uipc_socket.c revision 1.259.2.5
1 /* $NetBSD: uipc_socket.c,v 1.259.2.5 2018/11/26 01:52:50 pgoyette Exp $ */ 2 3 /*- 4 * Copyright (c) 2002, 2007, 2008, 2009 The NetBSD Foundation, Inc. 5 * All rights reserved. 6 * 7 * This code is derived from software contributed to The NetBSD Foundation 8 * by Jason R. Thorpe of Wasabi Systems, Inc, and 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) 2004 The FreeBSD Foundation 34 * Copyright (c) 2004 Robert Watson 35 * Copyright (c) 1982, 1986, 1988, 1990, 1993 36 * The Regents of the University of California. 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. Neither the name of the University nor the names of its contributors 47 * may be used to endorse or promote products derived from this software 48 * without specific prior written permission. 49 * 50 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 51 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 52 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 53 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 54 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 55 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 56 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 57 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 58 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 59 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 60 * SUCH DAMAGE. 61 * 62 * @(#)uipc_socket.c 8.6 (Berkeley) 5/2/95 63 */ 64 65 /* 66 * Socket operation routines. 67 * 68 * These routines are called by the routines in sys_socket.c or from a 69 * system process, and implement the semantics of socket operations by 70 * switching out to the protocol specific routines. 71 */ 72 73 #include <sys/cdefs.h> 74 __KERNEL_RCSID(0, "$NetBSD: uipc_socket.c,v 1.259.2.5 2018/11/26 01:52:50 pgoyette Exp $"); 75 76 #ifdef _KERNEL_OPT 77 #include "opt_compat_netbsd.h" 78 #include "opt_sock_counters.h" 79 #include "opt_sosend_loan.h" 80 #include "opt_mbuftrace.h" 81 #include "opt_somaxkva.h" 82 #include "opt_multiprocessor.h" /* XXX */ 83 #include "opt_sctp.h" 84 #endif 85 86 #include <sys/param.h> 87 #include <sys/systm.h> 88 #include <sys/proc.h> 89 #include <sys/file.h> 90 #include <sys/filedesc.h> 91 #include <sys/kmem.h> 92 #include <sys/mbuf.h> 93 #include <sys/domain.h> 94 #include <sys/kernel.h> 95 #include <sys/protosw.h> 96 #include <sys/socket.h> 97 #include <sys/socketvar.h> 98 #include <sys/signalvar.h> 99 #include <sys/resourcevar.h> 100 #include <sys/uidinfo.h> 101 #include <sys/event.h> 102 #include <sys/poll.h> 103 #include <sys/kauth.h> 104 #include <sys/mutex.h> 105 #include <sys/condvar.h> 106 #include <sys/kthread.h> 107 108 #ifdef COMPAT_50 109 #include <compat/sys/time.h> 110 #include <compat/sys/socket.h> 111 #endif 112 113 #include <uvm/uvm_extern.h> 114 #include <uvm/uvm_loan.h> 115 #include <uvm/uvm_page.h> 116 117 MALLOC_DEFINE(M_SONAME, "soname", "socket name"); 118 119 extern const struct fileops socketops; 120 121 static int sooptions; 122 extern int somaxconn; /* patchable (XXX sysctl) */ 123 int somaxconn = SOMAXCONN; 124 kmutex_t *softnet_lock; 125 126 #ifdef SOSEND_COUNTERS 127 #include <sys/device.h> 128 129 static struct evcnt sosend_loan_big = EVCNT_INITIALIZER(EVCNT_TYPE_MISC, 130 NULL, "sosend", "loan big"); 131 static struct evcnt sosend_copy_big = EVCNT_INITIALIZER(EVCNT_TYPE_MISC, 132 NULL, "sosend", "copy big"); 133 static struct evcnt sosend_copy_small = EVCNT_INITIALIZER(EVCNT_TYPE_MISC, 134 NULL, "sosend", "copy small"); 135 static struct evcnt sosend_kvalimit = EVCNT_INITIALIZER(EVCNT_TYPE_MISC, 136 NULL, "sosend", "kva limit"); 137 138 #define SOSEND_COUNTER_INCR(ev) (ev)->ev_count++ 139 140 EVCNT_ATTACH_STATIC(sosend_loan_big); 141 EVCNT_ATTACH_STATIC(sosend_copy_big); 142 EVCNT_ATTACH_STATIC(sosend_copy_small); 143 EVCNT_ATTACH_STATIC(sosend_kvalimit); 144 #else 145 146 #define SOSEND_COUNTER_INCR(ev) /* nothing */ 147 148 #endif /* SOSEND_COUNTERS */ 149 150 #if defined(SOSEND_NO_LOAN) || defined(MULTIPROCESSOR) 151 int sock_loan_thresh = -1; 152 #else 153 int sock_loan_thresh = 4096; 154 #endif 155 156 static kmutex_t so_pendfree_lock; 157 static struct mbuf *so_pendfree = NULL; 158 159 #ifndef SOMAXKVA 160 #define SOMAXKVA (16 * 1024 * 1024) 161 #endif 162 int somaxkva = SOMAXKVA; 163 static int socurkva; 164 static kcondvar_t socurkva_cv; 165 166 static kauth_listener_t socket_listener; 167 168 #define SOCK_LOAN_CHUNK 65536 169 170 static void sopendfree_thread(void *); 171 static kcondvar_t pendfree_thread_cv; 172 static lwp_t *sopendfree_lwp; 173 174 static void sysctl_kern_socket_setup(void); 175 static struct sysctllog *socket_sysctllog; 176 177 static vsize_t 178 sokvareserve(struct socket *so, vsize_t len) 179 { 180 int error; 181 182 mutex_enter(&so_pendfree_lock); 183 while (socurkva + len > somaxkva) { 184 SOSEND_COUNTER_INCR(&sosend_kvalimit); 185 error = cv_wait_sig(&socurkva_cv, &so_pendfree_lock); 186 if (error) { 187 len = 0; 188 break; 189 } 190 } 191 socurkva += len; 192 mutex_exit(&so_pendfree_lock); 193 return len; 194 } 195 196 static void 197 sokvaunreserve(vsize_t len) 198 { 199 200 mutex_enter(&so_pendfree_lock); 201 socurkva -= len; 202 cv_broadcast(&socurkva_cv); 203 mutex_exit(&so_pendfree_lock); 204 } 205 206 /* 207 * sokvaalloc: allocate kva for loan. 208 */ 209 210 vaddr_t 211 sokvaalloc(vaddr_t sva, vsize_t len, struct socket *so) 212 { 213 vaddr_t lva; 214 215 /* 216 * reserve kva. 217 */ 218 219 if (sokvareserve(so, len) == 0) 220 return 0; 221 222 /* 223 * allocate kva. 224 */ 225 226 lva = uvm_km_alloc(kernel_map, len, atop(sva) & uvmexp.colormask, 227 UVM_KMF_COLORMATCH | UVM_KMF_VAONLY | UVM_KMF_WAITVA); 228 if (lva == 0) { 229 sokvaunreserve(len); 230 return (0); 231 } 232 233 return lva; 234 } 235 236 /* 237 * sokvafree: free kva for loan. 238 */ 239 240 void 241 sokvafree(vaddr_t sva, vsize_t len) 242 { 243 244 /* 245 * free kva. 246 */ 247 248 uvm_km_free(kernel_map, sva, len, UVM_KMF_VAONLY); 249 250 /* 251 * unreserve kva. 252 */ 253 254 sokvaunreserve(len); 255 } 256 257 static void 258 sodoloanfree(struct vm_page **pgs, void *buf, size_t size) 259 { 260 vaddr_t sva, eva; 261 vsize_t len; 262 int npgs; 263 264 KASSERT(pgs != NULL); 265 266 eva = round_page((vaddr_t) buf + size); 267 sva = trunc_page((vaddr_t) buf); 268 len = eva - sva; 269 npgs = len >> PAGE_SHIFT; 270 271 pmap_kremove(sva, len); 272 pmap_update(pmap_kernel()); 273 uvm_unloan(pgs, npgs, UVM_LOAN_TOPAGE); 274 sokvafree(sva, len); 275 } 276 277 /* 278 * sopendfree_thread: free mbufs on "pendfree" list. 279 * unlock and relock so_pendfree_lock when freeing mbufs. 