1 /* $NetBSD: kern_exec.c,v 1.514 2021/11/26 08:06:12 ryo Exp $ */ 2 3 /*- 4 * Copyright (c) 2008, 2019, 2020 The NetBSD Foundation, Inc. 5 * All rights reserved. 6 * 7 * This code is derived from software contributed to The NetBSD Foundation 8 * by Andrew Doran. 9 * 10 * Redistribution and use in source and binary forms, with or without 11 * modification, are permitted provided that the following conditions 12 * are met: 13 * 1. Redistributions of source code must retain the above copyright 14 * notice, this list of conditions and the following disclaimer. 15 * 2. Redistributions in binary form must reproduce the above copyright 16 * notice, this list of conditions and the following disclaimer in the 17 * documentation and/or other materials provided with the distribution. 18 * 19 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS 20 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 21 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 22 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS 23 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 24 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 25 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 26 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 27 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 28 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 29 * POSSIBILITY OF SUCH DAMAGE. 30 */ 31 32 /*- 33 * Copyright (C) 1993, 1994, 1996 Christopher G. Demetriou 34 * Copyright (C) 1992 Wolfgang Solfrank. 35 * Copyright (C) 1992 TooLs GmbH. 36 * All rights reserved. 37 * 38 * Redistribution and use in source and binary forms, with or without 39 * modification, are permitted provided that the following conditions 40 * are met: 41 * 1. Redistributions of source code must retain the above copyright 42 * notice, this list of conditions and the following disclaimer. 43 * 2. Redistributions in binary form must reproduce the above copyright 44 * notice, this list of conditions and the following disclaimer in the 45 * documentation and/or other materials provided with the distribution. 46 * 3. All advertising materials mentioning features or use of this software 47 * must display the following acknowledgement: 48 * This product includes software developed by TooLs GmbH. 49 * 4. The name of TooLs GmbH may not be used to endorse or promote products 50 * derived from this software without specific prior written permission. 51 * 52 * THIS SOFTWARE IS PROVIDED BY TOOLS GMBH ``AS IS'' AND ANY EXPRESS OR 53 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES 54 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. 55 * IN NO EVENT SHALL TOOLS GMBH BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, 56 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, 57 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; 58 * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, 59 * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR 60 * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF 61 * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 62 */ 63 64 #include <sys/cdefs.h> 65 __KERNEL_RCSID(0, "$NetBSD: kern_exec.c,v 1.514 2021/11/26 08:06:12 ryo Exp $"); 66 67 #include "opt_exec.h" 68 #include "opt_execfmt.h" 69 #include "opt_ktrace.h" 70 #include "opt_modular.h" 71 #include "opt_syscall_debug.h" 72 #include "veriexec.h" 73 #include "opt_pax.h" 74 75 #include <sys/param.h> 76 #include <sys/systm.h> 77 #include <sys/filedesc.h> 78 #include <sys/kernel.h> 79 #include <sys/proc.h> 80 #include <sys/ptrace.h> 81 #include <sys/mount.h> 82 #include <sys/kmem.h> 83 #include <sys/namei.h> 84 #include <sys/vnode.h> 85 #include <sys/file.h> 86 #include <sys/filedesc.h> 87 #include <sys/acct.h> 88 #include <sys/atomic.h> 89 #include <sys/exec.h> 90 #include <sys/futex.h> 91 #include <sys/ktrace.h> 92 #include <sys/uidinfo.h> 93 #include <sys/wait.h> 94 #include <sys/mman.h> 95 #include <sys/ras.h> 96 #include <sys/signalvar.h> 97 #include <sys/stat.h> 98 #include <sys/syscall.h> 99 #include <sys/kauth.h> 100 #include <sys/lwpctl.h> 101 #include <sys/pax.h> 102 #include <sys/cpu.h> 103 #include <sys/module.h> 104 #include <sys/syscallvar.h> 105 #include <sys/syscallargs.h> 106 #include <sys/vfs_syscalls.h> 107 #if NVERIEXEC > 0 108 #include <sys/verified_exec.h> 109 #endif /* NVERIEXEC > 0 */ 110 #include <sys/sdt.h> 111 #include <sys/spawn.h> 112 #include <sys/prot.h> 113 #include <sys/cprng.h> 114 115 #include <uvm/uvm_extern.h> 116 117 #include <machine/reg.h> 118 119 #include <compat/common/compat_util.h> 120 121 #ifndef MD_TOPDOWN_INIT 122 #ifdef __USE_TOPDOWN_VM 123 #define MD_TOPDOWN_INIT(epp) (epp)->ep_flags |= EXEC_TOPDOWN_VM 124 #else 125 #define MD_TOPDOWN_INIT(epp) 126 #endif 127 #endif 128 129 struct execve_data; 130 131 extern int user_va0_disable; 132 133 static size_t calcargs(struct execve_data * restrict, const size_t); 134 static size_t calcstack(struct execve_data * restrict, const size_t); 135 static int copyoutargs(struct execve_data * restrict, struct lwp *, 136 char * const); 137 static int copyoutpsstrs(struct execve_data * restrict, struct proc *); 138 static int copyinargs(struct execve_data * restrict, char * const *, 139 char * const *, execve_fetch_element_t, char **); 140 static int copyinargstrs(struct execve_data * restrict, char * const *, 141 execve_fetch_element_t, char **, size_t *, void (*)(const void *, size_t)); 142 static int exec_sigcode_map(struct proc *, const struct emul *); 143 144 #if defined(DEBUG) && !defined(DEBUG_EXEC) 145 #define DEBUG_EXEC 146 #endif 147 #ifdef DEBUG_EXEC 148 #define DPRINTF(a) printf a 149 #define COPYPRINTF(s, a, b) printf("%s, %d: copyout%s @%p %zu\n", __func__, \ 150 __LINE__, (s), (a), (b)) 151 static void dump_vmcmds(const struct exec_package * const, size_t, int); 152 #define DUMPVMCMDS(p, x, e) do { dump_vmcmds((p), (x), (e)); } while (0) 153 #else 154 #define DPRINTF(a) 155 #define COPYPRINTF(s, a, b) 156 #define DUMPVMCMDS(p, x, e) do {} while (0) 157 #endif /* DEBUG_EXEC */ 158 159 /* 160 * DTrace SDT provider definitions 161 */ 162 SDT_PROVIDER_DECLARE(proc); 163 SDT_PROBE_DEFINE1(proc, kernel, , exec, "char *"); 164 SDT_PROBE_DEFINE1(proc, kernel, , exec__success, "char *"); 165 SDT_PROBE_DEFINE1(proc, kernel, , exec__failure, "int"); 166 167 /* 168 * Exec function switch: 169 * 170 * Note that each makecmds function is responsible for loading the 171 * exec package with the necessary functions for any exec-type-specific 172 * handling. 173 * 174 * Functions for specific exec types should be defined in their own 175 * header file. 176 */ 177 static const struct execsw **execsw = NULL; 178 static int nexecs; 179 180 u_int exec_maxhdrsz; /* must not be static - used by netbsd32 */ 181 182 /* list of dynamically loaded execsw entries */ 183 static LIST_HEAD(execlist_head, exec_entry) ex_head = 184 LIST_HEAD_INITIALIZER(ex_head); 185 struct exec_entry { 186 LIST_ENTRY(exec_entry) ex_list; 187 SLIST_ENTRY(exec_entry) ex_slist; 188 const struct execsw *ex_sw; 189 }; 190 191 #ifndef __HAVE_SYSCALL_INTERN 192 void syscall(void); 193 #endif 194 195 /* NetBSD autoloadable syscalls */ 196 #ifdef MODULAR 197 #include <kern/syscalls_autoload.c> 198 #endif 199 200 /* NetBSD emul struct */ 201 struct emul emul_netbsd = { 202 .e_name = "netbsd", 203 #ifdef EMUL_NATIVEROOT 204 .e_path = EMUL_NATIVEROOT, 205 #else 206 .e_path = NULL, 207 #endif 208 #ifndef __HAVE_MINIMAL_EMUL 209 .e_flags = EMUL_HAS_SYS___syscall, 210 .e_errno = NULL, 211 .e_nosys = SYS_syscall, 212 .e_nsysent = SYS_NSYSENT, 213 #endif 214 #ifdef MODULAR 215 .e_sc_autoload = netbsd_syscalls_autoload, 216 #endif 217 .e_sysent = sysent, 218 .e_nomodbits = sysent_nomodbits, 219 #ifdef SYSCALL_DEBUG 220 .e_syscallnames = syscallnames, 221 #else 222 .e_syscallnames = NULL, 223 #endif 224 .e_sendsig = sendsig, 225 .e_trapsignal = trapsignal, 226 .e_sigcode = NULL, 227 .e_esigcode = NULL, 228 .e_sigobject = NULL, 229 .e_setregs = setregs, 230 .e_proc_exec = NULL, 231 .e_proc_fork = NULL, 232 .e_proc_exit = NULL, 233 .e_lwp_fork = NULL, 234 .e_lwp_exit = NULL, 235 #ifdef __HAVE_SYSCALL_INTERN 236 .e_syscall_intern = syscall_intern, 237 #else 238 .e_syscall = syscall, 239 #endif 240 .e_sysctlovly = NULL, 241 .e_vm_default_addr = uvm_default_mapaddr, 242 .e_usertrap = NULL, 243 .e_ucsize = sizeof(ucontext_t), 244 .e_startlwp = startlwp 245 }; 246 247 /* 248 * Exec lock. Used to control access to execsw[] structures. 249 * This must not be static so that netbsd32 can access it, too. 250 */ 251 krwlock_t exec_lock __cacheline_aligned; 252 253 /* 254 * Data used between a loadvm and execve part of an "exec" operation 255 */ 256 struct execve_data { 257 struct exec_package ed_pack; 258 struct pathbuf *ed_pathbuf; 259 struct vattr ed_attr; 260 struct ps_strings ed_arginfo; 261 char *ed_argp; 262 const char *ed_pathstring; 263 char *ed_resolvedname; 264 size_t ed_ps_strings_sz; 265 int ed_szsigcode; 266 size_t ed_argslen; 267 long ed_argc; 268 long ed_envc; 269 }; 270 271 /* 272 * data passed from parent lwp to child during a posix_spawn() 273 */ 274 struct spawn_exec_data { 275 struct execve_data sed_exec; 276 struct posix_spawn_file_actions 277 *sed_actions; 278 struct posix_spawnattr *sed_attrs; 279 struct proc *sed_parent; 280 kcondvar_t sed_cv_child_ready; 281 kmutex_t sed_mtx_child; 282 int sed_error; 283 volatile uint32_t sed_refcnt; 284 }; 285 286 static struct vm_map *exec_map; 287 static struct pool exec_pool; 288 289 static void * 290 exec_pool_alloc(struct pool *pp, int flags) 291 { 292 293 return (void *)uvm_km_alloc(exec_map, NCARGS, 0, 294 UVM_KMF_PAGEABLE | UVM_KMF_WAITVA); 295 } 296 297 static void 298 exec_pool_free(struct pool *pp, void *addr) 299 { 300 301 uvm_km_free(exec_map, (vaddr_t)addr, NCARGS, UVM_KMF_PAGEABLE); 302 } 303 304 static struct pool_allocator exec_palloc = { 305 .pa_alloc = exec_pool_alloc, 306 .pa_free = exec_pool_free, 307 .pa_pagesz = NCARGS 308 }; 309 310 static void 311 exec_path_free(struct execve_data *data) 312 { 313 pathbuf_stringcopy_put(data->ed_pathbuf, data->ed_pathstring); 314 pathbuf_destroy(data->ed_pathbuf); 315 if (data->ed_resolvedname) 316 PNBUF_PUT(data->ed_resolvedname); 317 } 318 319 static int 320 exec_resolvename(struct lwp *l, struct exec_package *epp, struct vnode *vp, 321 char **rpath) 322 { 323 int error; 324 char *p; 325 326 KASSERT(rpath != NULL); 327 328 *rpath = PNBUF_GET(); 329 error = vnode_to_path(*rpath, MAXPATHLEN, vp, l, l->l_proc); 330 if (error) { 331 DPRINTF(("%s: can't resolve name for %s, error %d\n", 332 __func__, epp->ep_kname, error)); 333 PNBUF_PUT(*rpath); 334 *rpath = NULL; 335 return error; 336 } 337 epp->ep_resolvedname = *rpath; 338 if ((p = strrchr(*rpath, '/')) != NULL) 339 epp->ep_kname = p + 1; 340 return 0; 341 } 342 343 344 /* 345 * check exec: 346 * given an "executable" described in the exec package's namei info, 347 * see what we can do with it. 348 * 349 * ON ENTRY: 350 * exec package with appropriate namei info 351 * lwp pointer of exec'ing lwp 352 * NO SELF-LOCKED VNODES 353 * 354 * ON EXIT: 355 * error: nothing held, etc. exec header still allocated. 