
-				if (!p->deps[i]->relocated)
-					prepare_lazy(p->deps[i]);
-		}
+	load_deps(p);
+	extend_bfs_deps(p);
+	pthread_mutex_lock(&init_fini_lock);
+	if (!p->constructed) ctor_queue = queue_ctors(p);
+	pthread_mutex_unlock(&init_fini_lock);
+	if (!p->relocated && (mode & RTLD_LAZY)) {
+		prepare_lazy(p);
+		for (i=0; p->deps[i]; i++)
+			if (!p->deps[i]->relocated)
+				prepare_lazy(p->deps[i]);
 	}
-	if (first_load || (mode & RTLD_GLOBAL)) {
+	if (!p->relocated || (mode & RTLD_GLOBAL)) {
 		/* Make new symbols global, at least temporarily, so we can do
 		 * relocations. If not RTLD_GLOBAL, this is reverted below. */
 		add_syms(p);
 		for (i=0; p->deps[i]; i++)
 			add_syms(p->deps[i]);
 	}
-	if (first_load) {
+	if (!p->relocated) {
 		reloc_all(p);
 	}
 
@@ -1881,7 +2087,10 @@ end:
 	__release_ptc();
 	if (p) gencnt++;
 	pthread_rwlock_unlock(&lock);
-	if (p) do_init_fini(orig_tail);
+	if (ctor_queue) {
+		do_init_fini(ctor_queue);
+		free(ctor_queue);
+	}
 	pthread_setcancelstate(cs, 0);
 	return p;
 }
@@ -1929,58 +2138,27 @@ static void *addr2dso(size_t a)
 
 static void *do_dlsym(struct dso *p, const char *s, void *ra)
 {
-	size_t i;
-	uint32_t h = 0, gh = 0, *ght;
-	Sym *sym;
-	if (p == head || p == RTLD_DEFAULT || p == RTLD_NEXT) {
-		if (p == RTLD_DEFAULT) {
-			p = head;
-		} else if (p == RTLD_NEXT) {
-			p = addr2dso((size_t)ra);
-			if (!p) p=head;
-			p = p->next;
-		}
-		struct symdef def = find_sym(p, s, 0);
-		if (!def.sym) goto failed;
-		if ((def.sym->st_info&0xf) == STT_TLS)
-			return __tls_get_addr((tls_mod_off_t []){def.dso->tls_id, def.sym->st_value-DTP_OFFSET});
-		if (DL_FDPIC && (def.sym->st_info&0xf) == STT_FUNC)
-			return def.dso->funcdescs + (def.sym - def.dso->syms);
-		return laddr(def.dso, def.sym->st_value);
-	}
-	if (__dl_invalid_handle(p))
+	int use_deps = 0;
+	if (p == head || p == RTLD_DEFAULT) {
+		p = head;
+	} else if (p == RTLD_NEXT) {
+		p = addr2dso((size_t)ra);
+		if (!p) p=head;
+		p = p->next;
+	} else if (__dl_invalid_handle(p)) {
 		return 0;
-	if ((ght = p->ghashtab)) {
-		gh = gnu_hash(s);
-		sym = gnu_lookup(gh, ght, p, s);
-	} else {
-		h = sysv_hash(s);
-		sym = sysv_lookup(s, h, p);
-	}
-	if (sym && (sym->st_info&0xf) == STT_TLS)
-		return __tls_get_addr((tls_mod_off_t []){p->tls_id, sym->st_value-DTP_OFFSET});
-	if (DL_FDPIC && sym && sym->st_shndx && (sym->st_info&0xf) == STT_FUNC)
-		return p->funcdescs + (sym - p->syms);
-	if (sym && sym->st_value && (1<<(sym->st_info&0xf) & OK_TYPES))
-		return laddr(p, sym->st_value);
-	for (i=0; p->deps[i]; i++) {
-		if ((ght = p->deps[i]->ghashtab)) {
-			if (!gh) gh = gnu_hash(s);
-			sym = gnu_lookup(gh, ght, p->deps[i], s);
-		} else {
-			if (!h) h = sysv_hash(s);
-			sym = sysv_lookup(s, h, p->deps[i]);
-		}
-		if (sym && (sym->st_info&0xf) == STT_TLS)
-			return __tls_get_addr((tls_mod_off_t []){p->deps[i]->tls_id, sym->st_value-DTP_OFFSET});
-		if (DL_FDPIC && sym && sym->st_shndx && (sym->st_info&0xf) == STT_FUNC)
-			return p->deps[i]->funcdescs + (sym - p->deps[i]->syms);
-		if (sym && sym->st_value && (1<<(sym->st_info&0xf) & OK_TYPES))
-			return laddr(p->deps[i], sym->st_value);
-	}
-failed:
-	error("Symbol not found: %s", s);
-	return 0;
+	} else
+		use_deps = 1;
+	struct symdef def = find_sym2(p, s, 0, use_deps);
+	if (!def.sym) {
+		error("Symbol not found: %s", s);
+		return 0;
+	}
+	if ((def.sym->st_info&0xf) == STT_TLS)
+		return __tls_get_addr((tls_mod_off_t []){def.dso->tls_id, def.sym->st_value-DTP_OFFSET});
+	if (DL_FDPIC && (def.sym->st_info&0xf) == STT_FUNC)
+		return def.dso->funcdescs + (def.sym - def.dso->syms);
+	return laddr(def.dso, def.sym->st_value);
 }
 
 int dladdr(const void *addr_arg, Dl_info *info)
@@ -2028,7 +2206,7 @@ int dladdr(const void *addr_arg, Dl_info *info)
 		}
 	}
 
-	if (bestsym && besterr > bestsym->st_size-1) {
+	if (best && besterr > bestsym->st_size-1) {
 		best = 0;
 		bestsym = 0;
 	}