280 */ 281 282 static void 283 sopendfree_thread(void *v) 284 { 285 struct mbuf *m, *next; 286 size_t rv; 287 288 mutex_enter(&so_pendfree_lock); 289 290 for (;;) { 291 rv = 0; 292 while (so_pendfree != NULL) { 293 m = so_pendfree; 294 so_pendfree = NULL; 295 mutex_exit(&so_pendfree_lock); 296 297 for (; m != NULL; m = next) { 298 next = m->m_next; 299 KASSERT((~m->m_flags & (M_EXT|M_EXT_PAGES)) == 300 0); 301 KASSERT(m->m_ext.ext_refcnt == 0); 302 303 rv += m->m_ext.ext_size; 304 sodoloanfree(m->m_ext.ext_pgs, m->m_ext.ext_buf, 305 m->m_ext.ext_size); 306 pool_cache_put(mb_cache, m); 307 } 308 309 mutex_enter(&so_pendfree_lock); 310 } 311 if (rv) 312 cv_broadcast(&socurkva_cv); 313 cv_wait(&pendfree_thread_cv, &so_pendfree_lock); 314 } 315 panic("sopendfree_thread"); 316 /* NOTREACHED */ 317 } 318 319 void 320 soloanfree(struct mbuf *m, void *buf, size_t size, void *arg) 321 { 322 323 KASSERT(m != NULL); 324 325 /* 326 * postpone freeing mbuf. 327 * 328 * we can't do it in interrupt context 329 * because we need to put kva back to kernel_map. 330 */ 331 332 mutex_enter(&so_pendfree_lock); 333 m->m_next = so_pendfree; 334 so_pendfree = m; 335 cv_signal(&pendfree_thread_cv); 336 mutex_exit(&so_pendfree_lock); 337 } 338 339 static long 340 sosend_loan(struct socket *so, struct uio *uio, struct mbuf *m, long space) 341 { 342 struct iovec *iov = uio->uio_iov; 343 vaddr_t sva, eva; 344 vsize_t len; 345 vaddr_t lva; 346 int npgs, error; 347 vaddr_t va; 348 int i; 349 350 if (VMSPACE_IS_KERNEL_P(uio->uio_vmspace)) 351 return (0); 352 353 if (iov->iov_len < (size_t) space) 354 space = iov->iov_len; 355 if (space > SOCK_LOAN_CHUNK) 356 space = SOCK_LOAN_CHUNK; 357 358 eva = round_page((vaddr_t) iov->iov_base + space); 359 sva = trunc_page((vaddr_t) iov->iov_base); 360 len = eva - sva; 361 npgs = len >> PAGE_SHIFT; 362 363 KASSERT(npgs <= M_EXT_MAXPAGES); 364 365 lva = sokvaalloc(sva, len, so); 366 if (lva == 0) 367 return 0; 368 369 error = uvm_loan(&uio->uio_vmspace->vm_map, sva, len, 370 m->m_ext.ext_pgs, UVM_LOAN_TOPAGE); 371 if (error) { 372 sokvafree(lva, len); 373 return (0); 374 } 375 376 for (i = 0, va = lva; i < npgs; i++, va += PAGE_SIZE) 377 pmap_kenter_pa(va, VM_PAGE_TO_PHYS(m->m_ext.ext_pgs[i]), 378 VM_PROT_READ, 0); 379 pmap_update(pmap_kernel()); 380 381 lva += (vaddr_t) iov->iov_base & PAGE_MASK; 382 383 MEXTADD(m, (void *) lva, space, M_MBUF, soloanfree, so); 384 m->m_flags |= M_EXT_PAGES | M_EXT_ROMAP; 385 386 uio->uio_resid -= space; 387 /* uio_offset not updated, not set/used for write(2) */ 388 uio->uio_iov->iov_base = (char *)uio->uio_iov->iov_base + space; 389 uio->uio_iov->iov_len -= space; 390 if (uio->uio_iov->iov_len == 0) { 391 uio->uio_iov++; 392 uio->uio_iovcnt--; 393 } 394 395 return (space); 396 } 397 398 struct mbuf * 399 getsombuf(struct socket *so, int type) 400 { 401 struct mbuf *m; 402 403 m = m_get(M_WAIT, type); 404 MCLAIM(m, so->so_mowner); 405 return m; 406 } 407 408 static int 409 socket_listener_cb(kauth_cred_t cred, kauth_action_t action, void *cookie, 410 void *arg0, void *arg1, void *arg2, void *arg3) 411 { 412 int result; 413 enum kauth_network_req req; 414 415 result = KAUTH_RESULT_DEFER; 416 req = (enum kauth_network_req)arg0; 417 418 if ((action != KAUTH_NETWORK_SOCKET) && 419 (action != KAUTH_NETWORK_BIND)) 420 return result; 421 422 switch (req) { 423 case KAUTH_REQ_NETWORK_BIND_PORT: 424 result = KAUTH_RESULT_ALLOW; 425 break; 426 427 case KAUTH_REQ_NETWORK_SOCKET_DROP: { 428 /* Normal users can only drop their own connections. */ 429 struct socket *so = (struct socket *)arg1; 430 431 if (so->so_cred && proc_uidmatch(cred, so->so_cred) == 0) 432 result = KAUTH_RESULT_ALLOW; 433 434 break; 435 } 436 437 case KAUTH_REQ_NETWORK_SOCKET_OPEN: 438 /* We allow "raw" routing/bluetooth sockets to anyone. */ 439 switch ((u_long)arg1) { 440 case PF_ROUTE: 441 case PF_OROUTE: 442 case PF_BLUETOOTH: 443 case PF_CAN: 444 result = KAUTH_RESULT_ALLOW; 445 break; 446 default: 447 /* Privileged, let secmodel handle this. */ 448 if ((u_long)arg2 == SOCK_RAW) 449 break; 450 result = KAUTH_RESULT_ALLOW; 451 break; 452 } 453 break; 454 455 case KAUTH_REQ_NETWORK_SOCKET_CANSEE: 456 result = KAUTH_RESULT_ALLOW; 457 458 break; 459 460 default: 461 break; 462 } 463 464 return result; 465 } 466 467 void 468 soinit(void) 469 { 470 471 sysctl_kern_socket_setup(); 472 473 mutex_init(&so_pendfree_lock, MUTEX_DEFAULT, IPL_VM); 474 softnet_lock = mutex_obj_alloc(MUTEX_DEFAULT, IPL_NONE); 475 cv_init(&socurkva_cv, "sokva"); 476 cv_init(&pendfree_thread_cv, "sopendfr"); 477 soinit2(); 478 479 /* Set the initial adjusted socket buffer size. */ 480 if (sb_max_set(sb_max)) 481 panic("bad initial sb_max value: %lu", sb_max); 482 483 socket_listener = kauth_listen_scope(KAUTH_SCOPE_NETWORK, 484 socket_listener_cb, NULL); 485 } 486 487 void 488 soinit1(void) 489 { 490 int error = kthread_create(PRI_NONE, KTHREAD_MPSAFE, NULL, 491 sopendfree_thread, NULL, &sopendfree_lwp, "sopendfree"); 492 if (error) 493 panic("soinit1 %d", error); 494 } 495 496 /* 497 * socreate: create a new socket of the specified type and the protocol. 498 * 499 * => Caller may specify another socket for lock sharing (must not be held). 500 * => Returns the new socket without lock held. 501 */ 502 int 503 socreate(int dom, struct socket **aso, int type, int proto, struct lwp *l, 504 struct socket *lockso) 505 { 506 const struct protosw *prp; 507 struct socket *so; 508 uid_t uid; 509 int error; 510 kmutex_t *lock; 511 512 error = kauth_authorize_network(l->l_cred, KAUTH_NETWORK_SOCKET, 513 KAUTH_REQ_NETWORK_SOCKET_OPEN, KAUTH_ARG(dom), KAUTH_ARG(type), 514 KAUTH_ARG(proto)); 515 if (error != 0) 516 return error; 517 518 if (proto) 519 prp = pffindproto(dom, proto, type); 520 else 521 prp = pffindtype(dom, type); 522 if (prp == NULL) { 523 /* no support for domain */ 524 if (pffinddomain(dom) == 0) 525 return EAFNOSUPPORT; 526 /* no support for socket type */ 527 if (proto == 0 && type != 0) 528 return EPROTOTYPE; 529 return EPROTONOSUPPORT; 530 } 531 if (prp->pr_usrreqs == NULL) 532 return EPROTONOSUPPORT; 533 if (prp->pr_type != type) 534 return EPROTOTYPE; 535 536 so = soget(true); 537 so->so_type = type; 538 so->so_proto = prp; 539 so->so_send = sosend; 540 so->so_receive = soreceive; 541 so->so_options = sooptions; 542 #ifdef MBUFTRACE 543 so->so_rcv.sb_mowner = &prp->pr_domain->dom_mowner; 544 so->so_snd.sb_mowner = &prp->pr_domain->dom_mowner; 545 so->so_mowner = &prp->pr_domain->dom_mowner; 546 #endif 547 uid = kauth_cred_geteuid(l->l_cred); 548 so->so_uidinfo = uid_find(uid); 549 so->so_cpid = l->l_proc->p_pid; 550 551 /* 552 * Lock assigned and taken during PCB attach, unless we share 553 * the lock with another socket, e.g. socketpair(2) case. 554 */ 555 if (lockso) { 556 lock = lockso->so_lock; 557 so->so_lock = lock; 558 mutex_obj_hold(lock); 559 mutex_enter(lock); 560 } 561 562 /* Attach the PCB (returns with the socket lock held). */ 563 error = (*prp->pr_usrreqs->pr_attach)(so, proto); 564 KASSERT(solocked(so)); 565 566 if (error) { 567 KASSERT(so->so_pcb == NULL); 568 so->so_state |= SS_NOFDREF; 569 sofree(so); 570 return error; 571 } 572 so->so_cred = kauth_cred_dup(l->l_cred); 573 sounlock(so); 574 575 *aso = so; 576 return 0; 577 } 578 579 /* 580 * fsocreate: create a socket and a file descriptor associated with it. 581 * 582 * => On success, write file descriptor to fdout and return zero. 583 * => On failure, return non-zero; *fdout will be undefined. 