356 * ok: filled exec package, executable's vnode (unlocked). 357 * 358 * EXEC SWITCH ENTRY: 359 * Locked vnode to check, exec package, proc. 360 * 361 * EXEC SWITCH EXIT: 362 * ok: return 0, filled exec package, executable's vnode (unlocked). 363 * error: destructive: 364 * everything deallocated execept exec header. 365 * non-destructive: 366 * error code, executable's vnode (unlocked), 367 * exec header unmodified. 368 */ 369 int 370 /*ARGSUSED*/ 371 check_exec(struct lwp *l, struct exec_package *epp, struct pathbuf *pb, 372 char **rpath) 373 { 374 int error, i; 375 struct vnode *vp; 376 size_t resid; 377 378 if (epp->ep_resolvedname) { 379 struct nameidata nd; 380 381 // grab the absolute pathbuf here before namei() trashes it. 382 pathbuf_copystring(pb, epp->ep_resolvedname, PATH_MAX); 383 NDINIT(&nd, LOOKUP, FOLLOW | LOCKLEAF | TRYEMULROOT, pb); 384 385 /* first get the vnode */ 386 if ((error = namei(&nd)) != 0) 387 return error; 388 389 epp->ep_vp = vp = nd.ni_vp; 390 #ifdef DIAGNOSTIC 391 /* paranoia (take this out once namei stuff stabilizes) */ 392 memset(nd.ni_pnbuf, '~', PATH_MAX); 393 #endif 394 } else { 395 struct file *fp; 396 397 if ((error = fd_getvnode(epp->ep_xfd, &fp)) != 0) 398 return error; 399 epp->ep_vp = vp = fp->f_vnode; 400 vref(vp); 401 fd_putfile(epp->ep_xfd); 402 if ((error = exec_resolvename(l, epp, vp, rpath)) != 0) 403 return error; 404 vn_lock(vp, LK_EXCLUSIVE | LK_RETRY); 405 } 406 407 /* check access and type */ 408 if (vp->v_type != VREG) { 409 error = EACCES; 410 goto bad1; 411 } 412 if ((error = VOP_ACCESS(vp, VEXEC, l->l_cred)) != 0) 413 goto bad1; 414 415 /* get attributes */ 416 /* XXX VOP_GETATTR is the only thing that needs LK_EXCLUSIVE here */ 417 if ((error = VOP_GETATTR(vp, epp->ep_vap, l->l_cred)) != 0) 418 goto bad1; 419 420 /* Check mount point */ 421 if (vp->v_mount->mnt_flag & MNT_NOEXEC) { 422 error = EACCES; 423 goto bad1; 424 } 425 if (vp->v_mount->mnt_flag & MNT_NOSUID) 426 epp->ep_vap->va_mode &= ~(S_ISUID | S_ISGID); 427 428 /* try to open it */ 429 if ((error = VOP_OPEN(vp, FREAD, l->l_cred)) != 0) 430 goto bad1; 431 432 /* now we have the file, get the exec header */ 433 error = vn_rdwr(UIO_READ, vp, epp->ep_hdr, epp->ep_hdrlen, 0, 434 UIO_SYSSPACE, IO_NODELOCKED, l->l_cred, &resid, NULL); 435 if (error) 436 goto bad1; 437 438 /* unlock vp, since we need it unlocked from here on out. */ 439 VOP_UNLOCK(vp); 440 441 #if NVERIEXEC > 0 442 error = veriexec_verify(l, vp, 443 epp->ep_resolvedname ? epp->ep_resolvedname : epp->ep_kname, 444 epp->ep_flags & EXEC_INDIR ? VERIEXEC_INDIRECT : VERIEXEC_DIRECT, 445 NULL); 446 if (error) 447 goto bad2; 448 #endif /* NVERIEXEC > 0 */ 449 450 #ifdef PAX_SEGVGUARD 451 error = pax_segvguard(l, vp, epp->ep_resolvedname, false); 452 if (error) 453 goto bad2; 454 #endif /* PAX_SEGVGUARD */ 455 456 epp->ep_hdrvalid = epp->ep_hdrlen - resid; 457 458 /* 459 * Set up default address space limits. Can be overridden 460 * by individual exec packages. 461 */ 462 epp->ep_vm_minaddr = exec_vm_minaddr(VM_MIN_ADDRESS); 463 epp->ep_vm_maxaddr = VM_MAXUSER_ADDRESS; 464 465 /* 466 * set up the vmcmds for creation of the process 467 * address space 468 */ 469 error = ENOEXEC; 470 for (i = 0; i < nexecs; i++) { 471 int newerror; 472 473 epp->ep_esch = execsw[i]; 474 newerror = (*execsw[i]->es_makecmds)(l, epp); 475 476 if (!newerror) { 477 /* Seems ok: check that entry point is not too high */ 478 if (epp->ep_entry >= epp->ep_vm_maxaddr) { 479 #ifdef DIAGNOSTIC 480 printf("%s: rejecting %p due to " 481 "too high entry address (>= %p)\n", 482 __func__, (void *)epp->ep_entry, 483 (void *)epp->ep_vm_maxaddr); 484 #endif 485 error = ENOEXEC; 486 break; 487 } 488 /* Seems ok: check that entry point is not too low */ 489 if (epp->ep_entry < epp->ep_vm_minaddr) { 490 #ifdef DIAGNOSTIC 491 printf("%s: rejecting %p due to " 492 "too low entry address (< %p)\n", 493 __func__, (void *)epp->ep_entry, 494 (void *)epp->ep_vm_minaddr); 495 #endif 496 error = ENOEXEC; 497 break; 498 } 499 500 /* check limits */ 501 #ifdef DIAGNOSTIC 502 #define LMSG "%s: rejecting due to %s limit (%ju > %ju)\n" 503 #endif 504 #ifdef MAXTSIZ 505 if (epp->ep_tsize > MAXTSIZ) { 506 #ifdef DIAGNOSTIC 507 printf(LMSG, __func__, "text", 508 (uintmax_t)epp->ep_tsize, 509 (uintmax_t)MAXTSIZ); 510 #endif 511 error = ENOMEM; 512 break; 513 } 514 #endif 515 vsize_t dlimit = 516 (vsize_t)l->l_proc->p_rlimit[RLIMIT_DATA].rlim_cur; 517 if (epp->ep_dsize > dlimit) { 518 #ifdef DIAGNOSTIC 519 printf(LMSG, __func__, "data", 520 (uintmax_t)epp->ep_dsize, 521 (uintmax_t)dlimit); 522 #endif 523 error = ENOMEM; 524 break; 525 } 526 return 0; 527 } 528 529 /* 530 * Reset all the fields that may have been modified by the 531 * loader. 532 */ 533 KASSERT(epp->ep_emul_arg == NULL); 534 if (epp->ep_emul_root != NULL) { 535 vrele(epp->ep_emul_root); 536 epp->ep_emul_root = NULL; 537 } 538 if (epp->ep_interp != NULL) { 539 vrele(epp->ep_interp); 540 epp->ep_interp = NULL; 541 } 542 epp->ep_pax_flags = 0; 543 544 /* make sure the first "interesting" error code is saved. */ 545 if (error == ENOEXEC) 546 error = newerror; 547 548 if (epp->ep_flags & EXEC_DESTR) 549 /* Error from "#!" code, tidied up by recursive call */ 550 return error; 551 } 552 553 /* not found, error */ 554 555 /* 556 * free any vmspace-creation commands, 557 * and release their references 558 */ 559 kill_vmcmds(&epp->ep_vmcmds); 560 561 #if NVERIEXEC > 0 || defined(PAX_SEGVGUARD) 562 bad2: 563 #endif 564 /* 565 * close and release the vnode, restore the old one, free the 566 * pathname buf, and punt. 567 */ 568 vn_lock(vp, LK_EXCLUSIVE | LK_RETRY); 569 VOP_CLOSE(vp, FREAD, l->l_cred); 570 vput(vp); 571 return error; 572 573 bad1: 574 /* 575 * free the namei pathname buffer, and put the vnode 576 * (which we don't yet have open). 577 */ 578 vput(vp); /* was still locked */ 579 return error; 580 } 581 582 #ifdef __MACHINE_STACK_GROWS_UP 583 #define STACK_PTHREADSPACE NBPG 584 #else 585 #define STACK_PTHREADSPACE 0 586 #endif 587 588 static int 589 execve_fetch_element(char * const *array, size_t index, char **value) 590 { 591 return copyin(array + index, value, sizeof(*value)); 592 } 593 594 /* 595 * exec system call 596 */ 597 int 598 sys_execve(struct lwp *l, const struct sys_execve_args *uap, register_t *retval) 599 { 600 /* { 601 syscallarg(const char *) path; 602 syscallarg(char * const *) argp; 603 syscallarg(char * const *) envp; 604 } */ 605 606 return execve1(l, true, SCARG(uap, path), -1, SCARG(uap, argp), 607 SCARG(uap, envp), execve_fetch_element); 608 } 609 610 int 611 sys_fexecve(struct lwp *l, const struct sys_fexecve_args *uap, 612 register_t *retval) 613 { 614 /* { 615 syscallarg(int) fd; 616 syscallarg(char * const *) argp; 617 syscallarg(char * const *) envp; 618 } */ 619 620 return execve1(l, false, NULL, SCARG(uap, fd), SCARG(uap, argp), 621 SCARG(uap, envp), execve_fetch_element); 622 } 623 624 /* 625 * Load modules to try and execute an image that we do not understand. 626 * If no execsw entries are present, we load those likely to be needed 627 * in order to run native images only. Otherwise, we autoload all 628 * possible modules that could let us run the binary. XXX lame 629 */ 630 static void 631 exec_autoload(void) 632 { 633 #ifdef MODULAR 634 static const char * const native[] = { 635 "exec_elf32", 636 "exec_elf64", 637 "exec_script", 638 NULL 639 }; 640 static const char * const compat[] = { 641 "exec_elf32", 642 "exec_elf64", 643 "exec_script", 644 "exec_aout", 645 "exec_coff", 646 "exec_ecoff", 647 "compat_aoutm68k", 648 "compat_netbsd32", 649 #if 0 650 "compat_linux", 651 "compat_linux32", 652 #endif 653 "compat_sunos", 654 "compat_sunos32", 655 "compat_ultrix", 656 NULL 657 }; 658 char const * const *list; 659 int i; 660 661 list = nexecs == 0 ? native : compat; 662 for (i = 0; list[i] != NULL; i++) { 663 if (module_autoload(list[i], MODULE_CLASS_EXEC) != 0) { 664 continue; 665 } 666 yield(); 667 } 668 #endif 669 } 670 671 /* 672 * Copy the user or kernel supplied upath to the allocated pathbuffer pbp 673 * making it absolute in the process, by prepending the current working 674 * directory if it is not. If offs is supplied it will contain the offset 675 * where the original supplied copy of upath starts. 676 */ 677 int 678 exec_makepathbuf(struct lwp *l, const char *upath, enum uio_seg seg, 679 struct pathbuf **pbp, size_t *offs) 680 { 681 char *path, *bp; 682 size_t len, tlen; 683 int error; 684 struct cwdinfo *cwdi; 685 686 path = PNBUF_GET(); 687 if (seg == UIO_SYSSPACE) { 688 error = copystr(upath, path, MAXPATHLEN, &len); 689 } else { 690 error = copyinstr(upath, path, MAXPATHLEN, &len); 691 } 692 if (error) 693 goto err; 694 695 if (path[0] == '/') { 696 if (offs) 697 *offs = 0; 698 goto out; 699 } 700 701 len++; 702 if (len + 1 >= MAXPATHLEN) { 703 error = ENAMETOOLONG; 704 goto err; 705 } 706 bp = path + MAXPATHLEN - len; 707 memmove(bp, path, len); 708 *(--bp) = '/'; 709 710 cwdi = l->l_proc->p_cwdi; 711 rw_enter(&cwdi->cwdi_lock, RW_READER); 712 error = getcwd_common(cwdi->cwdi_cdir, NULL, &bp, path, MAXPATHLEN / 2, 713 GETCWD_CHECK_ACCESS, l); 714 rw_exit(&cwdi->cwdi_lock); 715 716 if (error) 717 goto err; 718 tlen = path + MAXPATHLEN - bp; 719 720 memmove(path, bp, tlen); 721 path[tlen - 1] = '\0'; 722 if (offs) 723 *offs = tlen - len; 724 out: 725 *pbp = pathbuf_assimilate(path); 726 return 0; 727 err: 728 PNBUF_PUT(path); 729 return error; 730 } 731 732 vaddr_t 733 exec_vm_minaddr(vaddr_t va_min) 734 { 735 /* 736 * Increase va_min if we don't want NULL to be mappable by the 737 * process. 