1 /* $NetBSD: linux_exec.c,v 1.106 2008/10/15 06:51:19 wrstuden Exp $ */ 2 3 /*- 4 * Copyright (c) 1994, 1995, 1998, 2000, 2007, 2008 The NetBSD Foundation, Inc. 5 * All rights reserved. 6 * 7 * This code is derived from software contributed to The NetBSD Foundation 8 * by Christos Zoulas, Frank van der Linden, Eric Haszlakiewicz and 9 * Thor Lancelot Simon. 10 * 11 * Redistribution and use in source and binary forms, with or without 12 * modification, are permitted provided that the following conditions 13 * are met: 14 * 1. Redistributions of source code must retain the above copyright 15 * notice, this list of conditions and the following disclaimer. 16 * 2. Redistributions in binary form must reproduce the above copyright 17 * notice, this list of conditions and the following disclaimer in the 18 * documentation and/or other materials provided with the distribution. 19 * 20 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS 21 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 22 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 23 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS 24 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 25 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 26 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 27 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 28 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 29 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 30 * POSSIBILITY OF SUCH DAMAGE. 31 */ 32 33 #include <sys/cdefs.h> 34 __KERNEL_RCSID(0, "$NetBSD: linux_exec.c,v 1.106 2008/10/15 06:51:19 wrstuden Exp $"); 35 36 #include <sys/param.h> 37 #include <sys/systm.h> 38 #include <sys/kernel.h> 39 #include <sys/proc.h> 40 #include <sys/malloc.h> 41 #include <sys/namei.h> 42 #include <sys/vnode.h> 43 #include <sys/mount.h> 44 #include <sys/exec.h> 45 #include <sys/exec_elf.h> 46 47 #include <sys/mman.h> 48 #include <sys/syscallargs.h> 49 50 #include <sys/ptrace.h> /* For proc_reparent() */ 51 52 #include <uvm/uvm_extern.h> 53 54 #include <sys/cpu.h> 55 #include <machine/reg.h> 56 57 #include <compat/linux/common/linux_types.h> 58 #include <compat/linux/common/linux_signal.h> 59 #include <compat/linux/common/linux_util.h> 60 #include <compat/linux/common/linux_sched.h> 61 #include <compat/linux/common/linux_machdep.h> 62 #include <compat/linux/common/linux_exec.h> 63 #include <compat/linux/common/linux_futex.h> 64 #include <compat/linux/common/linux_ipc.h> 65 #include <compat/linux/common/linux_sem.h> 66 67 #include <compat/linux/linux_syscallargs.h> 68 #include <compat/linux/linux_syscall.h> 69 #include <compat/linux/common/linux_misc.h> 70 #include <compat/linux/common/linux_errno.h> 71 #include <compat/linux/common/linux_emuldata.h> 72 73 extern struct sysent linux_sysent[]; 74 extern const char * const linux_syscallnames[]; 75 extern char linux_sigcode[], linux_esigcode[]; 76 77 static void linux_e_proc_exec(struct proc *, struct exec_package *); 78 static void linux_e_proc_fork(struct proc *, struct proc *, int); 79 static void linux_e_proc_exit(struct proc *); 80 static void linux_e_proc_init(struct proc *, struct proc *, int); 81 82 #ifdef LINUX_NPTL 83 void linux_userret(void); 84 #endif 85 86 /* 87 * Emulation switch. 88 */ 89 90 struct uvm_object *emul_linux_object; 91 92 const struct emul emul_linux = { 93 "linux", 94 "/emul/linux", 95 #ifndef __HAVE_MINIMAL_EMUL 96 0, 97 (const int *)native_to_linux_errno, 98 LINUX_SYS_syscall, 99 LINUX_SYS_NSYSENT, 100 #endif 101 linux_sysent, 102 linux_syscallnames, 103 linux_sendsig, 104 linux_trapsignal, 105 NULL, 106 linux_sigcode, 107 linux_esigcode, 108 &emul_linux_object, 109 linux_setregs, 110 linux_e_proc_exec, 111 linux_e_proc_fork, 112 linux_e_proc_exit, 113 NULL, 114 NULL, 115 #ifdef __HAVE_SYSCALL_INTERN 116 linux_syscall_intern, 117 #else 118 #error Implement __HAVE_SYSCALL_INTERN for this platform 119 #endif 120 NULL, 121 NULL, 122 123 