584 */ 585 int 586 fsocreate(int domain, struct socket **sop, int type, int proto, int *fdout) 587 { 588 lwp_t *l = curlwp; 589 int error, fd, flags; 590 struct socket *so; 591 struct file *fp; 592 593 if ((error = fd_allocfile(&fp, &fd)) != 0) { 594 return error; 595 } 596 flags = type & SOCK_FLAGS_MASK; 597 fd_set_exclose(l, fd, (flags & SOCK_CLOEXEC) != 0); 598 fp->f_flag = FREAD|FWRITE|((flags & SOCK_NONBLOCK) ? FNONBLOCK : 0)| 599 ((flags & SOCK_NOSIGPIPE) ? FNOSIGPIPE : 0); 600 fp->f_type = DTYPE_SOCKET; 601 fp->f_ops = &socketops; 602 603 type &= ~SOCK_FLAGS_MASK; 604 error = socreate(domain, &so, type, proto, l, NULL); 605 if (error) { 606 fd_abort(curproc, fp, fd); 607 return error; 608 } 609 if (flags & SOCK_NONBLOCK) { 610 so->so_state |= SS_NBIO; 611 } 612 fp->f_socket = so; 613 fd_affix(curproc, fp, fd); 614 615 if (sop != NULL) { 616 *sop = so; 617 } 618 *fdout = fd; 619 return error; 620 } 621 622 int 623 sofamily(const struct socket *so) 624 { 625 const struct protosw *pr; 626 const struct domain *dom; 627 628 if ((pr = so->so_proto) == NULL) 629 return AF_UNSPEC; 630 if ((dom = pr->pr_domain) == NULL) 631 return AF_UNSPEC; 632 return dom->dom_family; 633 } 634 635 int 636 sobind(struct socket *so, struct sockaddr *nam, struct lwp *l) 637 { 638 int error; 639 640 solock(so); 641 if (nam->sa_family != so->so_proto->pr_domain->dom_family) { 642 sounlock(so); 643 return EAFNOSUPPORT; 644 } 645 error = (*so->so_proto->pr_usrreqs->pr_bind)(so, nam, l); 646 sounlock(so); 647 return error; 648 } 649 650 int 651 solisten(struct socket *so, int backlog, struct lwp *l) 652 { 653 int error; 654 short oldopt, oldqlimit; 655 656 solock(so); 657 if ((so->so_state & (SS_ISCONNECTED | SS_ISCONNECTING | 658 SS_ISDISCONNECTING)) != 0) { 659 sounlock(so); 660 return EINVAL; 661 } 662 oldopt = so->so_options; 663 oldqlimit = so->so_qlimit; 664 if (TAILQ_EMPTY(&so->so_q)) 665 so->so_options |= SO_ACCEPTCONN; 666 if (backlog < 0) 667 backlog = 0; 668 so->so_qlimit = uimin(backlog, somaxconn); 669 670 error = (*so->so_proto->pr_usrreqs->pr_listen)(so, l); 671 if (error != 0) { 672 so->so_options = oldopt; 673 so->so_qlimit = oldqlimit; 674 sounlock(so); 675 return error; 676 } 677 sounlock(so); 678 return 0; 679 } 680 681 void 682 sofree(struct socket *so) 683 { 684 u_int refs; 685 686 KASSERT(solocked(so)); 687 688 if (so->so_pcb || (so->so_state & SS_NOFDREF) == 0) { 689 sounlock(so); 690 return; 691 } 692 if (so->so_head) { 693 /* 694 * We must not decommission a socket that's on the accept(2) 695 * queue. If we do, then accept(2) may hang after select(2) 696 * indicated that the listening socket was ready. 697 */ 698 if (!soqremque(so, 0)) { 699 sounlock(so); 700 return; 701 } 702 } 703 if (so->so_rcv.sb_hiwat) 704 (void)chgsbsize(so->so_uidinfo, &so->so_rcv.sb_hiwat, 0, 705 RLIM_INFINITY); 706 if (so->so_snd.sb_hiwat) 707 (void)chgsbsize(so->so_uidinfo, &so->so_snd.sb_hiwat, 0, 708 RLIM_INFINITY); 709 sbrelease(&so->so_snd, so); 710 KASSERT(!cv_has_waiters(&so->so_cv)); 711 KASSERT(!cv_has_waiters(&so->so_rcv.sb_cv)); 712 KASSERT(!cv_has_waiters(&so->so_snd.sb_cv)); 713 sorflush(so); 714 refs = so->so_aborting; /* XXX */ 715 /* Remove acccept filter if one is present. */ 716 if (so->so_accf != NULL) 717 (void)accept_filt_clear(so); 718 sounlock(so); 719 if (refs == 0) /* XXX */ 720 soput(so); 721 } 722 723 /* 724 * soclose: close a socket on last file table reference removal. 725 * Initiate disconnect if connected. Free socket when disconnect complete. 726 */ 727 int 728 soclose(struct socket *so) 729 { 730 struct socket *so2; 731 int error = 0; 732 733 solock(so); 734 if (so->so_options & SO_ACCEPTCONN) { 735 for (;;) { 736 if ((so2 = TAILQ_FIRST(&so->so_q0)) != 0) { 737 KASSERT(solocked2(so, so2)); 738 (void) soqremque(so2, 0); 739 /* soabort drops the lock. */ 740 (void) soabort(so2); 741 solock(so); 742 continue; 743 } 744 if ((so2 = TAILQ_FIRST(&so->so_q)) != 0) { 745 KASSERT(solocked2(so, so2)); 746 (void) soqremque(so2, 1); 747 /* soabort drops the lock. */ 748 (void) soabort(so2); 749 solock(so); 750 continue; 751 } 752 break; 753 } 754 } 755 if (so->so_pcb == NULL) 756 goto discard; 757 if (so->so_state & SS_ISCONNECTED) { 758 if ((so->so_state & SS_ISDISCONNECTING) == 0) { 759 error = sodisconnect(so); 760 if (error) 761 goto drop; 762 } 763 if (so->so_options & SO_LINGER) { 764 if ((so->so_state & (SS_ISDISCONNECTING|SS_NBIO)) == 765 (SS_ISDISCONNECTING|SS_NBIO)) 766 goto drop; 767 while (so->so_state & SS_ISCONNECTED) { 768 error = sowait(so, true, so->so_linger * hz); 769 if (error) 770 break; 771 } 772 } 773 } 774 drop: 775 if (so->so_pcb) { 776 KASSERT(solocked(so)); 777 (*so->so_proto->pr_usrreqs->pr_detach)(so); 778 } 779 discard: 780 KASSERT((so->so_state & SS_NOFDREF) == 0); 781 kauth_cred_free(so->so_cred); 782 so->so_state |= SS_NOFDREF; 783 sofree(so); 784 return error; 785 } 786 787 /* 788 * Must be called with the socket locked.. Will return with it unlocked. 789 */ 790 int 791 soabort(struct socket *so) 792 { 793 u_int refs; 794 int error; 795 796 KASSERT(solocked(so)); 797 KASSERT(so->so_head == NULL); 798 799 so->so_aborting++; /* XXX */ 800 error = (*so->so_proto->pr_usrreqs->pr_abort)(so); 801 refs = --so->so_aborting; /* XXX */ 802 if (error || (refs == 0)) { 803 sofree(so); 804 } else { 805 sounlock(so); 806 } 807 return error; 808 } 809 810 int 811 soaccept(struct socket *so, struct sockaddr *nam) 812 { 813 int error; 814 815 KASSERT(solocked(so)); 816 KASSERT((so->so_state & SS_NOFDREF) != 0); 817 818 so->so_state &= ~SS_NOFDREF; 819 if ((so->so_state & SS_ISDISCONNECTED) == 0 || 820 (so->so_proto->pr_flags & PR_ABRTACPTDIS) == 0) 821 error = (*so->so_proto->pr_usrreqs->pr_accept)(so, nam); 822 else 823 error = ECONNABORTED; 824 825 return error; 826 } 827 828 int 829 soconnect(struct socket *so, struct sockaddr *nam, struct lwp *l) 830 { 831 int error; 832 833 KASSERT(solocked(so)); 834 835 if (so->so_options & SO_ACCEPTCONN) 836 return EOPNOTSUPP; 837 /* 838 * If protocol is connection-based, can only connect once. 839 * Otherwise, if connected, try to disconnect first. 840 * This allows user to disconnect by connecting to, e.g., 841 * a null address. 842 */ 843 if (so->so_state & (SS_ISCONNECTED|SS_ISCONNECTING) && 844 ((so->so_proto->pr_flags & PR_CONNREQUIRED) || 845 (error = sodisconnect(so)))) { 846 error = EISCONN; 847 } else { 848 if (nam->sa_family != so->so_proto->pr_domain->dom_family) { 849 return EAFNOSUPPORT; 850 } 851 error = (*so->so_proto->pr_usrreqs->pr_connect)(so, nam, l); 852 } 853 854 return error; 855 } 856 857 int 858 soconnect2(struct socket *so1, struct socket *so2) 859 { 860 KASSERT(solocked2(so1, so2)); 861 862 return (*so1->so_proto->pr_usrreqs->pr_connect2)(so1, so2); 863 } 864 865 int 866 sodisconnect(struct socket *so) 867 { 868 int error; 869 870 KASSERT(solocked(so)); 871 872 if ((so->so_state & SS_ISCONNECTED) == 0) { 873 error = ENOTCONN; 874 } else if (so->so_state & SS_ISDISCONNECTING) { 875 error = EALREADY; 876 } else { 877 error = (*so->so_proto->pr_usrreqs->pr_disconnect)(so); 878 } 879 return (error); 880 } 881 882 #define SBLOCKWAIT(f) (((f) & MSG_DONTWAIT) ? M_NOWAIT : M_WAITOK) 883 /* 884 * Send on a socket. 885 * If send must go all at once and message is larger than 886 * send buffering, then hard error. 887 * Lock against other senders. 888 * If must go all at once and not enough room now, then 889 * inform user that this would block and do nothing. 890 * Otherwise, if nonblocking, send as much as possible. 891 * The data to be sent is described by "uio" if nonzero, 892 * otherwise by the mbuf chain "top" (which must be null 893 * if uio is not). Data provided in mbuf chain must be small 894 * enough to send all at once. 895 * 896 * Returns nonzero on error, timeout or signal; callers 897 * must check for short counts if EINTR/ERESTART are returned. 898 * Data and control buffers are freed on return. 899 */ 900 int 901 sosend(struct socket *so, struct sockaddr *addr, struct uio *uio, 902 struct mbuf *top, struct mbuf *control, int flags, struct lwp *l) 903 { 904 struct mbuf **mp, *m; 905 long space, len, resid, clen, mlen; 906 int error, s, dontroute, atomic; 907 short wakeup_state = 0; 908 909 clen = 0; 910 911 /* 912 * solock() provides atomicity of access. splsoftnet() prevents 913 * protocol processing soft interrupts from interrupting us and 914 * blocking (expensive). 