738 */ 739 #define VM_MIN_GUARD PAGE_SIZE 740 if (user_va0_disable && (va_min < VM_MIN_GUARD)) 741 return VM_MIN_GUARD; 742 return va_min; 743 } 744 745 static int 746 execve_loadvm(struct lwp *l, bool has_path, const char *path, int fd, 747 char * const *args, char * const *envs, 748 execve_fetch_element_t fetch_element, 749 struct execve_data * restrict data) 750 { 751 struct exec_package * const epp = &data->ed_pack; 752 int error; 753 struct proc *p; 754 char *dp; 755 u_int modgen; 756 757 KASSERT(data != NULL); 758 759 p = l->l_proc; 760 modgen = 0; 761 762 SDT_PROBE(proc, kernel, , exec, path, 0, 0, 0, 0); 763 764 /* 765 * Check if we have exceeded our number of processes limit. 766 * This is so that we handle the case where a root daemon 767 * forked, ran setuid to become the desired user and is trying 768 * to exec. The obvious place to do the reference counting check 769 * is setuid(), but we don't do the reference counting check there 770 * like other OS's do because then all the programs that use setuid() 771 * must be modified to check the return code of setuid() and exit(). 772 * It is dangerous to make setuid() fail, because it fails open and 773 * the program will continue to run as root. If we make it succeed 774 * and return an error code, again we are not enforcing the limit. 775 * The best place to enforce the limit is here, when the process tries 776 * to execute a new image, because eventually the process will need 777 * to call exec in order to do something useful. 778 */ 779 retry: 780 if (p->p_flag & PK_SUGID) { 781 if (kauth_authorize_process(l->l_cred, KAUTH_PROCESS_RLIMIT, 782 p, KAUTH_ARG(KAUTH_REQ_PROCESS_RLIMIT_BYPASS), 783 &p->p_rlimit[RLIMIT_NPROC], 784 KAUTH_ARG(RLIMIT_NPROC)) != 0 && 785 chgproccnt(kauth_cred_getuid(l->l_cred), 0) > 786 p->p_rlimit[RLIMIT_NPROC].rlim_cur) 787 return EAGAIN; 788 } 789 790 /* 791 * Drain existing references and forbid new ones. The process 792 * should be left alone until we're done here. This is necessary 793 * to avoid race conditions - e.g. in ptrace() - that might allow 794 * a local user to illicitly obtain elevated privileges. 795 */ 796 rw_enter(&p->p_reflock, RW_WRITER); 797 798 if (has_path) { 799 size_t offs; 800 /* 801 * Init the namei data to point the file user's program name. 802 * This is done here rather than in check_exec(), so that it's 803 * possible to override this settings if any of makecmd/probe 804 * functions call check_exec() recursively - for example, 805 * see exec_script_makecmds(). 806 */ 807 if ((error = exec_makepathbuf(l, path, UIO_USERSPACE, 808 &data->ed_pathbuf, &offs)) != 0) 809 goto clrflg; 810 data->ed_pathstring = pathbuf_stringcopy_get(data->ed_pathbuf); 811 epp->ep_kname = data->ed_pathstring + offs; 812 data->ed_resolvedname = PNBUF_GET(); 813 epp->ep_resolvedname = data->ed_resolvedname; 814 epp->ep_xfd = -1; 815 } else { 816 data->ed_pathbuf = pathbuf_assimilate(strcpy(PNBUF_GET(), "/")); 817 data->ed_pathstring = pathbuf_stringcopy_get(data->ed_pathbuf); 818 epp->ep_kname = "*fexecve*"; 819 data->ed_resolvedname = NULL; 820 epp->ep_resolvedname = NULL; 821 epp->ep_xfd = fd; 822 } 823 824 825 /* 826 * initialize the fields of the exec package. 827 */ 828 epp->ep_hdr = kmem_alloc(exec_maxhdrsz, KM_SLEEP); 829 epp->ep_hdrlen = exec_maxhdrsz; 830 epp->ep_hdrvalid = 0; 831 epp->ep_emul_arg = NULL; 832 epp->ep_emul_arg_free = NULL; 833 memset(&epp->ep_vmcmds, 0, sizeof(epp->ep_vmcmds)); 834 epp->ep_vap = &data->ed_attr; 835 epp->ep_flags = (p->p_flag & PK_32) ? EXEC_FROM32 : 0; 836 MD_TOPDOWN_INIT(epp); 837 epp->ep_emul_root = NULL; 838 epp->ep_interp = NULL; 839 epp->ep_esch = NULL; 840 epp->ep_pax_flags = 0; 841 memset(epp->ep_machine_arch, 0, sizeof(epp->ep_machine_arch)); 842 843 rw_enter(&exec_lock, RW_READER); 844 845 /* see if we can run it. */ 846 if ((error = check_exec(l, epp, data->ed_pathbuf, 847 &data->ed_resolvedname)) != 0) { 848 if (error != ENOENT && error != EACCES && error != ENOEXEC) { 849 DPRINTF(("%s: check exec failed for %s, error %d\n", 850 __func__, epp->ep_kname, error)); 851 } 852 goto freehdr; 853 } 854 855 /* allocate an argument buffer */ 856 data->ed_argp = pool_get(&exec_pool, PR_WAITOK); 857 KASSERT(data->ed_argp != NULL); 858 dp = data->ed_argp; 859 860 if ((error = copyinargs(data, args, envs, fetch_element, &dp)) != 0) { 861 goto bad; 862 } 863 864 /* 865 * Calculate the new stack size. 866 */ 867 868 #ifdef __MACHINE_STACK_GROWS_UP 869 /* 870 * copyargs() fills argc/argv/envp from the lower address even on 871 * __MACHINE_STACK_GROWS_UP machines. Reserve a few words just below the SP 872 * so that _rtld() use it. 873 */ 874 #define RTLD_GAP 32 875 #else 876 #define RTLD_GAP 0 877 #endif 878 879 const size_t argenvstrlen = (char *)ALIGN(dp) - data->ed_argp; 880 881 data->ed_argslen = calcargs(data, argenvstrlen); 882 883 const size_t len = calcstack(data, pax_aslr_stack_gap(epp) + RTLD_GAP); 884 885 if (len > epp->ep_ssize) { 886 /* in effect, compare to initial limit */ 887 DPRINTF(("%s: stack limit exceeded %zu\n", __func__, len)); 888 error = ENOMEM; 889 goto bad; 890 } 891 /* adjust "active stack depth" for process VSZ */ 892 epp->ep_ssize = len; 893 894 return 0; 895 896 bad: 897 /* free the vmspace-creation commands, and release their references */ 898 kill_vmcmds(&epp->ep_vmcmds); 899 /* kill any opened file descriptor, if necessary */ 900 if (epp->ep_flags & EXEC_HASFD) { 901 epp->ep_flags &= ~EXEC_HASFD; 902 fd_close(epp->ep_fd); 903 } 904 /* close and put the exec'd file */ 905 vn_lock(epp->ep_vp, LK_EXCLUSIVE | LK_RETRY); 906 VOP_CLOSE(epp->ep_vp, FREAD, l->l_cred); 907 vput(epp->ep_vp); 908 pool_put(&exec_pool, data->ed_argp); 909 910 freehdr: 911 kmem_free(epp->ep_hdr, epp->ep_hdrlen); 912 if (epp->ep_emul_root != NULL) 913 vrele(epp->ep_emul_root); 914 if (epp->ep_interp != NULL) 915 vrele(epp->ep_interp); 916 917 rw_exit(&exec_lock); 918 919 exec_path_free(data); 920 921 clrflg: 922 rw_exit(&p->p_reflock); 923 924 if (modgen != module_gen && error == ENOEXEC) { 925 modgen = module_gen; 926 exec_autoload(); 927 goto retry; 928 } 929 930 SDT_PROBE(proc, kernel, , exec__failure, error, 0, 0, 0, 0); 931 return error; 932 } 933 934 static int 935 execve_dovmcmds(struct lwp *l, struct execve_data * restrict data) 936 { 937 struct exec_package * const epp = &data->ed_pack; 938 struct proc *p = l->l_proc; 939 struct exec_vmcmd *base_vcp; 940 int error = 0; 941 size_t i; 942 943 /* record proc's vnode, for use by procfs and others */ 944 if (p->p_textvp) 945 vrele(p->p_textvp); 946 vref(epp->ep_vp); 947 p->p_textvp = epp->ep_vp; 948 949 /* create the new process's VM space by running the vmcmds */ 950 KASSERTMSG(epp->ep_vmcmds.evs_used != 0, "%s: no vmcmds", __func__); 951 952 #ifdef TRACE_EXEC 953 DUMPVMCMDS(epp, 0, 0); 954 #endif 955 956 base_vcp = NULL; 957 958 for (i = 0; i < epp->ep_vmcmds.evs_used && !error; i++) { 959 struct exec_vmcmd *vcp; 960 961 vcp = &epp->ep_vmcmds.evs_cmds[i]; 962 if (vcp->ev_flags & VMCMD_RELATIVE) { 963 KASSERTMSG(base_vcp != NULL, 964 "%s: relative vmcmd with no base", __func__); 965 KASSERTMSG((vcp->ev_flags & VMCMD_BASE) == 0, 966 "%s: illegal base & relative vmcmd", __func__); 967 vcp->ev_addr += base_vcp->ev_addr; 968 } 969 error = (*vcp->ev_proc)(l, vcp); 970 if (error) 971 DUMPVMCMDS(epp, i, error); 972 if (vcp->ev_flags & VMCMD_BASE) 973 base_vcp = vcp; 974 } 975 976 /* free the vmspace-creation commands, and release their references */ 977 kill_vmcmds(&epp->ep_vmcmds); 978 979 vn_lock(epp->ep_vp, LK_EXCLUSIVE | LK_RETRY); 980 VOP_CLOSE(epp->ep_vp, FREAD, l->l_cred); 981 vput(epp->ep_vp); 982 983 /* if an error happened, deallocate and punt */ 984 if (error != 0) { 985 DPRINTF(("%s: vmcmd %zu failed: %d\n", __func__, i - 1, error)); 986 } 987 return error; 988 } 989 990 static void 991 execve_free_data(struct execve_data *data) 992 { 993 struct exec_package * const epp = &data->ed_pack; 994 995 /* free the vmspace-creation commands, and release their references */ 996 kill_vmcmds(&epp->ep_vmcmds); 997 /* kill any opened file descriptor, if necessary */ 998 if (epp->ep_flags & EXEC_HASFD) { 999 epp->ep_flags &= ~EXEC_HASFD; 1000 fd_close(epp->ep_fd); 1001 } 1002 1003 /* close and put the exec'd file */ 1004 vn_lock(epp->ep_vp, LK_EXCLUSIVE | LK_RETRY); 1005 VOP_CLOSE(epp->ep_vp, FREAD, curlwp->l_cred); 1006 vput(epp->ep_vp); 1007 pool_put(&exec_pool, data->ed_argp); 1008 1009 kmem_free(epp->ep_hdr, epp->ep_hdrlen); 1010 if (epp->ep_emul_root != NULL) 1011 vrele(epp->ep_emul_root); 1012 if (epp->ep_interp != NULL) 1013 vrele(epp->ep_interp); 1014 1015 exec_path_free(data); 1016 } 1017 1018 static void 1019 pathexec(struct proc *p, const char *resolvedname) 1020 { 1021 /* set command name & other accounting info */ 1022 const char *cmdname; 1023 1024 if (resolvedname == NULL) { 1025 cmdname = "*fexecve*"; 1026 resolvedname = "/"; 1027 } else { 1028 cmdname = strrchr(resolvedname, '/') + 1; 1029 } 1030 KASSERTMSG(resolvedname[0] == '/', "bad resolvedname `%s'", 1031 resolvedname); 1032 1033 strlcpy(p->p_comm, cmdname, sizeof(p->p_comm)); 1034 1035 kmem_strfree(p->p_path); 1036 p->p_path = kmem_strdupsize(resolvedname, NULL, KM_SLEEP); 1037 } 1038 1039 /* XXX elsewhere */ 1040 static int 1041 credexec(struct lwp *l, struct vattr *attr) 1042 { 1043 struct proc *p = l->l_proc; 1044 int error; 1045 1046 /* 1047 * Deal with set[ug]id. MNT_NOSUID has already been used to disable 1048 * s[ug]id. It's OK to check for PSL_TRACED here as we have blocked 1049 * out additional references on the process for the moment. 1050 */ 1051 if ((p->p_slflag & PSL_TRACED) == 0 && 1052 1053 (((attr->va_mode & S_ISUID) != 0 && 1054 kauth_cred_geteuid(l->l_cred) != attr->va_uid) || 1055 1056 ((attr->va_mode & S_ISGID) != 0 && 1057 kauth_cred_getegid(l->l_cred) != attr->va_gid))) { 1058 /* 1059 * Mark the process as SUGID before we do 1060 * anything that might block. 1061 */ 1062 proc_crmod_enter(); 1063 proc_crmod_leave(NULL, NULL, true); 1064 1065 /* Make sure file descriptors 0..2 are in use. */ 1066 if ((error = fd_checkstd()) != 0) { 1067 DPRINTF(("%s: fdcheckstd failed %d\n", 1068 __func__, error)); 1069 return error; 1070 } 1071 1072 /* 1073 * Copy the credential so other references don't see our 1074 * changes. 1075 */ 1076 l->l_cred = kauth_cred_copy(l->l_cred); 1077 #ifdef KTRACE 1078 /* 1079 * If the persistent trace flag isn't set, turn off. 1080 */ 1081 if (p->p_tracep) { 1082 mutex_enter(&ktrace_lock); 1083 if (!(p->p_traceflag & KTRFAC_PERSISTENT)) 1084 ktrderef(p); 1085 mutex_exit(&ktrace_lock); 1086 } 1087 #endif 1088 if (attr->va_mode & S_ISUID) 1089 kauth_cred_seteuid(l->l_cred, attr->va_uid); 1090 if (attr->va_mode & S_ISGID) 1091 kauth_cred_setegid(l->l_cred, attr->va_gid); 1092 } else { 1093 if (kauth_cred_geteuid(l->l_cred) == 1094 kauth_cred_getuid(l->l_cred) && 1095 kauth_cred_getegid(l->l_cred) == 1096 kauth_cred_getgid(l->l_cred)) 1097 p->p_flag &= ~PK_SUGID; 1098 } 1099 1100 /* 1101 * Copy the credential so other references don't see our changes. 