uvm_default_mapaddr, 124 125 linux_usertrap, 126 NULL, /* e_sa */ 127 0, 128 NULL, /* e_startlwp */ 129 }; 130 131 static void 132 linux_e_proc_init(p, parent, forkflags) 133 struct proc *p, *parent; 134 int forkflags; 135 { 136 struct linux_emuldata *e = p->p_emuldata; 137 struct linux_emuldata_shared *s; 138 struct linux_emuldata *ep = NULL; 139 140 if (!e) { 141 /* allocate new Linux emuldata */ 142 MALLOC(e, void *, sizeof(struct linux_emuldata), 143 M_EMULDATA, M_WAITOK); 144 } else { 145 mutex_enter(proc_lock); 146 e->s->refs--; 147 if (e->s->refs == 0) 148 FREE(e->s, M_EMULDATA); 149 mutex_exit(proc_lock); 150 } 151 152 memset(e, '\0', sizeof(struct linux_emuldata)); 153 154 e->proc = p; 155 156 if (parent) 157 ep = parent->p_emuldata; 158 159 if (forkflags & FORK_SHAREVM) { 160 mutex_enter(proc_lock); 161 #ifdef DIAGNOSTIC 162 if (ep == NULL) { 163 killproc(p, "FORK_SHAREVM while emuldata is NULL\n"); 164 mutex_exit(proc_lock); 165 FREE(e, M_EMULDATA); 166 return; 167 } 168 #endif 169 s = ep->s; 170 s->refs++; 171 } else { 172 struct vmspace *vm; 173 174 MALLOC(s, void *, sizeof(struct linux_emuldata_shared), 175 M_EMULDATA, M_WAITOK); 176 s->refs = 1; 177 178 /* 179 * Set the process idea of the break to the real value. 180 * For fork, we use parent's vmspace since our's 181 * is not setup at the time of this call and is going 182 * to be copy of parent's anyway. For exec, just 183 * use our own vmspace. 184 */ 185 vm = (parent) ? parent->p_vmspace : p->p_vmspace; 186 s->p_break = (char *)vm->vm_daddr + ctob(vm->vm_dsize); 187 188 /* 189 * Linux threads are emulated as NetBSD processes (not lwp) 190 * We use native PID for Linux TID. The Linux TID is the 191 * PID of the first process in the group. It is stored 192 * here 193 */ 194 s->group_pid = p->p_pid; 195 196 /* 197 * Initialize the list of threads in the group 198 */ 199 LIST_INIT(&s->threads); 200 201 s->xstat = 0; 202 s->flags = 0; 203 mutex_enter(proc_lock); 204 } 205 206 e->s = s; 207 208 /* 209 * Add this thread in the group thread list 210 */ 211 LIST_INSERT_HEAD(&s->threads, e, threads); 212 213 #ifdef LINUX_NPTL 214 if (ep != NULL) { 215 e->parent_tidptr = ep->parent_tidptr; 216 e->child_tidptr = ep->child_tidptr; 217 e->clone_flags = ep->clone_flags; 218 } 219 #endif /* LINUX_NPTL */ 220 221 mutex_exit(proc_lock); 222 p->p_emuldata = e; 223 } 224 225 /* 226 * Allocate new per-process structures. Called when executing Linux 227 * process. We can reuse the old emuldata - if it's not null, 228 * the executed process is of same emulation as original forked one. 229 */ 230 static void 231 linux_e_proc_exec(struct proc *p, struct exec_package *epp) 232 { 233 234 /* exec, use our vmspace */ 235 linux_e_proc_init(p, NULL, 0); 236 } 237 238 /* 239 * Emulation per-process exit hook. 240 */ 241 static void 242 linux_e_proc_exit(struct proc *p) 243 { 244 struct linux_emuldata *e = p->p_emuldata; 245 246 #ifdef LINUX_NPTL 247 linux_nptl_proc_exit(p); 248 #endif 249 250 /* Remove the thread for the group thread list */ 251 mutex_enter(proc_lock); 252 LIST_REMOVE(e, threads); 253 254 /* free Linux emuldata and set the pointer to null */ 255 e->s->refs--; 256 if (e->s->refs == 0) 257 FREE(e->s, M_EMULDATA); 258 p->p_emuldata = NULL; 259 mutex_exit(proc_lock); 260 FREE(e, M_EMULDATA); 261 } 262 263 /* 264 * Emulation fork hook. 265 */ 266 static void 267 linux_e_proc_fork(p, parent, forkflags) 268 struct proc *p, *parent; 269 int forkflags; 270 { 271 272 /* 273 * The new process might share some vmspace-related stuff 274 * with parent, depending on fork flags (CLONE_VM et.al). 275 * Force allocation of new base emuldata, and share the 276 * VM-related parts only if necessary. 