915 */ 916 s = splsoftnet(); 917 solock(so); 918 atomic = sosendallatonce(so) || top; 919 if (uio) 920 resid = uio->uio_resid; 921 else 922 resid = top->m_pkthdr.len; 923 /* 924 * In theory resid should be unsigned. 925 * However, space must be signed, as it might be less than 0 926 * if we over-committed, and we must use a signed comparison 927 * of space and resid. On the other hand, a negative resid 928 * causes us to loop sending 0-length segments to the protocol. 929 */ 930 if (resid < 0) { 931 error = EINVAL; 932 goto out; 933 } 934 dontroute = 935 (flags & MSG_DONTROUTE) && (so->so_options & SO_DONTROUTE) == 0 && 936 (so->so_proto->pr_flags & PR_ATOMIC); 937 l->l_ru.ru_msgsnd++; 938 if (control) 939 clen = control->m_len; 940 restart: 941 if ((error = sblock(&so->so_snd, SBLOCKWAIT(flags))) != 0) 942 goto out; 943 do { 944 if (so->so_state & SS_CANTSENDMORE) { 945 error = EPIPE; 946 goto release; 947 } 948 if (so->so_error) { 949 error = so->so_error; 950 so->so_error = 0; 951 goto release; 952 } 953 if ((so->so_state & SS_ISCONNECTED) == 0) { 954 if (so->so_proto->pr_flags & PR_CONNREQUIRED) { 955 if (resid || clen == 0) { 956 error = ENOTCONN; 957 goto release; 958 } 959 } else if (addr == NULL) { 960 error = EDESTADDRREQ; 961 goto release; 962 } 963 } 964 space = sbspace(&so->so_snd); 965 if (flags & MSG_OOB) 966 space += 1024; 967 if ((atomic && resid > so->so_snd.sb_hiwat) || 968 clen > so->so_snd.sb_hiwat) { 969 error = EMSGSIZE; 970 goto release; 971 } 972 if (space < resid + clen && 973 (atomic || space < so->so_snd.sb_lowat || space < clen)) { 974 if ((so->so_state & SS_NBIO) || (flags & MSG_NBIO)) { 975 error = EWOULDBLOCK; 976 goto release; 977 } 978 sbunlock(&so->so_snd); 979 if (wakeup_state & SS_RESTARTSYS) { 980 error = ERESTART; 981 goto out; 982 } 983 error = sbwait(&so->so_snd); 984 if (error) 985 goto out; 986 wakeup_state = so->so_state; 987 goto restart; 988 } 989 wakeup_state = 0; 990 mp = ⊤ 991 space -= clen; 992 do { 993 if (uio == NULL) { 994 /* 995 * Data is prepackaged in "top". 996 */ 997 resid = 0; 998 if (flags & MSG_EOR) 999 top->m_flags |= M_EOR; 1000 } else do { 1001 sounlock(so); 1002 splx(s); 1003 if (top == NULL) { 1004 m = m_gethdr(M_WAIT, MT_DATA); 1005 mlen = MHLEN; 1006 m->m_pkthdr.len = 0; 1007 m_reset_rcvif(m); 1008 } else { 1009 m = m_get(M_WAIT, MT_DATA); 1010 mlen = MLEN; 1011 } 1012 MCLAIM(m, so->so_snd.sb_mowner); 1013 if (sock_loan_thresh >= 0 && 1014 uio->uio_iov->iov_len >= sock_loan_thresh && 1015 space >= sock_loan_thresh && 1016 (len = sosend_loan(so, uio, m, 1017 space)) != 0) { 1018 SOSEND_COUNTER_INCR(&sosend_loan_big); 1019 space -= len; 1020 goto have_data; 1021 } 1022 if (resid >= MINCLSIZE && space >= MCLBYTES) { 1023 SOSEND_COUNTER_INCR(&sosend_copy_big); 1024 m_clget(m, M_DONTWAIT); 1025 if ((m->m_flags & M_EXT) == 0) 1026 goto nopages; 1027 mlen = MCLBYTES; 1028 if (atomic && top == 0) { 1029 len = lmin(MCLBYTES - max_hdr, 1030 resid); 1031 m->m_data += max_hdr; 1032 } else 1033 len = lmin(MCLBYTES, resid); 1034 space -= len; 1035 } else { 1036 nopages: 1037 SOSEND_COUNTER_INCR(&sosend_copy_small); 1038 len = lmin(lmin(mlen, resid), space); 1039 space -= len; 1040 /* 1041 * For datagram protocols, leave room 1042 * for protocol headers in first mbuf. 1043 */ 1044 if (atomic && top == 0 && len < mlen) 1045 MH_ALIGN(m, len); 1046 } 1047 error = uiomove(mtod(m, void *), (int)len, uio); 1048 have_data: 1049 resid = uio->uio_resid; 1050 m->m_len = len; 1051 *mp = m; 1052 top->m_pkthdr.len += len; 1053 s = splsoftnet(); 1054 solock(so); 1055 if (error != 0) 1056 goto release; 1057 mp = &m->m_next; 1058 if (resid <= 0) { 1059 if (flags & MSG_EOR) 1060 top->m_flags |= M_EOR; 1061 break; 1062 } 1063 } while (space > 0 && atomic); 1064 1065 if (so->so_state & SS_CANTSENDMORE) { 1066 error = EPIPE; 1067 goto release; 1068 } 1069 if (dontroute) 1070 so->so_options |= SO_DONTROUTE; 1071 if (resid > 0) 1072 so->so_state |= SS_MORETOCOME; 1073 if (flags & MSG_OOB) { 1074 error = (*so->so_proto->pr_usrreqs->pr_sendoob)( 1075 so, top, control); 1076 } else { 1077 error = (*so->so_proto->pr_usrreqs->pr_send)(so, 1078 top, addr, control, l); 1079 } 1080 if (dontroute) 1081 so->so_options &= ~SO_DONTROUTE; 1082 if (resid > 0) 1083 so->so_state &= ~SS_MORETOCOME; 1084 clen = 0; 1085 control = NULL; 1086 top = NULL; 1087 mp = ⊤ 1088 if (error != 0) 1089 goto release; 1090 } while (resid && space > 0); 1091 } while (resid); 1092 1093 release: 1094 sbunlock(&so->so_snd); 1095 out: 1096 sounlock(so); 1097 splx(s); 1098 if (top) 1099 m_freem(top); 1100 if (control) 1101 m_freem(control); 1102 return (error); 1103 } 1104 1105 /* 1106 * Following replacement or removal of the first mbuf on the first 1107 * mbuf chain of a socket buffer, push necessary state changes back 1108 * into the socket buffer so that other consumers see the values 1109 * consistently. 'nextrecord' is the callers locally stored value of 1110 * the original value of sb->sb_mb->m_nextpkt which must be restored 1111 * when the lead mbuf changes. NOTE: 'nextrecord' may be NULL. 1112 */ 1113 static void 1114 sbsync(struct sockbuf *sb, struct mbuf *nextrecord) 1115 { 1116 1117 KASSERT(solocked(sb->sb_so)); 1118 1119 /* 1120 * First, update for the new value of nextrecord. If necessary, 1121 * make it the first record. 1122 */ 1123 if (sb->sb_mb != NULL) 1124 sb->sb_mb->m_nextpkt = nextrecord; 1125 else 1126 sb->sb_mb = nextrecord; 1127 1128 /* 1129 * Now update any dependent socket buffer fields to reflect 1130 * the new state. This is an inline of SB_EMPTY_FIXUP, with 1131 * the addition of a second clause that takes care of the 1132 * case where sb_mb has been updated, but remains the last 1133 * record. 1134 */ 1135 if (sb->sb_mb == NULL) { 1136 sb->sb_mbtail = NULL; 1137 sb->sb_lastrecord = NULL; 1138 } else if (sb->sb_mb->m_nextpkt == NULL) 1139 sb->sb_lastrecord = sb->sb_mb; 1140 } 1141 1142 /* 1143 * Implement receive operations on a socket. 1144 * We depend on the way that records are added to the sockbuf 1145 * by sbappend*. In particular, each record (mbufs linked through m_next) 1146 * must begin with an address if the protocol so specifies, 1147 * followed by an optional mbuf or mbufs containing ancillary data, 1148 * and then zero or more mbufs of data. 1149 * In order to avoid blocking network interrupts for the entire time here, 1150 * we splx() while doing the actual copy to user space. 1151 * Although the sockbuf is locked, new data may still be appended, 1152 * and thus we must maintain consistency of the sockbuf during that time. 1153 * 1154 * The caller may receive the data as a single mbuf chain by supplying 1155 * an mbuf **mp0 for use in returning the chain. The uio is then used 1156 * only for the count in uio_resid. 