1102 * Test to see if this is necessary first, since in the common case 1103 * we won't need a private reference. 1104 */ 1105 if (kauth_cred_geteuid(l->l_cred) != kauth_cred_getsvuid(l->l_cred) || 1106 kauth_cred_getegid(l->l_cred) != kauth_cred_getsvgid(l->l_cred)) { 1107 l->l_cred = kauth_cred_copy(l->l_cred); 1108 kauth_cred_setsvuid(l->l_cred, kauth_cred_geteuid(l->l_cred)); 1109 kauth_cred_setsvgid(l->l_cred, kauth_cred_getegid(l->l_cred)); 1110 } 1111 1112 /* Update the master credentials. */ 1113 if (l->l_cred != p->p_cred) { 1114 kauth_cred_t ocred; 1115 1116 kauth_cred_hold(l->l_cred); 1117 mutex_enter(p->p_lock); 1118 ocred = p->p_cred; 1119 p->p_cred = l->l_cred; 1120 mutex_exit(p->p_lock); 1121 kauth_cred_free(ocred); 1122 } 1123 1124 return 0; 1125 } 1126 1127 static void 1128 emulexec(struct lwp *l, struct exec_package *epp) 1129 { 1130 struct proc *p = l->l_proc; 1131 1132 /* The emulation root will usually have been found when we looked 1133 * for the elf interpreter (or similar), if not look now. */ 1134 if (epp->ep_esch->es_emul->e_path != NULL && 1135 epp->ep_emul_root == NULL) 1136 emul_find_root(l, epp); 1137 1138 /* Any old emulation root got removed by fdcloseexec */ 1139 rw_enter(&p->p_cwdi->cwdi_lock, RW_WRITER); 1140 p->p_cwdi->cwdi_edir = epp->ep_emul_root; 1141 rw_exit(&p->p_cwdi->cwdi_lock); 1142 epp->ep_emul_root = NULL; 1143 if (epp->ep_interp != NULL) 1144 vrele(epp->ep_interp); 1145 1146 /* 1147 * Call emulation specific exec hook. This can setup per-process 1148 * p->p_emuldata or do any other per-process stuff an emulation needs. 1149 * 1150 * If we are executing process of different emulation than the 1151 * original forked process, call e_proc_exit() of the old emulation 1152 * first, then e_proc_exec() of new emulation. If the emulation is 1153 * same, the exec hook code should deallocate any old emulation 1154 * resources held previously by this process. 1155 */ 1156 if (p->p_emul && p->p_emul->e_proc_exit 1157 && p->p_emul != epp->ep_esch->es_emul) 1158 (*p->p_emul->e_proc_exit)(p); 1159 1160 /* 1161 * Call exec hook. Emulation code may NOT store reference to anything 1162 * from &pack. 1163 */ 1164 if (epp->ep_esch->es_emul->e_proc_exec) 1165 (*epp->ep_esch->es_emul->e_proc_exec)(p, epp); 1166 1167 /* update p_emul, the old value is no longer needed */ 1168 p->p_emul = epp->ep_esch->es_emul; 1169 1170 /* ...and the same for p_execsw */ 1171 p->p_execsw = epp->ep_esch; 1172 1173 #ifdef __HAVE_SYSCALL_INTERN 1174 (*p->p_emul->e_syscall_intern)(p); 1175 #endif 1176 ktremul(); 1177 } 1178 1179 static int 1180 execve_runproc(struct lwp *l, struct execve_data * restrict data, 1181 bool no_local_exec_lock, bool is_spawn) 1182 { 1183 struct exec_package * const epp = &data->ed_pack; 1184 int error = 0; 1185 struct proc *p; 1186 struct vmspace *vm; 1187 1188 /* 1189 * In case of a posix_spawn operation, the child doing the exec 1190 * might not hold the reader lock on exec_lock, but the parent 1191 * will do this instead. 1192 */ 1193 KASSERT(no_local_exec_lock || rw_lock_held(&exec_lock)); 1194 KASSERT(!no_local_exec_lock || is_spawn); 1195 KASSERT(data != NULL); 1196 1197 p = l->l_proc; 1198 1199 /* Get rid of other LWPs. */ 1200 if (p->p_nlwps > 1) { 1201 mutex_enter(p->p_lock); 1202 exit_lwps(l); 1203 mutex_exit(p->p_lock); 1204 } 1205 KDASSERT(p->p_nlwps == 1); 1206 1207 /* 1208 * All of the other LWPs got rid of their robust futexes 1209 * when they exited above, but we might still have some 1210 * to dispose of. Do that now. 1211 */ 1212 if (__predict_false(l->l_robust_head != 0)) { 1213 futex_release_all_lwp(l); 1214 /* 1215 * Since this LWP will live on with a different 1216 * program image, we need to clear the robust 1217 * futex list pointer here. 1218 */ 1219 l->l_robust_head = 0; 1220 } 1221 1222 /* Destroy any lwpctl info. */ 1223 if (p->p_lwpctl != NULL) 1224 lwp_ctl_exit(); 1225 1226 /* Remove POSIX timers */ 1227 ptimers_free(p, TIMERS_POSIX); 1228 1229 /* Set the PaX flags. */ 1230 pax_set_flags(epp, p); 1231 1232 /* 1233 * Do whatever is necessary to prepare the address space 1234 * for remapping. Note that this might replace the current 1235 * vmspace with another! 1236 * 1237 * vfork(): do not touch any user space data in the new child 1238 * until we have awoken the parent below, or it will defeat 1239 * lazy pmap switching (on x86). 1240 */ 1241 if (is_spawn) 1242 uvmspace_spawn(l, epp->ep_vm_minaddr, 1243 epp->ep_vm_maxaddr, 1244 epp->ep_flags & EXEC_TOPDOWN_VM); 1245 else 1246 uvmspace_exec(l, epp->ep_vm_minaddr, 1247 epp->ep_vm_maxaddr, 1248 epp->ep_flags & EXEC_TOPDOWN_VM); 1249 vm = p->p_vmspace; 1250 1251 vm->vm_taddr = (void *)epp->ep_taddr; 1252 vm->vm_tsize = btoc(epp->ep_tsize); 1253 vm->vm_daddr = (void*)epp->ep_daddr; 1254 vm->vm_dsize = btoc(epp->ep_dsize); 1255 vm->vm_ssize = btoc(epp->ep_ssize); 1256 vm->vm_issize = 0; 1257 vm->vm_maxsaddr = (void *)epp->ep_maxsaddr; 1258 vm->vm_minsaddr = (void *)epp->ep_minsaddr; 1259 1260 pax_aslr_init_vm(l, vm, epp); 1261 1262 cwdexec(p); 1263 fd_closeexec(); /* handle close on exec */ 1264 1265 if (__predict_false(ktrace_on)) 1266 fd_ktrexecfd(); 1267 1268 execsigs(p); /* reset caught signals */ 1269 1270 mutex_enter(p->p_lock); 1271 l->l_ctxlink = NULL; /* reset ucontext link */ 1272 p->p_acflag &= ~AFORK; 1273 p->p_flag |= PK_EXEC; 1274 mutex_exit(p->p_lock); 1275 1276 error = credexec(l, &data->ed_attr); 1277 if (error) 1278 goto exec_abort; 1279 1280 #if defined(__HAVE_RAS) 1281 /* 1282 * Remove all RASs from the address space. 1283 */ 1284 ras_purgeall(); 1285 #endif 1286 1287 /* 1288 * Stop profiling. 1289 */ 1290 if ((p->p_stflag & PST_PROFIL) != 0) { 1291 mutex_spin_enter(&p->p_stmutex); 1292 stopprofclock(p); 1293 mutex_spin_exit(&p->p_stmutex); 1294 } 1295 1296 /* 1297 * It's OK to test PL_PPWAIT unlocked here, as other LWPs have 1298 * exited and exec()/exit() are the only places it will be cleared. 1299 * 1300 * Once the parent has been awoken, curlwp may teleport to a new CPU 1301 * in sched_vforkexec(), and it's then OK to start messing with user 1302 * data. See comment above. 1303 */ 1304 if ((p->p_lflag & PL_PPWAIT) != 0) { 1305 bool samecpu; 1306 lwp_t *lp; 1307 1308 mutex_enter(&proc_lock); 1309 lp = p->p_vforklwp; 1310 p->p_vforklwp = NULL; 1311 l->l_lwpctl = NULL; /* was on loan from blocked parent */ 1312 cv_broadcast(&lp->l_waitcv); 1313 1314 /* Clear flags after cv_broadcast() (scheduler needs them). */ 1315 p->p_lflag &= ~PL_PPWAIT; 1316 lp->l_vforkwaiting = false; 1317 1318 /* If parent is still on same CPU, teleport curlwp elsewhere. */ 1319 samecpu = (lp->l_cpu == curlwp->l_cpu); 1320 mutex_exit(&proc_lock); 1321 1322 /* Give the parent its CPU back - find a new home. */ 1323 KASSERT(!is_spawn); 1324 sched_vforkexec(l, samecpu); 1325 } 1326 1327 /* Now map address space. */ 1328 error = execve_dovmcmds(l, data); 1329 if (error != 0) 1330 goto exec_abort; 1331 1332 pathexec(p, epp->ep_resolvedname); 1333 1334 char * const newstack = STACK_GROW(vm->vm_minsaddr, epp->ep_ssize); 1335 1336 error = copyoutargs(data, l, newstack); 1337 if (error != 0) 1338 goto exec_abort; 1339 1340 doexechooks(p); 1341 1342 /* 1343 * Set initial SP at the top of the stack. 1344 * 1345 * Note that on machines where stack grows up (e.g. hppa), SP points to 1346 * the end of arg/env strings. Userland guesses the address of argc 1347 * via ps_strings::ps_argvstr. 1348 */ 1349 1350 /* Setup new registers and do misc. setup. */ 1351 (*epp->ep_esch->es_emul->e_setregs)(l, epp, (vaddr_t)newstack); 1352 if (epp->ep_esch->es_setregs) 1353 (*epp->ep_esch->es_setregs)(l, epp, (vaddr_t)newstack); 1354 1355 /* Provide a consistent LWP private setting */ 1356 (void)lwp_setprivate(l, NULL); 1357 1358 /* Discard all PCU state; need to start fresh */ 1359 pcu_discard_all(l); 1360 1361 /* map the process's signal trampoline code */ 1362 if ((error = exec_sigcode_map(p, epp->ep_esch->es_emul)) != 0) { 1363 DPRINTF(("%s: map sigcode failed %d\n", __func__, error)); 1364 goto exec_abort; 1365 } 1366 1367 pool_put(&exec_pool, data->ed_argp); 1368 1369 /* 1370 * Notify anyone who might care that we've exec'd. 1371 * 1372 * This is slightly racy; someone could sneak in and 1373 * attach a knote after we've decided not to notify, 1374 * or vice-versa, but that's not particularly bothersome. 1375 * knote_proc_exec() will acquire p->p_lock as needed. 1376 */ 1377 if (!SLIST_EMPTY(&p->p_klist)) { 1378 knote_proc_exec(p); 1379 } 1380 1381 kmem_free(epp->ep_hdr, epp->ep_hdrlen); 1382 1383 SDT_PROBE(proc, kernel, , exec__success, epp->ep_kname, 0, 0, 0, 0); 1384 1385 emulexec(l, epp); 1386 1387 /* Allow new references from the debugger/procfs. */ 1388 rw_exit(&p->p_reflock); 1389 if (!no_local_exec_lock) 1390 rw_exit(&exec_lock); 1391 1392 mutex_enter(&proc_lock); 1393 1394 /* posix_spawn(3) reports a single event with implied exec(3) */ 1395 if ((p->p_slflag & PSL_TRACED) && !is_spawn) { 1396 mutex_enter(p->p_lock); 1397 eventswitch(TRAP_EXEC, 0, 0); 1398 mutex_enter(&proc_lock); 1399 } 1400 1401 if (p->p_sflag & PS_STOPEXEC) { 1402 ksiginfoq_t kq; 1403 1404 KERNEL_UNLOCK_ALL(l, &l->l_biglocks); 1405 p->p_pptr->p_nstopchild++; 1406 p->p_waited = 0; 1407 mutex_enter(p->p_lock); 1408 ksiginfo_queue_init(&kq); 1409 sigclearall(p, &contsigmask, &kq); 1410 lwp_lock(l); 1411 l->l_stat = LSSTOP; 1412 p->p_stat = SSTOP; 1413 p->p_nrlwps--; 1414 lwp_unlock(l); 1415 mutex_exit(p->p_lock); 1416 mutex_exit(&proc_lock); 1417 lwp_lock(l); 1418 spc_lock(l->l_cpu); 1419 mi_switch(l); 1420 ksiginfo_queue_drain(&kq); 1421 } else { 1422 mutex_exit(&proc_lock); 1423 } 1424 1425 exec_path_free(data); 1426 #ifdef TRACE_EXEC 1427 DPRINTF(("%s finished\n", __func__)); 1428 #endif 1429 return EJUSTRETURN; 1430 1431 exec_abort: 1432 SDT_PROBE(proc, kernel, , exec__failure, error, 0, 0, 0, 0); 1433 rw_exit(&p->p_reflock); 1434 if (!no_local_exec_lock) 1435 rw_exit(&exec_lock); 1436 1437 exec_path_free(data); 1438 1439 /* 1440 * the old process doesn't exist anymore. exit gracefully. 