277 */ 278 p->p_emuldata = NULL; 279 linux_e_proc_init(p, parent, forkflags); 280 281 #ifdef LINUX_NPTL 282 linux_nptl_proc_fork(p, parent, linux_userret); 283 #endif 284 285 return; 286 } 287 288 #ifdef LINUX_NPTL 289 void 290 linux_userret(void) 291 { 292 struct lwp *l = curlwp; 293 struct proc *p = l->l_proc; 294 struct linux_emuldata *led = p->p_emuldata; 295 int error; 296 297 /* LINUX_CLONE_CHILD_SETTID: copy child's TID to child's memory */ 298 if (led->clone_flags & LINUX_CLONE_CHILD_SETTID) { 299 if ((error = copyout(&l->l_proc->p_pid, 300 led->child_tidptr, sizeof(l->l_proc->p_pid))) != 0) 301 printf("linux_userret: LINUX_CLONE_CHILD_SETTID " 302 "failed (led->child_tidptr = %p, p->p_pid = %d)\n", 303 led->child_tidptr, p->p_pid); 304 } 305 306 /* LINUX_CLONE_SETTLS: allocate a new TLS */ 307 if (led->clone_flags & LINUX_CLONE_SETTLS) { 308 if (linux_set_newtls(l, linux_get_newtls(l)) != 0) 309 printf("linux_userret: linux_set_tls failed"); 310 } 311 312 return; 313 } 314 315 void 316 linux_nptl_proc_exit(struct proc *p) 317 { 318 struct linux_emuldata *e = p->p_emuldata; 319 320 mutex_enter(proc_lock); 321 322 /* 323 * Check if we are a thread group leader victim of another 324 * thread doing exit_group(). If we are, change the exit code. 325 */ 326 if ((e->s->group_pid == p->p_pid) && 327 (e->s->flags & LINUX_LES_INEXITGROUP)) { 328 p->p_xstat = e->s->xstat; 329 } 330 331 /* 332 * Members of the thread groups others than the leader should 333 * exit quietely: no zombie stage, no signal. We do that by 334 * reparenting to init. init will collect us and nobody will 335 * notice what happened. 336 */ 337 #ifdef DEBUG_LINUX 338 printf("%s:%d e->s->group_pid = %d, p->p_pid = %d, flags = 0x%x\n", 339 __func__, __LINE__, e->s->group_pid, p->p_pid, e->s->flags); 340 #endif 341 if ((e->s->group_pid != p->p_pid) && 342 (e->clone_flags & LINUX_CLONE_THREAD)) { 343 proc_reparent(p, initproc); 344 cv_broadcast(&initproc->p_waitcv); 345 } 346 347 mutex_exit(proc_lock); 348 349 /* Emulate LINUX_CLONE_CHILD_CLEARTID */ 350 if (e->clear_tid != NULL) { 351 int error; 352 int null = 0; 353 struct linux_sys_futex_args cup; 354 register_t retval; 355 356 error = copyout(&null, e->clear_tid, sizeof(null)); 357 #ifdef DEBUG_LINUX 358 if (error != 0) 359 printf("%s: cannot clear TID\n", __func__); 360 #endif 361 362 SCARG(&cup, uaddr) = e->clear_tid; 363 SCARG(&cup, op) = LINUX_FUTEX_WAKE; 364 SCARG(&cup, val) = 0x7fffffff; /* Awake everyone */ 365 SCARG(&cup, timeout) = NULL; 366 SCARG(&cup, uaddr2) = NULL; 367 SCARG(&cup, val3) = 0; 368 if ((error = linux_sys_futex(curlwp, &cup, &retval)) != 0) 369 printf("%s: linux_sys_futex failed\n", __func__); 370 } 371 372 return; 373 } 374 375 void 376 linux_nptl_proc_fork(p, parent, luserret) 377 struct proc *p; 378 struct proc *parent; 379 void (*luserret)(void); 380 { 381 struct linux_emuldata *e = p->p_emuldata; 382 383 /* LINUX_CLONE_CHILD_CLEARTID: clear TID in child's memory on exit() */ 384 if (e->clone_flags & LINUX_CLONE_CHILD_CLEARTID) 385 e->clear_tid = e->child_tidptr; 386 387 /* LINUX_CLONE_PARENT_SETTID: set child's TID in parent's memory */ 388 if (e->clone_flags & LINUX_CLONE_PARENT_SETTID) { 389 if (copyout_proc(parent, &p->p_pid, 390 e->parent_tidptr, sizeof(p->p_pid)) != 0) 391 printf("%s: LINUX_CLONE_PARENT_SETTID " 392 "failed (e->parent_tidptr = %p, " 393 "parent->p_pid = %d, p->p_pid = %d)\n", 394 __func__, e->parent_tidptr, 395 parent->p_pid, p->p_pid); 396 } 397 398 /* 399 * CLONE_CHILD_SETTID and LINUX_CLONE_SETTLS require child's VM 400 * setup to be completed, we postpone them until userret time. 401 */ 402 if (e->clone_flags & 403 (LINUX_CLONE_CHILD_CLEARTID | LINUX_CLONE_SETTLS)) 404 p->p_userret = luserret; 405 406 return; 407 } 408 409 void 410 linux_nptl_proc_init(struct proc *p, struct proc *parent) 411 { 412 struct linux_emuldata *e = p->p_emuldata; 413 struct linux_emuldata *ep; 414 415 if ((parent != NULL) && (parent->p_emuldata != NULL)) { 416 ep = parent->p_emuldata; 417 418 e->parent_tidptr = ep->parent_tidptr; 419 e->child_tidptr = ep->child_tidptr; 420 e->clone_flags = ep->clone_flags; 421 } 422 423 return; 424 } 425 #endif /* LINUX_NPTL */ 426