1157 */ 1158 int 1159 soreceive(struct socket *so, struct mbuf **paddr, struct uio *uio, 1160 struct mbuf **mp0, struct mbuf **controlp, int *flagsp) 1161 { 1162 struct lwp *l = curlwp; 1163 struct mbuf *m, **mp, *mt; 1164 size_t len, offset, moff, orig_resid; 1165 int atomic, flags, error, s, type; 1166 const struct protosw *pr; 1167 struct mbuf *nextrecord; 1168 int mbuf_removed = 0; 1169 const struct domain *dom; 1170 short wakeup_state = 0; 1171 1172 pr = so->so_proto; 1173 atomic = pr->pr_flags & PR_ATOMIC; 1174 dom = pr->pr_domain; 1175 mp = mp0; 1176 type = 0; 1177 orig_resid = uio->uio_resid; 1178 1179 if (paddr != NULL) 1180 *paddr = NULL; 1181 if (controlp != NULL) 1182 *controlp = NULL; 1183 if (flagsp != NULL) 1184 flags = *flagsp &~ MSG_EOR; 1185 else 1186 flags = 0; 1187 1188 if (flags & MSG_OOB) { 1189 m = m_get(M_WAIT, MT_DATA); 1190 solock(so); 1191 error = (*pr->pr_usrreqs->pr_recvoob)(so, m, flags & MSG_PEEK); 1192 sounlock(so); 1193 if (error) 1194 goto bad; 1195 do { 1196 error = uiomove(mtod(m, void *), 1197 MIN(uio->uio_resid, m->m_len), uio); 1198 m = m_free(m); 1199 } while (uio->uio_resid > 0 && error == 0 && m); 1200 bad: 1201 if (m != NULL) 1202 m_freem(m); 1203 return error; 1204 } 1205 if (mp != NULL) 1206 *mp = NULL; 1207 1208 /* 1209 * solock() provides atomicity of access. splsoftnet() prevents 1210 * protocol processing soft interrupts from interrupting us and 1211 * blocking (expensive). 1212 */ 1213 s = splsoftnet(); 1214 solock(so); 1215 restart: 1216 if ((error = sblock(&so->so_rcv, SBLOCKWAIT(flags))) != 0) { 1217 sounlock(so); 1218 splx(s); 1219 return error; 1220 } 1221 1222 m = so->so_rcv.sb_mb; 1223 /* 1224 * If we have less data than requested, block awaiting more 1225 * (subject to any timeout) if: 1226 * 1. the current count is less than the low water mark, 1227 * 2. MSG_WAITALL is set, and it is possible to do the entire 1228 * receive operation at once if we block (resid <= hiwat), or 1229 * 3. MSG_DONTWAIT is not set. 1230 * If MSG_WAITALL is set but resid is larger than the receive buffer, 1231 * we have to do the receive in sections, and thus risk returning 1232 * a short count if a timeout or signal occurs after we start. 1233 */ 1234 if (m == NULL || 1235 ((flags & MSG_DONTWAIT) == 0 && 1236 so->so_rcv.sb_cc < uio->uio_resid && 1237 (so->so_rcv.sb_cc < so->so_rcv.sb_lowat || 1238 ((flags & MSG_WAITALL) && 1239 uio->uio_resid <= so->so_rcv.sb_hiwat)) && 1240 m->m_nextpkt == NULL && !atomic)) { 1241 #ifdef DIAGNOSTIC 1242 if (m == NULL && so->so_rcv.sb_cc) 1243 panic("receive 1"); 1244 #endif 1245 if (so->so_error || so->so_rerror) { 1246 if (m != NULL) 1247 goto dontblock; 1248 if (so->so_error) { 1249 error = so->so_error; 1250 so->so_error = 0; 1251 } else { 1252 error = so->so_rerror; 1253 so->so_rerror = 0; 1254 } 1255 goto release; 1256 } 1257 if (so->so_state & SS_CANTRCVMORE) { 1258 if (m != NULL) 1259 goto dontblock; 1260 else 1261 goto release; 1262 } 1263 for (; m != NULL; m = m->m_next) 1264 if (m->m_type == MT_OOBDATA || (m->m_flags & M_EOR)) { 1265 m = so->so_rcv.sb_mb; 1266 goto dontblock; 1267 } 1268 if ((so->so_state & (SS_ISCONNECTED|SS_ISCONNECTING)) == 0 && 1269 (so->so_proto->pr_flags & PR_CONNREQUIRED)) { 1270 error = ENOTCONN; 1271 goto release; 1272 } 1273 if (uio->uio_resid == 0) 1274 goto release; 1275 if ((so->so_state & SS_NBIO) || 1276 (flags & (MSG_DONTWAIT|MSG_NBIO))) { 1277 error = EWOULDBLOCK; 1278 goto release; 1279 } 1280 SBLASTRECORDCHK(&so->so_rcv, "soreceive sbwait 1"); 1281 SBLASTMBUFCHK(&so->so_rcv, "soreceive sbwait 1"); 1282 sbunlock(&so->so_rcv); 1283 if (wakeup_state & SS_RESTARTSYS) 1284 error = ERESTART; 1285 else 1286 error = sbwait(&so->so_rcv); 1287 if (error != 0) { 1288 sounlock(so); 1289 splx(s); 1290 return error; 1291 } 1292 wakeup_state = so->so_state; 1293 goto restart; 1294 } 1295 dontblock: 1296 /* 1297 * On entry here, m points to the first record of the socket buffer. 1298 * From this point onward, we maintain 'nextrecord' as a cache of the 1299 * pointer to the next record in the socket buffer. We must keep the 1300 * various socket buffer pointers and local stack versions of the 1301 * pointers in sync, pushing out modifications before dropping the 1302 * socket lock, and re-reading them when picking it up. 1303 * 1304 * Otherwise, we will race with the network stack appending new data 1305 * or records onto the socket buffer by using inconsistent/stale 1306 * versions of the field, possibly resulting in socket buffer 1307 * corruption. 1308 * 1309 * By holding the high-level sblock(), we prevent simultaneous 1310 * readers from pulling off the front of the socket buffer. 1311 */ 1312 if (l != NULL) 1313 l->l_ru.ru_msgrcv++; 1314 KASSERT(m == so->so_rcv.sb_mb); 1315 SBLASTRECORDCHK(&so->so_rcv, "soreceive 1"); 1316 SBLASTMBUFCHK(&so->so_rcv, "soreceive 1"); 1317 nextrecord = m->m_nextpkt; 1318 if (pr->pr_flags & PR_ADDR) { 1319 #ifdef DIAGNOSTIC 1320 if (m->m_type != MT_SONAME) 1321 panic("receive 1a"); 1322 #endif 1323 orig_resid = 0; 1324 if (flags & MSG_PEEK) { 1325 if (paddr) 1326 *paddr = m_copym(m, 0, m->m_len, M_DONTWAIT); 1327 m = m->m_next; 1328 } else { 1329 sbfree(&so->so_rcv, m); 1330 mbuf_removed = 1; 1331 if (paddr != NULL) { 1332 *paddr = m; 1333 so->so_rcv.sb_mb = m->m_next; 1334 m->m_next = NULL; 1335 m = so->so_rcv.sb_mb; 1336 } else { 1337 m = so->so_rcv.sb_mb = m_free(m); 1338 } 1339 sbsync(&so->so_rcv, nextrecord); 1340 } 1341 } 1342 if (pr->pr_flags & PR_ADDR_OPT) { 1343 /* 1344 * For SCTP we may be getting a 1345 * whole message OR a partial delivery. 1346 */ 1347 if (m->m_type == MT_SONAME) { 1348 orig_resid = 0; 1349 if (flags & MSG_PEEK) { 1350 if (paddr) 1351 *paddr = m_copym(m, 0, m->m_len, M_DONTWAIT); 1352 m = m->m_next; 1353 } else { 1354 sbfree(&so->so_rcv, m); 1355 if (paddr) { 1356 *paddr = m; 1357 so->so_rcv.sb_mb = m->m_next; 1358 m->m_next = 0; 1359 m = so->so_rcv.sb_mb; 1360 } else { 1361 m = so->so_rcv.sb_mb = m_free(m); 1362 } 1363 } 1364 } 1365 } 1366 1367 /* 1368 * Process one or more MT_CONTROL mbufs present before any data mbufs 1369 * in the first mbuf chain on the socket buffer. If MSG_PEEK, we 1370 * just copy the data; if !MSG_PEEK, we call into the protocol to 1371 * perform externalization (or freeing if controlp == NULL). 1372 */ 1373 if (__predict_false(m != NULL && m->m_type == MT_CONTROL)) { 1374 struct mbuf *cm = NULL, *cmn; 1375 struct mbuf **cme = &cm; 1376 1377 do { 1378 if (flags & MSG_PEEK) { 1379 if (controlp != NULL) { 1380 *controlp = m_copym(m, 0, m->m_len, M_DONTWAIT); 1381 controlp = &(*controlp)->m_next; 1382 } 1383 m = m->m_next; 1384 } else { 1385 sbfree(&so->so_rcv, m); 1386 so->so_rcv.sb_mb = m->m_next; 1387 m->m_next = NULL; 1388 *cme = m; 1389 cme = &(*cme)->m_next; 1390 m = so->so_rcv.sb_mb; 1391 } 1392 } while (m != NULL && m->m_type == MT_CONTROL); 1393 if ((flags & MSG_PEEK) == 0) 1394 sbsync(&so->so_rcv, nextrecord); 1395 for (; cm != NULL; cm = cmn) { 1396 cmn = cm->m_next; 1397 cm->m_next = NULL; 1398 type = mtod(cm, struct cmsghdr *)->cmsg_type; 1399 if (controlp != NULL) { 1400 if (dom->dom_externalize != NULL && 1401 type == SCM_RIGHTS) { 1402 sounlock(so); 1403 splx(s); 1404 error = (*dom->dom_externalize)(cm, l, 1405 (flags & MSG_CMSG_CLOEXEC) ? 