1441 * get rid of the (new) address space we have created, if any, get rid 1442 * of our namei data and vnode, and exit noting failure 1443 */ 1444 if (vm != NULL) { 1445 uvm_deallocate(&vm->vm_map, VM_MIN_ADDRESS, 1446 VM_MAXUSER_ADDRESS - VM_MIN_ADDRESS); 1447 } 1448 1449 exec_free_emul_arg(epp); 1450 pool_put(&exec_pool, data->ed_argp); 1451 kmem_free(epp->ep_hdr, epp->ep_hdrlen); 1452 if (epp->ep_emul_root != NULL) 1453 vrele(epp->ep_emul_root); 1454 if (epp->ep_interp != NULL) 1455 vrele(epp->ep_interp); 1456 1457 /* Acquire the sched-state mutex (exit1() will release it). */ 1458 if (!is_spawn) { 1459 mutex_enter(p->p_lock); 1460 exit1(l, error, SIGABRT); 1461 } 1462 1463 return error; 1464 } 1465 1466 int 1467 execve1(struct lwp *l, bool has_path, const char *path, int fd, 1468 char * const *args, char * const *envs, 1469 execve_fetch_element_t fetch_element) 1470 { 1471 struct execve_data data; 1472 int error; 1473 1474 error = execve_loadvm(l, has_path, path, fd, args, envs, fetch_element, 1475 &data); 1476 if (error) 1477 return error; 1478 error = execve_runproc(l, &data, false, false); 1479 return error; 1480 } 1481 1482 static size_t 1483 fromptrsz(const struct exec_package *epp) 1484 { 1485 return (epp->ep_flags & EXEC_FROM32) ? sizeof(int) : sizeof(char *); 1486 } 1487 1488 static size_t 1489 ptrsz(const struct exec_package *epp) 1490 { 1491 return (epp->ep_flags & EXEC_32) ? sizeof(int) : sizeof(char *); 1492 } 1493 1494 static size_t 1495 calcargs(struct execve_data * restrict data, const size_t argenvstrlen) 1496 { 1497 struct exec_package * const epp = &data->ed_pack; 1498 1499 const size_t nargenvptrs = 1500 1 + /* long argc */ 1501 data->ed_argc + /* char *argv[] */ 1502 1 + /* \0 */ 1503 data->ed_envc + /* char *env[] */ 1504 1; /* \0 */ 1505 1506 return (nargenvptrs * ptrsz(epp)) /* pointers */ 1507 + argenvstrlen /* strings */ 1508 + epp->ep_esch->es_arglen; /* auxinfo */ 1509 } 1510 1511 static size_t 1512 calcstack(struct execve_data * restrict data, const size_t gaplen) 1513 { 1514 struct exec_package * const epp = &data->ed_pack; 1515 1516 data->ed_szsigcode = epp->ep_esch->es_emul->e_esigcode - 1517 epp->ep_esch->es_emul->e_sigcode; 1518 1519 data->ed_ps_strings_sz = (epp->ep_flags & EXEC_32) ? 1520 sizeof(struct ps_strings32) : sizeof(struct ps_strings); 1521 1522 const size_t sigcode_psstr_sz = 1523 data->ed_szsigcode + /* sigcode */ 1524 data->ed_ps_strings_sz + /* ps_strings */ 1525 STACK_PTHREADSPACE; /* pthread space */ 1526 1527 const size_t stacklen = 1528 data->ed_argslen + 1529 gaplen + 1530 sigcode_psstr_sz; 1531 1532 /* make the stack "safely" aligned */ 1533 return STACK_LEN_ALIGN(stacklen, STACK_ALIGNBYTES); 1534 } 1535 1536 static int 1537 copyoutargs(struct execve_data * restrict data, struct lwp *l, 1538 char * const newstack) 1539 { 1540 struct exec_package * const epp = &data->ed_pack; 1541 struct proc *p = l->l_proc; 1542 int error; 1543 1544 memset(&data->ed_arginfo, 0, sizeof(data->ed_arginfo)); 1545 1546 /* remember information about the process */ 1547 data->ed_arginfo.ps_nargvstr = data->ed_argc; 1548 data->ed_arginfo.ps_nenvstr = data->ed_envc; 1549 1550 /* 1551 * Allocate the stack address passed to the newly execve()'ed process. 1552 * 1553 * The new stack address will be set to the SP (stack pointer) register 1554 * in setregs(). 1555 */ 1556 1557 char *newargs = STACK_ALLOC( 1558 STACK_SHRINK(newstack, data->ed_argslen), data->ed_argslen); 1559 1560 error = (*epp->ep_esch->es_copyargs)(l, epp, 1561 &data->ed_arginfo, &newargs, data->ed_argp); 1562 1563 if (error) { 1564 DPRINTF(("%s: copyargs failed %d\n", __func__, error)); 1565 return error; 1566 } 1567 1568 error = copyoutpsstrs(data, p); 1569 if (error != 0) 1570 return error; 1571 1572 return 0; 1573 } 1574 1575 static int 1576 copyoutpsstrs(struct execve_data * restrict data, struct proc *p) 1577 { 1578 struct exec_package * const epp = &data->ed_pack; 1579 struct ps_strings32 arginfo32; 1580 void *aip; 1581 int error; 1582 1583 /* fill process ps_strings info */ 1584 p->p_psstrp = (vaddr_t)STACK_ALLOC(STACK_GROW(epp->ep_minsaddr, 1585 STACK_PTHREADSPACE), data->ed_ps_strings_sz); 1586 1587 if (epp->ep_flags & EXEC_32) { 1588 aip = &arginfo32; 1589 arginfo32.ps_argvstr = (vaddr_t)data->ed_arginfo.ps_argvstr; 1590 arginfo32.ps_nargvstr = data->ed_arginfo.ps_nargvstr; 1591 arginfo32.ps_envstr = (vaddr_t)data->ed_arginfo.ps_envstr; 1592 arginfo32.ps_nenvstr = data->ed_arginfo.ps_nenvstr; 1593 } else 1594 aip = &data->ed_arginfo; 1595 1596 /* copy out the process's ps_strings structure */ 1597 if ((error = copyout(aip, (void *)p->p_psstrp, data->ed_ps_strings_sz)) 1598 != 0) { 1599 DPRINTF(("%s: ps_strings copyout %p->%p size %zu failed\n", 1600 __func__, aip, (void *)p->p_psstrp, data->ed_ps_strings_sz)); 1601 return error; 1602 } 1603 1604 return 0; 1605 } 1606 1607 static int 1608 copyinargs(struct execve_data * restrict data, char * const *args, 1609 char * const *envs, execve_fetch_element_t fetch_element, char **dpp) 1610 { 1611 struct exec_package * const epp = &data->ed_pack; 1612 char *dp; 1613 size_t i; 1614 int error; 1615 1616 dp = *dpp; 1617 1618 data->ed_argc = 0; 1619 1620 /* copy the fake args list, if there's one, freeing it as we go */ 1621 if (epp->ep_flags & EXEC_HASARGL) { 1622 struct exec_fakearg *fa = epp->ep_fa; 1623 1624 while (fa->fa_arg != NULL) { 1625 const size_t maxlen = ARG_MAX - (dp - data->ed_argp); 1626 size_t len; 1627 1628 len = strlcpy(dp, fa->fa_arg, maxlen); 1629 /* Count NUL into len. */ 1630 if (len < maxlen) 1631 len++; 1632 else { 1633 while (fa->fa_arg != NULL) { 1634 kmem_free(fa->fa_arg, fa->fa_len); 1635 fa++; 1636 } 1637 kmem_free(epp->ep_fa, epp->ep_fa_len); 1638 epp->ep_flags &= ~EXEC_HASARGL; 1639 return E2BIG; 1640 } 1641 ktrexecarg(fa->fa_arg, len - 1); 1642 dp += len; 1643 1644 kmem_free(fa->fa_arg, fa->fa_len); 1645 fa++; 1646 data->ed_argc++; 1647 } 1648 kmem_free(epp->ep_fa, epp->ep_fa_len); 1649 epp->ep_flags &= ~EXEC_HASARGL; 1650 } 1651 1652 /* 1653 * Read and count argument strings from user. 1654 */ 1655 1656 if (args == NULL) { 1657 DPRINTF(("%s: null args\n", __func__)); 1658 return EINVAL; 1659 } 1660 if (epp->ep_flags & EXEC_SKIPARG) 1661 args = (const void *)((const char *)args + fromptrsz(epp)); 1662 i = 0; 1663 error = copyinargstrs(data, args, fetch_element, &dp, &i, ktr_execarg); 1664 if (error != 0) { 1665 DPRINTF(("%s: copyin arg %d\n", __func__, error)); 1666 return error; 1667 } 1668 data->ed_argc += i; 1669 1670 /* 1671 * Read and count environment strings from user. 1672 */ 1673 1674 data->ed_envc = 0; 1675 /* environment need not be there */ 1676 if (envs == NULL) 1677 goto done; 1678 i = 0; 1679 error = copyinargstrs(data, envs, fetch_element, &dp, &i, ktr_execenv); 1680 if (error != 0) { 1681 DPRINTF(("%s: copyin env %d\n", __func__, error)); 1682 return error; 1683 } 1684 data->ed_envc += i; 1685 1686 done: 1687 *dpp = dp; 1688 1689 return 0; 1690 } 1691 1692 static int 1693 copyinargstrs(struct execve_data * restrict data, char * const *strs, 1694 execve_fetch_element_t fetch_element, char **dpp, size_t *ip, 1695 void (*ktr)(const void *, size_t)) 1696 { 1697 char *dp, *sp; 1698 size_t i; 1699 int error; 1700 1701 dp = *dpp; 1702 1703 i = 0; 1704 while (1) { 1705 const size_t maxlen = ARG_MAX - (dp - data->ed_argp); 1706 size_t len; 1707 1708 if ((error = (*fetch_element)(strs, i, &sp)) != 0) { 1709 return error; 1710 } 1711 if (!sp) 1712 break; 1713 if ((error = copyinstr(sp, dp, maxlen, &len)) != 0) { 1714 if (error == ENAMETOOLONG) 1715 error = E2BIG; 1716 return error; 1717 } 1718 if (__predict_false(ktrace_on)) 1719 (*ktr)(dp, len - 1); 1720 dp += len; 1721 i++; 1722 } 1723 1724 *dpp = dp; 1725 *ip = i; 1726 1727 return 0; 1728 } 1729 1730 /* 1731 * Copy argv and env strings from kernel buffer (argp) to the new stack. 1732 * Those strings are located just after auxinfo. 1733 */ 1734 int 1735 copyargs(struct lwp *l, struct exec_package *pack, struct ps_strings *arginfo, 1736 char **stackp, void *argp) 1737 { 1738 char **cpp, *dp, *sp; 1739 size_t len; 1740 void *nullp; 1741 long argc, envc; 1742 int error; 1743 1744 cpp = (char **)*stackp; 1745 nullp = NULL; 1746 argc = arginfo->ps_nargvstr; 1747 envc = arginfo->ps_nenvstr; 1748 1749 /* argc on stack is long */ 1750 CTASSERT(sizeof(*cpp) == sizeof(argc)); 1751 1752 dp = (char *)(cpp + 1753 1 + /* long argc */ 1754 argc + /* char *argv[] */ 1755 1 + /* \0 */ 1756 envc + /* char *env[] */ 1757 1) + /* \0 */ 1758 pack->ep_esch->es_arglen; /* auxinfo */ 1759 sp = argp; 1760 1761 if ((error = copyout(&argc, cpp++, sizeof(argc))) != 0) { 1762 COPYPRINTF("", cpp - 1, sizeof(argc)); 1763 return error; 1764 } 1765 1766 /* XXX don't copy them out, remap them! */ 1767 arginfo->ps_argvstr = cpp; /* remember location of argv for later */ 1768 1769 for (; --argc >= 0; sp += len, dp += len) { 1770 if ((error = copyout(&dp, cpp++, sizeof(dp))) != 0) { 1771 COPYPRINTF("", cpp - 1, sizeof(dp)); 1772 return error; 1773 } 1774 if ((error = copyoutstr(sp, dp, ARG_MAX, &len)) != 0) { 1775 COPYPRINTF("str", dp, (size_t)ARG_MAX); 1776 return error; 1777 } 1778 } 1779 1780 if ((error = copyout(&nullp, cpp++, sizeof(nullp))) != 0) { 1781 COPYPRINTF("", cpp - 1, sizeof(nullp)); 1782 return error; 1783 } 1784 1785 arginfo->ps_envstr = cpp; /* remember location of envp for later */ 1786 1787 for (; --envc >= 0; sp += len, dp += len) { 1788 if ((error = copyout(&dp, cpp++, sizeof(dp))) != 0) { 1789 COPYPRINTF("", cpp - 1, sizeof(dp)); 1790 return error; 1791 } 1792 if ((error = copyoutstr(sp, dp, ARG_MAX, &len)) != 0) { 1793 COPYPRINTF("str", dp, (size_t)ARG_MAX); 1794 return error; 1795 } 1796 1797 } 1798 1799 if ((error = copyout(&nullp, cpp++, sizeof(nullp))) != 0) { 1800 COPYPRINTF("", cpp - 1, sizeof(nullp)); 1801 return error; 1802 } 1803 1804 *stackp = (char *)cpp; 1805 return 0; 1806 } 1807 1808 1809 /* 1810 * Add execsw[] entries. 1811 */ 1812 int 1813 exec_add(struct execsw *esp, int count) 1814 { 1815 struct exec_entry *it; 1816 int i, error = 0; 1817 1818 if (count == 0) { 1819 return 0; 1820 } 1821 1822 /* Check for duplicates. */ 1823 rw_enter(&exec_lock, RW_WRITER); 1824 for (i = 0; i < count; i++) { 1825 LIST_FOREACH(it, &ex_head, ex_list) { 1826 /* assume unique (makecmds, probe_func, emulation) */ 1827 if (it->ex_sw->es_makecmds == esp[i].es_makecmds && 1828 it->ex_sw->u.elf_probe_func == 1829 esp[i].u.elf_probe_func && 1830 it->ex_sw->es_emul == esp[i].es_emul) { 1831 rw_exit(&exec_lock); 1832 return EEXIST; 1833 } 1834 } 1835 } 1836 1837 /* Allocate new entries. */ 1838 for (i = 0; i < count; i++) { 1839 it = kmem_alloc(sizeof(*it), KM_SLEEP); 1840 it->ex_sw = &esp[i]; 1841 error = exec_sigcode_alloc(it->ex_sw->es_emul); 1842 if (error != 0) { 1843 kmem_free(it, sizeof(*it)); 1844 break; 1845 } 1846 LIST_INSERT_HEAD(&ex_head, it, ex_list); 1847 } 1848 /* If even one fails, remove them all back. */ 1849 if (error != 0) { 1850 for (i--; i >= 0; i--) { 1851 it = LIST_FIRST(&ex_head); 1852 LIST_REMOVE(it, ex_list); 1853 exec_sigcode_free(it->ex_sw->es_emul); 1854 kmem_free(it, sizeof(*it)); 1855 } 1856 return error; 1857 } 1858 1859 /* update execsw[] */ 1860 exec_init(0); 1861 rw_exit(&exec_lock); 1862 return 0; 1863 } 1864 1865 /* 1866 * Remove execsw[] entry. 1867 */ 1868 int 1869 exec_remove(struct execsw *esp, int count) 1870 { 1871 struct exec_entry *it, *next; 1872 int i; 1873 const struct proclist_desc *pd; 1874 proc_t *p; 1875 1876 if (count == 0) { 1877 return 0; 1878 } 1879 1880 /* Abort if any are busy. */ 1881 rw_enter(&exec_lock, RW_WRITER); 1882 for (i = 0; i < count; i++) { 1883 mutex_enter(&proc_lock); 1884 for (pd = proclists; pd->pd_list != NULL; pd++) { 1885 PROCLIST_FOREACH(p, pd->pd_list) { 1886 if (p->p_execsw == &esp[i]) { 1887 mutex_exit(&proc_lock); 1888 rw_exit(&exec_lock); 1889 return EBUSY; 1890 } 1891 } 1892 } 1893 mutex_exit(&proc_lock); 1894 } 1895 1896 /* None are busy, so remove them all. */ 1897 for (i = 0; i < count; i++) { 1898 for (it = LIST_FIRST(&ex_head); it != NULL; it = next) { 1899 next = LIST_NEXT(it, ex_list); 1900 if (it->ex_sw == &esp[i]) { 1901 LIST_REMOVE(it, ex_list); 1902 exec_sigcode_free(it->ex_sw->es_emul); 1903 kmem_free(it, sizeof(*it)); 1904 break; 1905 } 1906 } 1907 } 1908 1909 /* update execsw[] */ 1910 exec_init(0); 1911 rw_exit(&exec_lock); 1912 return 0; 1913 } 1914 1915 /* 1916 * Initialize exec structures. If init_boot is true, also does necessary 1917 * one-time initialization (it's called from main() that way). 1918 * Once system is multiuser, this should be called with exec_lock held, 1919 * i.e. via exec_{add|remove}(). 1920 */ 1921 int 1922 exec_init(int init_boot) 1923 { 1924 const struct execsw **sw; 1925 struct exec_entry *ex; 1926 SLIST_HEAD(,exec_entry) first; 1927 SLIST_HEAD(,exec_entry) any; 1928 SLIST_HEAD(,exec_entry) last; 1929 int i, sz; 1930 1931 if (init_boot) { 1932 /* do one-time initializations */ 1933 vaddr_t vmin = 0, vmax; 1934 1935 rw_init(&exec_lock); 1936 exec_map = uvm_km_suballoc(kernel_map, &vmin, &vmax, 1937 maxexec*NCARGS, VM_MAP_PAGEABLE, false, NULL); 1938 pool_init(&exec_pool, NCARGS, 0, 0, PR_NOALIGN|PR_NOTOUCH, 1939 "execargs", &exec_palloc, IPL_NONE); 1940 pool_sethardlimit(&exec_pool, maxexec, "should not happen", 0); 1941 } else { 1942 KASSERT(rw_write_held(&exec_lock)); 1943 } 1944 1945 /* Sort each entry onto the appropriate queue. */ 1946 SLIST_INIT(&first); 1947 SLIST_INIT(&any); 1948 SLIST_INIT(&last); 1949 sz = 0; 1950 LIST_FOREACH(ex, &ex_head, ex_list) { 1951 switch(ex->ex_sw->es_prio) { 1952 case EXECSW_PRIO_FIRST: 1953 SLIST_INSERT_HEAD(&first, ex, ex_slist); 1954 break; 1955 case EXECSW_PRIO_ANY: 1956 SLIST_INSERT_HEAD(&any, ex, ex_slist); 1957 break; 1958 case EXECSW_PRIO_LAST: 1959 SLIST_INSERT_HEAD(&last, ex, ex_slist); 1960 break; 1961 default: 1962 panic("%s", __func__); 1963 break; 1964 } 1965 sz++; 1966 } 1967 1968 /* 1969 * Create new execsw[]. Ensure we do not try a zero-sized 1970 * allocation. 1971 */ 1972 sw = kmem_alloc(sz * sizeof(struct execsw *) + 1, KM_SLEEP); 1973 i = 0; 1974 SLIST_FOREACH(ex, &first, ex_slist) { 1975 sw[i++] = ex->ex_sw; 1976 } 1977 SLIST_FOREACH(ex, &any, ex_slist) { 1978 sw[i++] = ex->ex_sw; 1979 } 1980 SLIST_FOREACH(ex, &last, ex_slist) { 1981 sw[i++] = ex->ex_sw; 1982 } 1983 1984 /* Replace old execsw[] and free used memory. */ 1985 if (execsw != NULL) { 1986 kmem_free(__UNCONST(execsw), 1987 nexecs * sizeof(struct execsw *) + 1); 1988 } 1989 execsw = sw; 1990 nexecs = sz; 1991 1992 /* Figure out the maximum size of an exec header. */ 1993 exec_maxhdrsz = sizeof(int); 1994 for (i = 0; i < nexecs; i++) { 1995 if (execsw[i]->es_hdrsz > exec_maxhdrsz) 1996 exec_maxhdrsz = execsw[i]->es_hdrsz; 1997 } 1998 1999 return 0; 2000 } 2001 2002 int 2003 exec_sigcode_alloc(const struct emul *e) 2004 { 2005 vaddr_t va; 2006 vsize_t sz; 2007 int error; 2008 struct uvm_object *uobj; 2009 2010 KASSERT(rw_lock_held(&exec_lock)); 2011 2012 if (e == NULL || e->e_sigobject == NULL) 2013 return 0; 2014 2015 sz = (vaddr_t)e->e_esigcode - (vaddr_t)e->e_sigcode; 2016 if (sz == 0) 2017 return 0; 2018 2019 /* 2020 * Create a sigobject for this emulation. 2021 * 2022 * sigobject is an anonymous memory object (just like SYSV shared 2023 * memory) that we keep a permanent reference to and that we map 2024 * in all processes that need this sigcode. The creation is simple, 2025 * we create an object, add a permanent reference to it, map it in 2026 * kernel space, copy out the sigcode to it and unmap it. 2027 * We map it with PROT_READ|PROT_EXEC into the process just 2028 * the way sys_mmap() would map it. 2029 */ 2030 if (*e->e_sigobject == NULL) { 2031 uobj = uao_create(sz, 0); 2032 (*uobj->pgops->pgo_reference)(uobj); 2033 va = vm_map_min(kernel_map); 2034 if ((error = uvm_map(kernel_map, &va, round_page(sz), 2035 uobj, 0, 0, 2036 UVM_MAPFLAG(UVM_PROT_RW, UVM_PROT_RW, 2037 UVM_INH_SHARE, UVM_ADV_RANDOM, 0)))) { 2038 printf("sigcode kernel mapping failed %d\n", error); 2039 (*uobj->pgops->pgo_detach)(uobj); 2040 return error; 2041 } 2042 memcpy((void *)va, e->e_sigcode, sz); 2043 #ifdef PMAP_NEED_PROCWR 2044 pmap_procwr(&proc0, va, sz); 2045 #endif 2046 uvm_unmap(kernel_map, va, va + round_page(sz)); 2047 *e->e_sigobject = uobj; 2048 KASSERT(uobj->uo_refs == 1); 2049 } else { 2050 /* if already created, reference++ */ 2051 uobj = *e->e_sigobject; 2052 (*uobj->pgops->pgo_reference)(uobj); 2053 } 2054 2055 return 0; 2056 } 2057 2058 void 2059 exec_sigcode_free(const struct emul *e) 2060 { 2061 struct uvm_object *uobj; 2062 2063 KASSERT(rw_lock_held(&exec_lock)); 2064 2065 if (e == NULL || e->e_sigobject == NULL) 2066 return; 2067 2068 uobj = *e->e_sigobject; 2069 if (uobj == NULL) 2070 return; 2071 2072 if (uobj->uo_refs == 1) 2073 *e->e_sigobject = NULL; /* I'm the last person to reference. */ 2074 (*uobj->pgops->pgo_detach)(uobj); 2075 } 2076 2077 static int 2078 exec_sigcode_map(struct proc *p, const struct emul *e) 2079 { 2080 vaddr_t va; 2081 vsize_t sz; 2082 int error; 2083 struct uvm_object *uobj; 2084 2085 sz = (vaddr_t)e->e_esigcode - (vaddr_t)e->e_sigcode; 2086 if (e->e_sigobject == NULL || sz == 0) 2087 return 0; 2088 2089 uobj = *e->e_sigobject; 2090 if (uobj == NULL) 2091 return 0; 2092 2093 /* Just a hint to uvm_map where to put it. */ 2094 va = e->e_vm_default_addr(p, (vaddr_t)p->p_vmspace->vm_daddr, 2095 round_page(sz), p->p_vmspace->vm_map.flags & VM_MAP_TOPDOWN); 2096 2097 #ifdef __alpha__ 2098 /* 2099 * Tru64 puts /sbin/loader at the end of user virtual memory, 2100 * which causes the above calculation to put the sigcode at 2101 * an invalid address. Put it just below the text instead. 2102 */ 2103 if (va == (vaddr_t)vm_map_max(&p->p_vmspace->vm_map)) { 2104 va = (vaddr_t)p->p_vmspace->vm_taddr - round_page(sz); 2105 } 2106 #endif 2107 2108 (*uobj->pgops->pgo_reference)(uobj); 2109 error = uvm_map(&p->p_vmspace->vm_map, &va, round_page(sz), 2110 uobj, 0, 0, 2111 UVM_MAPFLAG(UVM_PROT_RX, UVM_PROT_RX, UVM_INH_SHARE, 2112 UVM_ADV_RANDOM, 0)); 2113 if (error) { 2114 DPRINTF(("%s, %d: map %p " 2115 "uvm_map %#"PRIxVSIZE"@%#"PRIxVADDR" failed %d\n", 2116 __func__, __LINE__, &p->p_vmspace->vm_map, round_page(sz), 2117 va, error)); 2118 (*uobj->pgops->pgo_detach)(uobj); 2119 return error; 2120 } 2121 p->p_sigctx.ps_sigcode = (void *)va; 2122 return 0; 2123 } 2124 2125 /* 2126 * Release a refcount on spawn_exec_data and destroy memory, if this 2127 * was the last one. 2128 */ 2129 static void 2130 spawn_exec_data_release(struct spawn_exec_data *data) 2131 { 2132 if (atomic_dec_32_nv(&data->sed_refcnt) != 0) 2133 return; 2134 2135 cv_destroy(&data->sed_cv_child_ready); 2136 mutex_destroy(&data->sed_mtx_child); 2137 2138 if (data->sed_actions) 2139 posix_spawn_fa_free(data->sed_actions, 2140 data->sed_actions->len); 2141 if (data->sed_attrs) 2142 kmem_free(data->sed_attrs, 2143 sizeof(*data->sed_attrs)); 2144 kmem_free(data, sizeof(*data)); 2145 } 2146 2147 static int 2148 handle_posix_spawn_file_actions(struct posix_spawn_file_actions *actions) 2149 { 2150 struct lwp *l = curlwp; 2151 register_t retval; 2152 int error, newfd; 2153 2154 if (actions == NULL) 2155 return 0; 2156 2157 for (size_t i = 0; i < actions->len; i++) { 2158 const struct posix_spawn_file_actions_entry *fae = 2159 &actions->fae[i]; 2160 switch (fae->fae_action) { 2161 case FAE_OPEN: 2162 if (fd_getfile(fae->fae_fildes) != NULL) { 2163 error = fd_close(fae->fae_fildes); 2164 if (error) 2165 return error; 2166 } 2167 error = fd_open(fae->fae_path, fae->fae_oflag, 2168 fae->fae_mode, &newfd); 2169 if (error) 2170 return error; 2171 if (newfd != fae->fae_fildes) { 2172 error = dodup(l, newfd, 2173 fae->fae_fildes, 0, &retval); 2174 if (fd_getfile(newfd) != NULL) 2175 fd_close(newfd); 2176 } 2177 break; 2178 case FAE_DUP2: 2179 error = dodup(l, fae->fae_fildes, 2180 fae->fae_newfildes, 0, &retval); 2181 break; 2182 case FAE_CLOSE: 2183 if (fd_getfile(fae->fae_fildes) == NULL) { 2184 return EBADF; 2185 } 2186 error = fd_close(fae->fae_fildes); 2187 break; 2188 case FAE_CHDIR: 2189 error = do_sys_chdir(l, fae->fae_chdir_path, 2190 UIO_SYSSPACE, &retval); 2191 break; 2192 case FAE_FCHDIR: 2193 error = do_sys_fchdir(l, fae->fae_fildes, &retval); 2194 break; 2195 } 2196 if (error) 2197 return error; 2198 } 2199 return 0; 2200 } 2201 2202 static int 2203 handle_posix_spawn_attrs(struct posix_spawnattr *attrs, struct proc *parent) 2204 { 2205 struct sigaction sigact; 2206 int error; 2207 struct proc *p = curproc; 2208 struct lwp *l = curlwp; 2209 2210 if (attrs == NULL) 2211 return 0; 2212 2213 memset(&sigact, 0, sizeof(sigact)); 2214 sigact._sa_u._sa_handler = SIG_DFL; 2215 sigact.sa_flags = 0; 2216 2217 /* 2218 * set state to SSTOP so that this proc can be found by pid. 2219 * see proc_enterprp, do_sched_setparam below 2220 */ 2221 mutex_enter(&proc_lock); 2222 /* 2223 * p_stat should be SACTIVE, so we need to adjust the 2224 * parent's p_nstopchild here. For safety, just make 2225 * we're on the good side of SDEAD before we adjust. 