1406 O_CLOEXEC : 0); 1407 s = splsoftnet(); 1408 solock(so); 1409 } 1410 *controlp = cm; 1411 while (*controlp != NULL) 1412 controlp = &(*controlp)->m_next; 1413 } else { 1414 /* 1415 * Dispose of any SCM_RIGHTS message that went 1416 * through the read path rather than recv. 1417 */ 1418 if (dom->dom_dispose != NULL && 1419 type == SCM_RIGHTS) { 1420 sounlock(so); 1421 (*dom->dom_dispose)(cm); 1422 solock(so); 1423 } 1424 m_freem(cm); 1425 } 1426 } 1427 if (m != NULL) 1428 nextrecord = so->so_rcv.sb_mb->m_nextpkt; 1429 else 1430 nextrecord = so->so_rcv.sb_mb; 1431 orig_resid = 0; 1432 } 1433 1434 /* If m is non-NULL, we have some data to read. */ 1435 if (__predict_true(m != NULL)) { 1436 type = m->m_type; 1437 if (type == MT_OOBDATA) 1438 flags |= MSG_OOB; 1439 } 1440 SBLASTRECORDCHK(&so->so_rcv, "soreceive 2"); 1441 SBLASTMBUFCHK(&so->so_rcv, "soreceive 2"); 1442 1443 moff = 0; 1444 offset = 0; 1445 while (m != NULL && uio->uio_resid > 0 && error == 0) { 1446 if (m->m_type == MT_OOBDATA) { 1447 if (type != MT_OOBDATA) 1448 break; 1449 } else if (type == MT_OOBDATA) 1450 break; 1451 #ifdef DIAGNOSTIC 1452 else if (m->m_type != MT_DATA && m->m_type != MT_HEADER) 1453 panic("receive 3"); 1454 #endif 1455 so->so_state &= ~SS_RCVATMARK; 1456 wakeup_state = 0; 1457 len = uio->uio_resid; 1458 if (so->so_oobmark && len > so->so_oobmark - offset) 1459 len = so->so_oobmark - offset; 1460 if (len > m->m_len - moff) 1461 len = m->m_len - moff; 1462 /* 1463 * If mp is set, just pass back the mbufs. 1464 * Otherwise copy them out via the uio, then free. 1465 * Sockbuf must be consistent here (points to current mbuf, 1466 * it points to next record) when we drop priority; 1467 * we must note any additions to the sockbuf when we 1468 * block interrupts again. 1469 */ 1470 if (mp == NULL) { 1471 SBLASTRECORDCHK(&so->so_rcv, "soreceive uiomove"); 1472 SBLASTMBUFCHK(&so->so_rcv, "soreceive uiomove"); 1473 sounlock(so); 1474 splx(s); 1475 error = uiomove(mtod(m, char *) + moff, len, uio); 1476 s = splsoftnet(); 1477 solock(so); 1478 if (error != 0) { 1479 /* 1480 * If any part of the record has been removed 1481 * (such as the MT_SONAME mbuf, which will 1482 * happen when PR_ADDR, and thus also 1483 * PR_ATOMIC, is set), then drop the entire 1484 * record to maintain the atomicity of the 1485 * receive operation. 1486 * 1487 * This avoids a later panic("receive 1a") 1488 * when compiled with DIAGNOSTIC. 1489 */ 1490 if (m && mbuf_removed && atomic) 1491 (void) sbdroprecord(&so->so_rcv); 1492 1493 goto release; 1494 } 1495 } else 1496 uio->uio_resid -= len; 1497 if (len == m->m_len - moff) { 1498 if (m->m_flags & M_EOR) 1499 flags |= MSG_EOR; 1500 #ifdef SCTP 1501 if (m->m_flags & M_NOTIFICATION) 1502 flags |= MSG_NOTIFICATION; 1503 #endif /* SCTP */ 1504 if (flags & MSG_PEEK) { 1505 m = m->m_next; 1506 moff = 0; 1507 } else { 1508 nextrecord = m->m_nextpkt; 1509 sbfree(&so->so_rcv, m); 1510 if (mp) { 1511 *mp = m; 1512 mp = &m->m_next; 1513 so->so_rcv.sb_mb = m = m->m_next; 1514 *mp = NULL; 1515 } else { 1516 m = so->so_rcv.sb_mb = m_free(m); 1517 } 1518 /* 1519 * If m != NULL, we also know that 1520 * so->so_rcv.sb_mb != NULL. 1521 */ 1522 KASSERT(so->so_rcv.sb_mb == m); 1523 if (m) { 1524 m->m_nextpkt = nextrecord; 1525 if (nextrecord == NULL) 1526 so->so_rcv.sb_lastrecord = m; 1527 } else { 1528 so->so_rcv.sb_mb = nextrecord; 1529 SB_EMPTY_FIXUP(&so->so_rcv); 1530 } 1531 SBLASTRECORDCHK(&so->so_rcv, "soreceive 3"); 1532 SBLASTMBUFCHK(&so->so_rcv, "soreceive 3"); 1533 } 1534 } else if (flags & MSG_PEEK) 1535 moff += len; 1536 else { 1537 if (mp != NULL) { 1538 mt = m_copym(m, 0, len, M_NOWAIT); 1539 if (__predict_false(mt == NULL)) { 1540 sounlock(so); 1541 mt = m_copym(m, 0, len, M_WAIT); 1542 solock(so); 1543 } 1544 *mp = mt; 1545 } 1546 m->m_data += len; 1547 m->m_len -= len; 1548 so->so_rcv.sb_cc -= len; 1549 } 1550 if (so->so_oobmark) { 1551 if ((flags & MSG_PEEK) == 0) { 1552 so->so_oobmark -= len; 1553 if (so->so_oobmark == 0) { 1554 so->so_state |= SS_RCVATMARK; 1555 break; 1556 } 1557 } else { 1558 offset += len; 1559 if (offset == so->so_oobmark) 1560 break; 1561 } 1562 } 1563 if (flags & MSG_EOR) 1564 break; 1565 /* 1566 * If the MSG_WAITALL flag is set (for non-atomic socket), 1567 * we must not quit until "uio->uio_resid == 0" or an error 1568 * termination. If a signal/timeout occurs, return 1569 * with a short count but without error. 1570 * Keep sockbuf locked against other readers. 1571 */ 1572 while (flags & MSG_WAITALL && m == NULL && uio->uio_resid > 0 && 1573 !sosendallatonce(so) && !nextrecord) { 1574 if (so->so_error || so->so_rerror || 1575 so->so_state & SS_CANTRCVMORE) 1576 break; 1577 /* 1578 * If we are peeking and the socket receive buffer is 1579 * full, stop since we can't get more data to peek at. 1580 */ 1581 if ((flags & MSG_PEEK) && sbspace(&so->so_rcv) <= 0) 1582 break; 1583 /* 1584 * If we've drained the socket buffer, tell the 1585 * protocol in case it needs to do something to 1586 * get it filled again. 1587 */ 1588 if ((pr->pr_flags & PR_WANTRCVD) && so->so_pcb) 1589 (*pr->pr_usrreqs->pr_rcvd)(so, flags, l); 1590 SBLASTRECORDCHK(&so->so_rcv, "soreceive sbwait 2"); 1591 SBLASTMBUFCHK(&so->so_rcv, "soreceive sbwait 2"); 1592 if (wakeup_state & SS_RESTARTSYS) 1593 error = ERESTART; 1594 else 1595 error = sbwait(&so->so_rcv); 1596 if (error != 0) { 1597 sbunlock(&so->so_rcv); 1598 sounlock(so); 1599 splx(s); 1600 return 0; 1601 } 1602 if ((m = so->so_rcv.sb_mb) != NULL) 1603 nextrecord = m->m_nextpkt; 1604 wakeup_state = so->so_state; 1605 } 1606 } 1607 1608 if (m && atomic) { 1609 flags |= MSG_TRUNC; 1610 if ((flags & MSG_PEEK) == 0) 1611 (void) sbdroprecord(&so->so_rcv); 1612 } 1613 if ((flags & MSG_PEEK) == 0) { 1614 if (m == NULL) { 1615 /* 1616 * First part is an inline SB_EMPTY_FIXUP(). Second 1617 * part makes sure sb_lastrecord is up-to-date if 1618 * there is still data in the socket buffer. 1619 */ 1620 so->so_rcv.sb_mb = nextrecord; 1621 if (so->so_rcv.sb_mb == NULL) { 1622 so->so_rcv.sb_mbtail = NULL; 1623 so->so_rcv.sb_lastrecord = NULL; 1624 } else if (nextrecord->m_nextpkt == NULL) 1625 so->so_rcv.sb_lastrecord = nextrecord; 1626 } 1627 SBLASTRECORDCHK(&so->so_rcv, "soreceive 4"); 1628 SBLASTMBUFCHK(&so->so_rcv, "soreceive 4"); 1629 if (pr->pr_flags & PR_WANTRCVD && so->so_pcb) 1630 (*pr->pr_usrreqs->pr_rcvd)(so, flags, l); 1631 } 1632 if (orig_resid == uio->uio_resid && orig_resid && 1633 (flags & MSG_EOR) == 0 && (so->so_state & SS_CANTRCVMORE) == 0) { 1634 sbunlock(&so->so_rcv); 1635 goto restart; 1636 } 1637 1638 if (flagsp != NULL) 1639 *flagsp |= flags; 1640 release: 1641 sbunlock(&so->so_rcv); 1642 sounlock(so); 1643 splx(s); 1644 return error; 1645 } 1646 1647 int 1648 soshutdown(struct socket *so, int how) 1649 { 1650 const struct protosw *pr; 1651 int error; 1652 1653 KASSERT(solocked(so)); 1654 1655 pr = so->so_proto; 1656 if (!(how == SHUT_RD || how == SHUT_WR || how == SHUT_RDWR)) 1657 return (EINVAL); 1658 1659 if (how == SHUT_RD || how == SHUT_RDWR) { 1660 sorflush(so); 1661 error = 0; 1662 } 1663 if (how == SHUT_WR || how == SHUT_RDWR) 1664 error = (*pr->pr_usrreqs->pr_shutdown)(so); 1665 1666 return error; 1667 } 1668 1669 void 1670 sorestart(struct socket *so) 1671 { 1672 /* 1673 * An application has called close() on an fd on which another 1674 * of its threads has called a socket system call. 1675 * Mark this and wake everyone up, and code that would block again 1676 * instead returns ERESTART. 1677 * On system call re-entry the fd is validated and EBADF returned. 1678 * Any other fd will block again on the 2nd syscall. 1679 */ 1680 solock(so); 1681 so->so_state |= SS_RESTARTSYS; 1682 cv_broadcast(&so->so_cv); 1683 cv_broadcast(&so->so_snd.sb_cv); 1684 cv_broadcast(&so->so_rcv.sb_cv); 1685 sounlock(so); 1686 } 1687 1688 void 1689 sorflush(struct socket *so) 1690 { 1691 struct sockbuf *sb, asb; 1692 const struct protosw *pr; 1693 1694 KASSERT(solocked(so)); 1695 1696 sb = &so->so_rcv; 1697 pr = so->so_proto; 1698 socantrcvmore(so); 1699 sb->sb_flags |= SB_NOINTR; 1700 (void )sblock(sb, M_WAITOK); 1701 sbunlock(sb); 1702 asb = *sb; 1703 /* 1704 * Clear most of the sockbuf structure, but leave some of the 1705 * fields valid. 