2226 */ 2227 int ostat = p->p_stat; 2228 KASSERT(ostat < SSTOP); 2229 p->p_stat = SSTOP; 2230 p->p_waited = 0; 2231 p->p_pptr->p_nstopchild++; 2232 mutex_exit(&proc_lock); 2233 2234 /* Set process group */ 2235 if (attrs->sa_flags & POSIX_SPAWN_SETPGROUP) { 2236 pid_t mypid = p->p_pid; 2237 pid_t pgrp = attrs->sa_pgroup; 2238 2239 if (pgrp == 0) 2240 pgrp = mypid; 2241 2242 error = proc_enterpgrp(parent, mypid, pgrp, false); 2243 if (error) 2244 goto out; 2245 } 2246 2247 /* Set scheduler policy */ 2248 if (attrs->sa_flags & POSIX_SPAWN_SETSCHEDULER) 2249 error = do_sched_setparam(p->p_pid, 0, attrs->sa_schedpolicy, 2250 &attrs->sa_schedparam); 2251 else if (attrs->sa_flags & POSIX_SPAWN_SETSCHEDPARAM) { 2252 error = do_sched_setparam(parent->p_pid, 0, 2253 SCHED_NONE, &attrs->sa_schedparam); 2254 } 2255 if (error) 2256 goto out; 2257 2258 /* Reset user ID's */ 2259 if (attrs->sa_flags & POSIX_SPAWN_RESETIDS) { 2260 error = do_setresgid(l, -1, kauth_cred_getgid(l->l_cred), -1, 2261 ID_E_EQ_R | ID_E_EQ_S); 2262 if (error) 2263 return error; 2264 error = do_setresuid(l, -1, kauth_cred_getuid(l->l_cred), -1, 2265 ID_E_EQ_R | ID_E_EQ_S); 2266 if (error) 2267 goto out; 2268 } 2269 2270 /* Set signal masks/defaults */ 2271 if (attrs->sa_flags & POSIX_SPAWN_SETSIGMASK) { 2272 mutex_enter(p->p_lock); 2273 error = sigprocmask1(l, SIG_SETMASK, &attrs->sa_sigmask, NULL); 2274 mutex_exit(p->p_lock); 2275 if (error) 2276 goto out; 2277 } 2278 2279 if (attrs->sa_flags & POSIX_SPAWN_SETSIGDEF) { 2280 /* 2281 * The following sigaction call is using a sigaction 2282 * version 0 trampoline which is in the compatibility 2283 * code only. This is not a problem because for SIG_DFL 2284 * and SIG_IGN, the trampolines are now ignored. If they 2285 * were not, this would be a problem because we are 2286 * holding the exec_lock, and the compat code needs 2287 * to do the same in order to replace the trampoline 2288 * code of the process. 2289 */ 2290 for (int i = 1; i <= NSIG; i++) { 2291 if (sigismember(&attrs->sa_sigdefault, i)) 2292 sigaction1(l, i, &sigact, NULL, NULL, 0); 2293 } 2294 } 2295 error = 0; 2296 out: 2297 mutex_enter(&proc_lock); 2298 p->p_stat = ostat; 2299 p->p_pptr->p_nstopchild--; 2300 mutex_exit(&proc_lock); 2301 return error; 2302 } 2303 2304 /* 2305 * A child lwp of a posix_spawn operation starts here and ends up in 2306 * cpu_spawn_return, dealing with all filedescriptor and scheduler 2307 * manipulations in between. 2308 * The parent waits for the child, as it is not clear whether the child 2309 * will be able to acquire its own exec_lock. If it can, the parent can 2310 * be released early and continue running in parallel. If not (or if the 2311 * magic debug flag is passed in the scheduler attribute struct), the 2312 * child rides on the parent's exec lock until it is ready to return to 2313 * to userland - and only then releases the parent. This method loses 2314 * concurrency, but improves error reporting. 2315 */ 2316 static void 2317 spawn_return(void *arg) 2318 { 2319 struct spawn_exec_data *spawn_data = arg; 2320 struct lwp *l = curlwp; 2321 struct proc *p = l->l_proc; 2322 int error; 2323 bool have_reflock; 2324 bool parent_is_waiting = true; 2325 2326 /* 2327 * Check if we can release parent early. 2328 * We either need to have no sed_attrs, or sed_attrs does not 2329 * have POSIX_SPAWN_RETURNERROR or one of the flags, that require 2330 * safe access to the parent proc (passed in sed_parent). 2331 * We then try to get the exec_lock, and only if that works, we can 2332 * release the parent here already. 2333 */ 2334 struct posix_spawnattr *attrs = spawn_data->sed_attrs; 2335 if ((!attrs || (attrs->sa_flags 2336 & (POSIX_SPAWN_RETURNERROR|POSIX_SPAWN_SETPGROUP)) == 0) 2337 && rw_tryenter(&exec_lock, RW_READER)) { 2338 parent_is_waiting = false; 2339 mutex_enter(&spawn_data->sed_mtx_child); 2340 cv_signal(&spawn_data->sed_cv_child_ready); 2341 mutex_exit(&spawn_data->sed_mtx_child); 2342 } 2343 2344 /* don't allow debugger access yet */ 2345 rw_enter(&p->p_reflock, RW_WRITER); 2346 have_reflock = true; 2347 2348 /* handle posix_spawnattr */ 2349 error = handle_posix_spawn_attrs(attrs, spawn_data->sed_parent); 2350 if (error) 2351 goto report_error; 2352 2353 /* handle posix_spawn_file_actions */ 2354 error = handle_posix_spawn_file_actions(spawn_data->sed_actions); 2355 if (error) 2356 goto report_error; 2357 2358 /* now do the real exec */ 2359 error = execve_runproc(l, &spawn_data->sed_exec, parent_is_waiting, 2360 true); 2361 have_reflock = false; 2362 if (error == EJUSTRETURN) 2363 error = 0; 2364 else if (error) 2365 goto report_error; 2366 2367 if (parent_is_waiting) { 2368 mutex_enter(&spawn_data->sed_mtx_child); 2369 cv_signal(&spawn_data->sed_cv_child_ready); 2370 mutex_exit(&spawn_data->sed_mtx_child); 2371 } 2372 2373 /* release our refcount on the data */ 2374 spawn_exec_data_release(spawn_data); 2375 2376 if ((p->p_slflag & (PSL_TRACED|PSL_TRACEDCHILD)) == 2377 (PSL_TRACED|PSL_TRACEDCHILD)) { 2378 eventswitchchild(p, TRAP_CHLD, PTRACE_POSIX_SPAWN); 2379 } 2380 2381 /* and finally: leave to userland for the first time */ 2382 cpu_spawn_return(l); 2383 2384 /* NOTREACHED */ 2385 return; 2386 2387 report_error: 2388 if (have_reflock) { 2389 /* 2390 * We have not passed through execve_runproc(), 2391 * which would have released the p_reflock and also 2392 * taken ownership of the sed_exec part of spawn_data, 2393 * so release/free both here. 2394 */ 2395 rw_exit(&p->p_reflock); 2396 execve_free_data(&spawn_data->sed_exec); 2397 } 2398 2399 if (parent_is_waiting) { 2400 /* pass error to parent */ 2401 mutex_enter(&spawn_data->sed_mtx_child); 2402 spawn_data->sed_error = error; 2403 cv_signal(&spawn_data->sed_cv_child_ready); 2404 mutex_exit(&spawn_data->sed_mtx_child); 2405 } else { 2406 rw_exit(&exec_lock); 2407 } 2408 2409 /* release our refcount on the data */ 2410 spawn_exec_data_release(spawn_data); 2411 2412 /* done, exit */ 2413 mutex_enter(p->p_lock); 2414 /* 2415 * Posix explicitly asks for an exit code of 127 if we report 2416 * errors from the child process - so, unfortunately, there 2417 * is no way to report a more exact error code. 2418 * A NetBSD specific workaround is POSIX_SPAWN_RETURNERROR as 2419 * flag bit in the attrp argument to posix_spawn(2), see above. 2420 */ 2421 exit1(l, 127, 0); 2422 } 2423 2424 static __inline char ** 2425 posix_spawn_fae_path(struct posix_spawn_file_actions_entry *fae) 2426 { 2427 switch (fae->fae_action) { 2428 case FAE_OPEN: 2429 return &fae->fae_path; 2430 case FAE_CHDIR: 2431 return &fae->fae_chdir_path; 2432 default: 2433 return NULL; 2434 } 2435 } 2436 2437 void 2438 posix_spawn_fa_free(struct posix_spawn_file_actions *fa, size_t len) 2439 { 2440 2441 for (size_t i = 0; i < len; i++) { 2442 char **pathp = posix_spawn_fae_path(&fa->fae[i]); 2443 if (pathp) 2444 kmem_strfree(*pathp); 2445 } 2446 if (fa->len > 0) 2447 kmem_free(fa->fae, sizeof(*fa->fae) * fa->len); 2448 kmem_free(fa, sizeof(*fa)); 2449 } 2450 2451 static int 2452 posix_spawn_fa_alloc(struct posix_spawn_file_actions **fap, 2453 const struct posix_spawn_file_actions *ufa, rlim_t lim) 2454 { 2455 struct posix_spawn_file_actions *fa; 2456 struct posix_spawn_file_actions_entry *fae; 2457 char *pbuf = NULL; 2458 int error; 2459 size_t i = 0; 2460 2461 fa = kmem_alloc(sizeof(*fa), KM_SLEEP); 2462 error = copyin(ufa, fa, sizeof(*fa)); 2463 if (error || fa->len == 0) { 2464 kmem_free(fa, sizeof(*fa)); 2465 return error; /* 0 if not an error, and len == 0 */ 2466 } 2467 2468 if (fa->len > lim) { 2469 kmem_free(fa, sizeof(*fa)); 2470 return EINVAL; 2471 } 2472 2473 fa->size = fa->len; 2474 size_t fal = fa->len * sizeof(*fae); 2475 fae = fa->fae; 2476 fa->fae = kmem_alloc(fal, KM_SLEEP); 2477 error = copyin(fae, fa->fae, fal); 2478 if (error) 2479 goto out; 2480 2481 pbuf = PNBUF_GET(); 2482 for (; i < fa->len; i++) { 2483 char **pathp = posix_spawn_fae_path(&fa->fae[i]); 2484 if (pathp == NULL) 2485 continue; 2486 error = copyinstr(*pathp, pbuf, MAXPATHLEN, &fal); 2487 if (error) 2488 goto out; 2489 *pathp = kmem_alloc(fal, KM_SLEEP); 2490 memcpy(*pathp, pbuf, fal); 2491 } 2492 PNBUF_PUT(pbuf); 2493 2494 *fap = fa; 2495 return 0; 2496 out: 2497 if (pbuf) 2498 PNBUF_PUT(pbuf); 2499 posix_spawn_fa_free(fa, i); 2500 return error; 2501 } 2502 2503 /* 2504 * N.B. increments nprocs upon success. Callers need to drop nprocs if 2505 * they fail for some other reason. 2506 */ 2507 int 2508 check_posix_spawn(struct lwp *l1) 2509 { 2510 int error, tnprocs, count; 2511 uid_t uid; 2512 struct proc *p1; 2513 2514 p1 = l1->l_proc; 2515 uid = kauth_cred_getuid(l1->l_cred); 2516 tnprocs = atomic_inc_uint_nv(&nprocs); 2517 2518 /* 2519 * Although process entries are dynamically created, we still keep 2520 * a global limit on the maximum number we will create. 2521 */ 2522 if (__predict_false(tnprocs >= maxproc)) 2523 error = -1; 2524 else 2525 error = kauth_authorize_process(l1->l_cred, 2526 KAUTH_PROCESS_FORK, p1, KAUTH_ARG(tnprocs), NULL, NULL); 2527 2528 if (error) { 2529 atomic_dec_uint(&nprocs); 2530 return EAGAIN; 2531 } 2532 2533 /* 2534 * Enforce limits. 2535 */ 2536 count = chgproccnt(uid, 1); 2537 if (kauth_authorize_process(l1->l_cred, KAUTH_PROCESS_RLIMIT, 2538 p1, KAUTH_ARG(KAUTH_REQ_PROCESS_RLIMIT_BYPASS), 2539 &p1->p_rlimit[RLIMIT_NPROC], KAUTH_ARG(RLIMIT_NPROC)) != 0 && 2540 __predict_false(count > p1->p_rlimit[RLIMIT_NPROC].rlim_cur)) { 2541 (void)chgproccnt(uid, -1); 2542 atomic_dec_uint(&nprocs); 2543 return EAGAIN; 2544 } 2545 2546 return 0; 2547 } 2548 2549 int 2550 do_posix_spawn(struct lwp *l1, pid_t *pid_res, bool *child_ok, const char *path, 2551 struct posix_spawn_file_actions *fa, 2552 struct posix_spawnattr *sa, 2553 char *const *argv, char *const *envp, 2554 execve_fetch_element_t fetch) 2555 { 2556 2557 struct proc *p1, *p2; 2558 struct lwp *l2; 2559 int error; 2560 struct spawn_exec_data *spawn_data; 2561 vaddr_t uaddr; 2562 pid_t pid; 2563 bool have_exec_lock = false; 2564 2565 p1 = l1->l_proc; 2566 2567 /* Allocate and init spawn_data */ 2568 spawn_data = kmem_zalloc(sizeof(*spawn_data), KM_SLEEP); 2569 spawn_data->sed_refcnt = 1; /* only parent so far */ 2570 cv_init(&spawn_data->sed_cv_child_ready, "pspawn"); 2571 mutex_init(&spawn_data->sed_mtx_child, MUTEX_DEFAULT, IPL_NONE); 2572 mutex_enter(&spawn_data->sed_mtx_child); 2573 2574 /* 2575 * Do the first part of the exec now, collect state 2576 * in spawn_data. 2577 */ 2578 error = execve_loadvm(l1, true, path, -1, argv, 2579 envp, fetch, &spawn_data->sed_exec); 2580 if (error == EJUSTRETURN) 2581 error = 0; 2582 else if (error) 2583 goto error_exit; 2584 2585 have_exec_lock = true; 2586 2587 /* 2588 * Allocate virtual address space for the U-area now, while it 2589 * is still easy to abort the fork operation if we're out of 2590 * kernel virtual address space. 2591 */ 2592 uaddr = uvm_uarea_alloc(); 2593 if (__predict_false(uaddr == 0)) { 2594 error = ENOMEM; 2595 goto error_exit; 2596 } 2597 2598 /* 2599 * Allocate new proc. Borrow proc0 vmspace for it, we will 2600 * replace it with its own before returning to userland 2601 * in the child. 2602 */ 2603 p2 = proc_alloc(); 2604 if (p2 == NULL) { 2605 /* We were unable to allocate a process ID. */ 2606 error = EAGAIN; 2607 goto error_exit; 2608 } 2609 2610 /* 2611 * This is a point of no return, we will have to go through 2612 * the child proc to properly clean it up past this point. 2613 */ 2614 pid = p2->p_pid; 2615 2616 /* 2617 * Make a proc table entry for the new process. 2618 * Start by zeroing the section of proc that is zero-initialized, 2619 * then copy the section that is copied directly from the parent. 2620 */ 2621 memset(&p2->p_startzero, 0, 2622 (unsigned) ((char *)&p2->p_endzero - (char *)&p2->p_startzero)); 2623 memcpy(&p2->p_startcopy, &p1->p_startcopy, 2624 (unsigned) ((char *)&p2->p_endcopy - (char *)&p2->p_startcopy)); 2625 p2->p_vmspace = proc0.p_vmspace; 2626 2627 TAILQ_INIT(&p2->p_sigpend.sp_info); 2628 2629 LIST_INIT(&p2->p_lwps); 2630 LIST_INIT(&p2->p_sigwaiters); 2631 2632 /* 2633 * Duplicate sub-structures as needed. 2634 * Increase reference counts on shared objects. 2635 * Inherit flags we want to keep. The flags related to SIGCHLD 2636 * handling are important in order to keep a consistent behaviour 2637 * for the child after the fork. If we are a 32-bit process, the 2638 * child will be too. 2639 */ 2640 p2->p_flag = 2641 p1->p_flag & (PK_SUGID | PK_NOCLDWAIT | PK_CLDSIGIGN | PK_32); 2642 p2->p_emul = p1->p_emul; 2643 p2->p_execsw = p1->p_execsw; 2644 2645 mutex_init(&p2->p_stmutex, MUTEX_DEFAULT, IPL_HIGH); 2646 mutex_init(&p2->p_auxlock, MUTEX_DEFAULT, IPL_NONE); 2647 rw_init(&p2->p_reflock); 2648 cv_init(&p2->p_waitcv, "wait"); 2649 cv_init(&p2->p_lwpcv, "lwpwait"); 2650 2651 p2->p_lock = mutex_obj_alloc(MUTEX_DEFAULT, IPL_NONE); 2652 2653 kauth_proc_fork(p1, p2); 2654 2655 p2->p_raslist = NULL; 2656 p2->p_fd = fd_copy(); 2657 2658 /* XXX racy */ 2659 p2->p_mqueue_cnt = p1->p_mqueue_cnt; 2660 2661 p2->p_cwdi = cwdinit(); 2662 2663 /* 2664 * Note: p_limit (rlimit stuff) is copy-on-write, so normally 2665 * we just need increase pl_refcnt. 2666 */ 2667 if (!p1->p_limit->pl_writeable) { 2668 lim_addref(p1->p_limit); 2669 p2->p_limit = p1->p_limit; 2670 } else { 2671 p2->p_limit = lim_copy(p1->p_limit); 2672 } 2673 2674 p2->p_lflag = 0; 2675 l1->l_vforkwaiting = false; 2676 p2->p_sflag = 0; 2677 p2->p_slflag = 0; 2678 p2->p_pptr = p1; 2679 p2->p_ppid = p1->p_pid; 2680 LIST_INIT(&p2->p_children); 2681 2682 p2->p_aio = NULL; 2683 2684 #ifdef KTRACE 2685 /* 2686 * Copy traceflag and tracefile if enabled. 2687 * If not inherited, these were zeroed above. 2688 */ 2689 if (p1->p_traceflag & KTRFAC_INHERIT) { 2690 mutex_enter(&ktrace_lock); 2691 p2->p_traceflag = p1->p_traceflag; 2692 if ((p2->p_tracep = p1->p_tracep) != NULL) 2693 ktradref(p2); 2694 mutex_exit(&ktrace_lock); 2695 } 2696 #endif 2697 2698 /* 2699 * Create signal actions for the child process. 2700 */ 2701 p2->p_sigacts = sigactsinit(p1, 0); 2702 mutex_enter(p1->p_lock); 2703 p2->p_sflag |= 2704 (p1->p_sflag & (PS_STOPFORK | PS_STOPEXEC | PS_NOCLDSTOP)); 2705 sched_proc_fork(p1, p2); 2706 mutex_exit(p1->p_lock); 2707 2708 p2->p_stflag = p1->p_stflag; 2709 2710 /* 2711 * p_stats. 2712 * Copy parts of p_stats, and zero out the rest. 2713 */ 2714 p2->p_stats = pstatscopy(p1->p_stats); 2715 2716 /* copy over machdep flags to the new proc */ 2717 cpu_proc_fork(p1, p2); 2718 2719 /* 2720 * Prepare remaining parts of spawn data 2721 */ 2722 spawn_data->sed_actions = fa; 2723 spawn_data->sed_attrs = sa; 2724 2725 spawn_data->sed_parent = p1; 2726 2727 /* create LWP */ 2728 lwp_create(l1, p2, uaddr, 0, NULL, 0, spawn_return, spawn_data, 2729 &l2, l1->l_class, &l1->l_sigmask, &l1->l_sigstk); 2730 l2->l_ctxlink = NULL; /* reset ucontext link */ 2731 2732 /* 2733 * Copy the credential so other references don't see our changes. 2734 * Test to see if this is necessary first, since in the common case 2735 * we won't need a private reference. 2736 */ 2737 if (kauth_cred_geteuid(l2->l_cred) != kauth_cred_getsvuid(l2->l_cred) || 2738 kauth_cred_getegid(l2->l_cred) != kauth_cred_getsvgid(l2->l_cred)) { 2739 l2->l_cred = kauth_cred_copy(l2->l_cred); 2740 kauth_cred_setsvuid(l2->l_cred, kauth_cred_geteuid(l2->l_cred)); 2741 kauth_cred_setsvgid(l2->l_cred, kauth_cred_getegid(l2->l_cred)); 2742 } 2743 2744 /* Update the master credentials. */ 2745 if (l2->l_cred != p2->p_cred) { 2746 kauth_cred_t ocred; 2747 2748 kauth_cred_hold(l2->l_cred); 2749 mutex_enter(p2->p_lock); 2750 ocred = p2->p_cred; 2751 p2->p_cred = l2->l_cred; 2752 mutex_exit(p2->p_lock); 2753 kauth_cred_free(ocred); 2754 } 2755 2756 *child_ok = true; 2757 spawn_data->sed_refcnt = 2; /* child gets it as well */ 2758 #if 0 2759 l2->l_nopreempt = 1; /* start it non-preemptable */ 2760 #endif 2761 2762 /* 2763 * It's now safe for the scheduler and other processes to see the 2764 * child process. 2765 */ 2766 mutex_enter(&proc_lock); 2767 2768 if (p1->p_session->s_ttyvp != NULL && p1->p_lflag & PL_CONTROLT) 2769 p2->p_lflag |= PL_CONTROLT; 2770 2771 LIST_INSERT_HEAD(&p1->p_children, p2, p_sibling); 2772 p2->p_exitsig = SIGCHLD; /* signal for parent on exit */ 2773 2774 if ((p1->p_slflag & (PSL_TRACEPOSIX_SPAWN|PSL_TRACED)) == 2775 (PSL_TRACEPOSIX_SPAWN|PSL_TRACED)) { 2776 proc_changeparent(p2, p1->p_pptr); 2777 SET(p2->p_slflag, PSL_TRACEDCHILD); 2778 } 2779 2780 p2->p_oppid = p1->p_pid; /* Remember the original parent id. */ 2781 2782 LIST_INSERT_AFTER(p1, p2, p_pglist); 2783 LIST_INSERT_HEAD(&allproc, p2, p_list); 2784 2785 p2->p_trace_enabled = trace_is_enabled(p2); 2786 #ifdef __HAVE_SYSCALL_INTERN 2787 (*p2->p_emul->e_syscall_intern)(p2); 2788 #endif 2789 2790 /* 2791 * Make child runnable, set start time, and add to run queue except 2792 * if the parent requested the child to start in SSTOP state. 2793 */ 2794 mutex_enter(p2->p_lock); 2795 2796 getmicrotime(&p2->p_stats->p_start); 2797 2798 lwp_lock(l2); 2799 KASSERT(p2->p_nrlwps == 1); 2800 KASSERT(l2->l_stat == LSIDL); 2801 p2->p_nrlwps = 1; 2802 p2->p_stat = SACTIVE; 2803 setrunnable(l2); 2804 /* LWP now unlocked */ 2805 2806 mutex_exit(p2->p_lock); 2807 mutex_exit(&proc_lock); 2808 2809 cv_wait(&spawn_data->sed_cv_child_ready, &spawn_data->sed_mtx_child); 2810 error = spawn_data->sed_error; 2811 mutex_exit(&spawn_data->sed_mtx_child); 2812 spawn_exec_data_release(spawn_data); 2813 2814 rw_exit(&p1->p_reflock); 2815 rw_exit(&exec_lock); 2816 have_exec_lock = false; 2817 2818 *pid_res = pid; 2819 2820 if (error) 2821 return error; 2822 2823 if (p1->p_slflag & PSL_TRACED) { 2824 /* Paranoid check */ 2825 mutex_enter(&proc_lock); 2826 if ((p1->p_slflag & (PSL_TRACEPOSIX_SPAWN|PSL_TRACED)) != 2827 (PSL_TRACEPOSIX_SPAWN|PSL_TRACED)) { 2828 mutex_exit(&proc_lock); 2829 return 0; 2830 } 2831 2832 mutex_enter(p1->p_lock); 2833 eventswitch(TRAP_CHLD, PTRACE_POSIX_SPAWN, pid); 2834 } 2835 return 0; 2836 2837 error_exit: 2838 if (have_exec_lock) { 2839 execve_free_data(&spawn_data->sed_exec); 2840 rw_exit(&p1->p_reflock); 2841 rw_exit(&exec_lock); 2842 } 2843 mutex_exit(&spawn_data->sed_mtx_child); 2844 spawn_exec_data_release(spawn_data); 2845 2846 return error; 2847 } 2848 2849 int 2850 sys_posix_spawn(struct lwp *l1, const struct sys_posix_spawn_args *uap, 2851 register_t *retval) 2852 { 2853 /* { 2854 syscallarg(pid_t *) pid; 2855 syscallarg(const char *) path; 2856 syscallarg(const struct posix_spawn_file_actions *) file_actions; 2857 syscallarg(const struct posix_spawnattr *) attrp; 2858 syscallarg(char *const *) argv; 2859 syscallarg(char *const *) envp; 2860 } */ 2861 2862 int error; 2863 struct posix_spawn_file_actions *fa = NULL; 2864 struct posix_spawnattr *sa = NULL; 2865 pid_t pid; 2866 bool child_ok = false; 2867 rlim_t max_fileactions; 2868 proc_t *p = l1->l_proc; 2869 2870 /* check_posix_spawn() increments nprocs for us. */ 2871 error = check_posix_spawn(l1); 2872 if (error) { 2873 *retval = error; 2874 return 0; 2875 } 2876 2877 /* copy in file_actions struct */ 2878 if (SCARG(uap, file_actions) != NULL) { 2879 max_fileactions = 2 * uimin(p->p_rlimit[RLIMIT_NOFILE].rlim_cur, 2880 maxfiles); 2881 error = posix_spawn_fa_alloc(&fa, SCARG(uap, file_actions), 2882 max_fileactions); 2883 if (error) 2884 goto error_exit; 2885 } 2886 2887 /* copyin posix_spawnattr struct */ 2888 if (SCARG(uap, attrp) != NULL) { 2889 sa = kmem_alloc(sizeof(*sa), KM_SLEEP); 2890 error = copyin(SCARG(uap, attrp), sa, sizeof(*sa)); 2891 if (error) 2892 goto error_exit; 2893 } 2894 2895 /* 2896 * Do the spawn 2897 */ 2898 error = do_posix_spawn(l1, &pid, &child_ok, SCARG(uap, path), fa, sa, 2899 SCARG(uap, argv), SCARG(uap, envp), execve_fetch_element); 2900 if (error) 2901 goto error_exit; 2902 2903 if (error == 0 && SCARG(uap, pid) != NULL) 2904 error = copyout(&pid, SCARG(uap, pid), sizeof(pid)); 2905 2906 *retval = error; 2907 return 0; 2908 2909 error_exit: 2910 if (!child_ok) { 2911 (void)chgproccnt(kauth_cred_getuid(l1->l_cred), -1); 2912 atomic_dec_uint(&nprocs); 2913 2914 if (sa) 2915 kmem_free(sa, sizeof(*sa)); 2916 if (fa) 2917 posix_spawn_fa_free(fa, fa->len); 2918 } 2919 2920 *retval = error; 2921 return 0; 2922 } 2923 2924 void 2925 exec_free_emul_arg(struct exec_package *epp) 2926 { 2927 if (epp->ep_emul_arg_free != NULL) { 2928 KASSERT(epp->ep_emul_arg != NULL); 2929 (*epp->ep_emul_arg_free)(epp->ep_emul_arg); 2930 epp->ep_emul_arg_free = NULL; 2931 epp->ep_emul_arg = NULL; 2932 } else { 2933 KASSERT(epp->ep_emul_arg == NULL); 2934 } 2935 } 2936 2937 #ifdef DEBUG_EXEC 2938 static void 2939 dump_vmcmds(const struct exec_package * const epp, size_t x, int error) 2940 { 2941 struct exec_vmcmd *vp = &epp->ep_vmcmds.evs_cmds[0]; 2942 size_t j; 2943 2944 if (error == 0) 2945 DPRINTF(("vmcmds %u\n", epp->ep_vmcmds.evs_used)); 2946 else 2947 DPRINTF(("vmcmds %zu/%u, error %d\n", x, 2948 epp->ep_vmcmds.evs_used, error)); 2949 2950 for (j = 0; j < epp->ep_vmcmds.evs_used; j++) { 2951 DPRINTF(("vmcmd[%zu] = vmcmd_map_%s %#" 2952 PRIxVADDR"/%#"PRIxVSIZE" fd@%#" 2953 PRIxVSIZE" prot=0%o flags=%d\n", j, 2954 vp[j].ev_proc == vmcmd_map_pagedvn ? 2955 "pagedvn" : 2956 vp[j].ev_proc == vmcmd_map_readvn ? 2957 "readvn" : 2958 vp[j].ev_proc == vmcmd_map_zero ? 2959 "zero" : "*unknown*", 2960 vp[j].ev_addr, vp[j].ev_len, 2961 vp[j].ev_offset, vp[j].ev_prot, 2962 vp[j].ev_flags)); 2963 if (error != 0 && j == x) 2964 DPRINTF((" ^--- failed\n")); 2965 } 2966 } 2967 #endif 2968