1706 */ 1707 memset(&sb->sb_startzero, 0, 1708 sizeof(*sb) - offsetof(struct sockbuf, sb_startzero)); 1709 if (pr->pr_flags & PR_RIGHTS && pr->pr_domain->dom_dispose) { 1710 sounlock(so); 1711 (*pr->pr_domain->dom_dispose)(asb.sb_mb); 1712 solock(so); 1713 } 1714 sbrelease(&asb, so); 1715 } 1716 1717 /* 1718 * internal set SOL_SOCKET options 1719 */ 1720 static int 1721 sosetopt1(struct socket *so, const struct sockopt *sopt) 1722 { 1723 int error = EINVAL, opt; 1724 int optval = 0; /* XXX: gcc */ 1725 struct linger l; 1726 struct timeval tv; 1727 1728 switch ((opt = sopt->sopt_name)) { 1729 1730 case SO_ACCEPTFILTER: 1731 error = accept_filt_setopt(so, sopt); 1732 KASSERT(solocked(so)); 1733 break; 1734 1735 case SO_LINGER: 1736 error = sockopt_get(sopt, &l, sizeof(l)); 1737 solock(so); 1738 if (error) 1739 break; 1740 if (l.l_linger < 0 || l.l_linger > USHRT_MAX || 1741 l.l_linger > (INT_MAX / hz)) { 1742 error = EDOM; 1743 break; 1744 } 1745 so->so_linger = l.l_linger; 1746 if (l.l_onoff) 1747 so->so_options |= SO_LINGER; 1748 else 1749 so->so_options &= ~SO_LINGER; 1750 break; 1751 1752 case SO_DEBUG: 1753 case SO_KEEPALIVE: 1754 case SO_DONTROUTE: 1755 case SO_USELOOPBACK: 1756 case SO_BROADCAST: 1757 case SO_REUSEADDR: 1758 case SO_REUSEPORT: 1759 case SO_OOBINLINE: 1760 case SO_TIMESTAMP: 1761 case SO_NOSIGPIPE: 1762 case SO_RERROR: 1763 #ifdef SO_OTIMESTAMP 1764 case SO_OTIMESTAMP: 1765 #endif 1766 error = sockopt_getint(sopt, &optval); 1767 solock(so); 1768 if (error) 1769 break; 1770 if (optval) 1771 so->so_options |= opt; 1772 else 1773 so->so_options &= ~opt; 1774 break; 1775 1776 case SO_SNDBUF: 1777 case SO_RCVBUF: 1778 case SO_SNDLOWAT: 1779 case SO_RCVLOWAT: 1780 error = sockopt_getint(sopt, &optval); 1781 solock(so); 1782 if (error) 1783 break; 1784 1785 /* 1786 * Values < 1 make no sense for any of these 1787 * options, so disallow them. 1788 */ 1789 if (optval < 1) { 1790 error = EINVAL; 1791 break; 1792 } 1793 1794 switch (opt) { 1795 case SO_SNDBUF: 1796 if (sbreserve(&so->so_snd, (u_long)optval, so) == 0) { 1797 error = ENOBUFS; 1798 break; 1799 } 1800 so->so_snd.sb_flags &= ~SB_AUTOSIZE; 1801 break; 1802 1803 case SO_RCVBUF: 1804 if (sbreserve(&so->so_rcv, (u_long)optval, so) == 0) { 1805 error = ENOBUFS; 1806 break; 1807 } 1808 so->so_rcv.sb_flags &= ~SB_AUTOSIZE; 1809 break; 1810 1811 /* 1812 * Make sure the low-water is never greater than 1813 * the high-water. 1814 */ 1815 case SO_SNDLOWAT: 1816 if (optval > so->so_snd.sb_hiwat) 1817 optval = so->so_snd.sb_hiwat; 1818 1819 so->so_snd.sb_lowat = optval; 1820 break; 1821 1822 case SO_RCVLOWAT: 1823 if (optval > so->so_rcv.sb_hiwat) 1824 optval = so->so_rcv.sb_hiwat; 1825 1826 so->so_rcv.sb_lowat = optval; 1827 break; 1828 } 1829 break; 1830 1831 #ifdef COMPAT_50 1832 case SO_OSNDTIMEO: 1833 case SO_ORCVTIMEO: { 1834 struct timeval50 otv; 1835 error = sockopt_get(sopt, &otv, sizeof(otv)); 1836 if (error) { 1837 solock(so); 1838 break; 1839 } 1840 timeval50_to_timeval(&otv, &tv); 1841 opt = opt == SO_OSNDTIMEO ? SO_SNDTIMEO : SO_RCVTIMEO; 1842 error = 0; 1843 /*FALLTHROUGH*/ 1844 } 1845 #endif /* COMPAT_50 */ 1846 1847 case SO_SNDTIMEO: 1848 case SO_RCVTIMEO: 1849 if (error) 1850 error = sockopt_get(sopt, &tv, sizeof(tv)); 1851 solock(so); 1852 if (error) 1853 break; 1854 1855 if (tv.tv_sec > (INT_MAX - tv.tv_usec / tick) / hz) { 1856 error = EDOM; 1857 break; 1858 } 1859 1860 optval = tv.tv_sec * hz + tv.tv_usec / tick; 1861 if (optval == 0 && tv.tv_usec != 0) 1862 optval = 1; 1863 1864 switch (opt) { 1865 case SO_SNDTIMEO: 1866 so->so_snd.sb_timeo = optval; 1867 break; 1868 case SO_RCVTIMEO: 1869 so->so_rcv.sb_timeo = optval; 1870 break; 1871 } 1872 break; 1873 1874 default: 1875 solock(so); 1876 error = ENOPROTOOPT; 1877 break; 1878 } 1879 KASSERT(solocked(so)); 1880 return error; 1881 } 1882 1883 int 1884 sosetopt(struct socket *so, struct sockopt *sopt) 1885 { 1886 int error, prerr; 1887 1888 if (sopt->sopt_level == SOL_SOCKET) { 1889 error = sosetopt1(so, sopt); 1890 KASSERT(solocked(so)); 1891 } else { 1892 error = ENOPROTOOPT; 1893 solock(so); 1894 } 1895 1896 if ((error == 0 || error == ENOPROTOOPT) && 1897 so->so_proto != NULL && so->so_proto->pr_ctloutput != NULL) { 1898 /* give the protocol stack a shot */ 1899 prerr = (*so->so_proto->pr_ctloutput)(PRCO_SETOPT, so, sopt); 1900 if (prerr == 0) 1901 error = 0; 1902 else if (prerr != ENOPROTOOPT) 1903 error = prerr; 1904 } 1905 sounlock(so); 1906 return error; 1907 } 1908 1909 /* 1910 * so_setsockopt() is a wrapper providing a sockopt structure for sosetopt() 1911 */ 1912 int 1913 so_setsockopt(struct lwp *l, struct socket *so, int level, int name, 1914 const void *val, size_t valsize) 1915 { 1916 struct sockopt sopt; 1917 int error; 1918 1919 KASSERT(valsize == 0 || val != NULL); 1920 1921 sockopt_init(&sopt, level, name, valsize); 1922 sockopt_set(&sopt, val, valsize); 1923 1924 error = sosetopt(so, &sopt); 1925 1926 sockopt_destroy(&sopt); 1927 1928 return error; 1929 } 1930 1931 /* 1932 * internal get SOL_SOCKET options 1933 */ 1934 static int 1935 sogetopt1(struct socket *so, struct sockopt *sopt) 1936 { 1937 int error, optval, opt; 1938 struct linger l; 1939 struct timeval tv; 1940 1941 switch ((opt = sopt->sopt_name)) { 1942 1943 case SO_ACCEPTFILTER: 1944 error = accept_filt_getopt(so, sopt); 1945 break; 1946 1947 case SO_LINGER: 1948 l.l_onoff = (so->so_options & SO_LINGER) ? 1 : 0; 1949 l.l_linger = so->so_linger; 1950 1951 error = sockopt_set(sopt, &l, sizeof(l)); 1952 break; 1953 1954 case SO_USELOOPBACK: 1955 case SO_DONTROUTE: 1956 case SO_DEBUG: 1957 case SO_KEEPALIVE: 1958 case SO_REUSEADDR: 1959 case SO_REUSEPORT: 1960 case SO_BROADCAST: 1961 case SO_OOBINLINE: 1962 case SO_TIMESTAMP: 1963 case SO_NOSIGPIPE: 1964 case SO_RERROR: 1965 #ifdef SO_OTIMESTAMP 1966 case SO_OTIMESTAMP: 1967 #endif 1968 case SO_ACCEPTCONN: 1969 error = sockopt_setint(sopt, (so->so_options & opt) ? 1 : 0); 1970 break; 1971 1972 case SO_TYPE: 1973 error = sockopt_setint(sopt, so->so_type); 1974 break; 1975 1976 case SO_ERROR: 1977 if (so->so_error == 0) { 1978 so->so_error = so->so_rerror; 1979 so->so_rerror = 0; 1980 } 1981 error = sockopt_setint(sopt, so->so_error); 1982 so->so_error = 0; 1983 break; 1984 1985 case SO_SNDBUF: 1986 error = sockopt_setint(sopt, so->so_snd.sb_hiwat); 1987 break; 1988 1989 case SO_RCVBUF: 1990 error = sockopt_setint(sopt, so->so_rcv.sb_hiwat); 1991 break; 1992 1993 case SO_SNDLOWAT: 1994 error = sockopt_setint(sopt, so->so_snd.sb_lowat); 1995 break; 1996 1997 case SO_RCVLOWAT: 1998 error = sockopt_setint(sopt, so->so_rcv.sb_lowat); 1999 break; 2000 2001 #ifdef COMPAT_50 2002 case SO_OSNDTIMEO: 2003 case SO_ORCVTIMEO: { 2004 struct timeval50 otv; 2005 2006 optval = (opt == SO_OSNDTIMEO ? 2007 so->so_snd.sb_timeo : so->so_rcv.sb_timeo); 2008 2009 otv.tv_sec = optval / hz; 2010 otv.tv_usec = (optval % hz) * tick; 2011 2012 error = sockopt_set(sopt, &otv, sizeof(otv)); 2013 break; 2014 } 2015 #endif /* COMPAT_50 */ 2016 2017 case SO_SNDTIMEO: 2018 case SO_RCVTIMEO: 2019 optval = (opt == SO_SNDTIMEO ? 2020 so->so_snd.sb_timeo : so->so_rcv.sb_timeo); 2021 2022 tv.tv_sec = optval / hz; 2023 tv.tv_usec = (optval % hz) * tick; 2024 2025 error = sockopt_set(sopt, &tv, sizeof(tv)); 2026 break; 2027 2028 case SO_OVERFLOWED: 2029 error = sockopt_setint(sopt, so->so_rcv.sb_overflowed); 2030 break; 2031 2032 default: 2033 error = ENOPROTOOPT; 2034 break; 2035 } 2036 2037 return (error); 2038 } 2039 2040 int 2041 sogetopt(struct socket *so, struct sockopt *sopt) 2042 { 2043 int error; 2044 2045 solock(so); 2046 if (sopt->sopt_level != SOL_SOCKET) { 2047 if (so->so_proto && so->so_proto->pr_ctloutput) { 2048 error = ((*so->so_proto->pr_ctloutput) 2049 (PRCO_GETOPT, so, sopt)); 2050 } else 2051 error = (ENOPROTOOPT); 2052 } else { 2053 error = sogetopt1(so, sopt); 2054 } 2055 sounlock(so); 2056 return (error); 2057 } 2058 2059 /* 2060 * alloc sockopt data buffer buffer 2061 * - will be released at destroy 2062 */ 2063 static int 2064 sockopt_alloc(struct sockopt *sopt, size_t len, km_flag_t kmflag) 2065 { 2066 2067 KASSERT(sopt->sopt_size == 0); 2068 2069 if (len > sizeof(sopt->sopt_buf)) { 2070 sopt->sopt_data = kmem_zalloc(len, kmflag); 2071 if (sopt->sopt_data == NULL) 2072 return ENOMEM; 2073 } else 2074 sopt->sopt_data = sopt->sopt_buf; 2075 2076 sopt->sopt_size = len; 2077 return 0; 2078 } 2079 2080 /* 2081 * initialise sockopt storage 2082 * - MAY sleep during allocation 2083 */ 2084 void 2085 sockopt_init(struct sockopt *sopt, int level, int name, size_t size) 2086 { 2087 2088 memset(sopt, 0, sizeof(*sopt)); 2089 2090 sopt->sopt_level = level; 2091 sopt->sopt_name = name; 2092 (void)sockopt_alloc(sopt, size, KM_SLEEP); 2093 } 2094 2095 /* 2096 * destroy sockopt storage 2097 * - will release any held memory references 2098 */ 2099 void 2100 sockopt_destroy(struct sockopt *sopt) 2101 { 2102 2103 if (sopt->sopt_data != sopt->sopt_buf) 2104 kmem_free(sopt->sopt_data, sopt->sopt_size); 2105 2106 memset(sopt, 0, sizeof(*sopt)); 2107 } 2108 2109 /* 2110 * set sockopt value 2111 * - value is copied into sockopt 2112 * - memory is allocated when necessary, will not sleep 2113 */ 2114 int 2115 sockopt_set(struct sockopt *sopt, const void *buf, size_t len) 2116 { 2117 int error; 2118 2119 if (sopt->sopt_size == 0) { 2120 error = sockopt_alloc(sopt, len, KM_NOSLEEP); 2121 if (error) 2122 return error; 2123 } 2124 2125 if (sopt->sopt_size < len) 2126 return EINVAL; 2127 2128 memcpy(sopt->sopt_data, buf, len); 2129 sopt->sopt_retsize = len; 2130 2131 return 0; 2132 } 2133 2134 /* 2135 * common case of set sockopt integer value 2136 */ 2137 int 2138 sockopt_setint(struct sockopt *sopt, int val) 2139 { 2140 2141 return sockopt_set(sopt, &val, sizeof(int)); 2142 } 2143 2144 /* 2145 * get sockopt value 2146 * - correct size must be given 2147 */ 2148 int 2149 sockopt_get(const struct sockopt *sopt, void *buf, size_t len) 2150 { 2151 2152 if (sopt->sopt_size != len) 2153 return EINVAL; 2154 2155 memcpy(buf, sopt->sopt_data, len); 2156 return 0; 2157 } 2158 2159 /* 2160 * common case of get sockopt integer value 2161 */ 2162 int 2163 sockopt_getint(const struct sockopt *sopt, int *valp) 2164 { 2165 2166 return sockopt_get(sopt, valp, sizeof(int)); 2167 } 2168 2169 /* 2170 * set sockopt value from mbuf 2171 * - ONLY for legacy code 2172 * - mbuf is released by sockopt 2173 * - will not sleep 2174 */ 2175 int 2176 sockopt_setmbuf(struct sockopt *sopt, struct mbuf *m) 2177 { 2178 size_t len; 2179 int error; 2180 2181 len = m_length(m); 2182 2183 if (sopt->sopt_size == 0) { 2184 error = sockopt_alloc(sopt, len, KM_NOSLEEP); 2185 if (error) 2186 return error; 2187 } 2188 2189 if (sopt->sopt_size < len) 2190 return EINVAL; 2191 2192 m_copydata(m, 0, len, sopt->sopt_data); 2193 m_freem(m); 2194 sopt->sopt_retsize = len; 2195 2196 return 0; 2197 } 2198 2199 /* 2200 * get sockopt value into mbuf 2201 * - ONLY for legacy code 2202 * - mbuf to be released by the caller 2203 * - will not sleep 2204 */ 2205 struct mbuf * 2206 sockopt_getmbuf(const struct sockopt *sopt) 2207 { 2208 struct mbuf *m; 2209 2210 if (sopt->sopt_size > MCLBYTES) 2211 return NULL; 2212 2213 m = m_get(M_DONTWAIT, MT_SOOPTS); 2214 if (m == NULL) 2215 return NULL; 2216 2217 if (sopt->sopt_size > MLEN) { 2218 MCLGET(m, M_DONTWAIT); 2219 if ((m->m_flags & M_EXT) == 0) { 2220 m_free(m); 2221 return NULL; 2222 } 2223 } 2224 2225 memcpy(mtod(m, void *), sopt->sopt_data, sopt->sopt_size); 2226 m->m_len = sopt->sopt_size; 2227 2228 return m; 2229 } 2230 2231 void 2232 sohasoutofband(struct socket *so) 2233 { 2234 2235 fownsignal(so->so_pgid, SIGURG, POLL_PRI, POLLPRI|POLLRDBAND, so); 2236 selnotify(&so->so_rcv.sb_sel, POLLPRI | POLLRDBAND, NOTE_SUBMIT); 2237 } 2238 2239 static void 2240 filt_sordetach(struct knote *kn) 2241 { 2242 struct socket *so; 2243 2244 so = ((file_t *)kn->kn_obj)->f_socket; 2245 solock(so); 2246 SLIST_REMOVE(&so->so_rcv.sb_sel.sel_klist, kn, knote, kn_selnext); 2247 if (SLIST_EMPTY(&so->so_rcv.sb_sel.sel_klist)) 2248 so->so_rcv.sb_flags &= ~SB_KNOTE; 2249 sounlock(so); 2250 } 2251 2252 /*ARGSUSED*/ 2253 static int 2254 filt_soread(struct knote *kn, long hint) 2255 { 2256 struct socket *so; 2257 int rv; 2258 2259 so = ((file_t *)kn->kn_obj)->f_socket; 2260 if (hint != NOTE_SUBMIT) 2261 solock(so); 2262 kn->kn_data = so->so_rcv.sb_cc; 2263 if (so->so_state & SS_CANTRCVMORE) { 2264 kn->kn_flags |= EV_EOF; 2265 kn->kn_fflags = so->so_error; 2266 rv = 1; 2267 } else if (so->so_error || so->so_rerror) 2268 rv = 1; 2269 else if (kn->kn_sfflags & NOTE_LOWAT) 2270 rv = (kn->kn_data >= kn->kn_sdata); 2271 else 2272 rv = (kn->kn_data >= so->so_rcv.sb_lowat); 2273 if (hint != NOTE_SUBMIT) 2274 sounlock(so); 2275 return rv; 2276 } 2277 2278 static void 2279 filt_sowdetach(struct knote *kn) 2280 { 2281 struct socket *so; 2282 2283 so = ((file_t *)kn->kn_obj)->f_socket; 2284 solock(so); 2285 SLIST_REMOVE(&so->so_snd.sb_sel.sel_klist, kn, knote, kn_selnext); 2286 if (SLIST_EMPTY(&so->so_snd.sb_sel.sel_klist)) 2287 so->so_snd.sb_flags &= ~SB_KNOTE; 2288 sounlock(so); 2289 } 2290 2291 /*ARGSUSED*/ 2292 static int 2293 filt_sowrite(struct knote *kn, long hint) 2294 { 2295 struct socket *so; 2296 int rv; 2297 2298 so = ((file_t *)kn->kn_obj)->f_socket; 2299 if (hint != NOTE_SUBMIT) 2300 solock(so); 2301 kn->kn_data = sbspace(&so->so_snd); 2302 if (so->so_state & SS_CANTSENDMORE) { 2303 kn->kn_flags |= EV_EOF; 2304 kn->kn_fflags = so->so_error; 2305 rv = 1; 2306 } else if (so->so_error) 2307 rv = 1; 2308 else if (((so->so_state & SS_ISCONNECTED) == 0) && 2309 (so->so_proto->pr_flags & PR_CONNREQUIRED)) 2310 rv = 0; 2311 else if (kn->kn_sfflags & NOTE_LOWAT) 2312 rv = (kn->kn_data >= kn->kn_sdata); 2313 else 2314 rv = (kn->kn_data >= so->so_snd.sb_lowat); 2315 if (hint != NOTE_SUBMIT) 2316 sounlock(so); 2317 return rv; 2318 } 2319 2320 /*ARGSUSED*/ 2321 static int 2322 filt_solisten(struct knote *kn, long hint) 2323 { 2324 struct socket *so; 2325 int rv; 2326 2327 so = ((file_t *)kn->kn_obj)->f_socket; 2328 2329 /* 2330 * Set kn_data to number of incoming connections, not 2331 * counting partial (incomplete) connections. 2332 */ 2333 if (hint != NOTE_SUBMIT) 2334 solock(so); 2335 kn->kn_data = so->so_qlen; 2336 rv = (kn->kn_data > 0); 2337 if (hint != NOTE_SUBMIT) 2338 sounlock(so); 2339 return rv; 2340 } 2341 2342 static const struct filterops solisten_filtops = { 2343 .f_isfd = 1, 2344 .f_attach = NULL, 2345 .f_detach = filt_sordetach, 2346 .f_event = filt_solisten, 2347 }; 2348 2349 static const struct filterops soread_filtops = { 2350 .f_isfd = 1, 2351 .f_attach = NULL, 2352 .f_detach = filt_sordetach, 2353 .f_event = filt_soread, 2354 }; 2355 2356 static const struct filterops sowrite_filtops = { 2357 .f_isfd = 1, 2358 .f_attach = NULL, 2359 .f_detach = filt_sowdetach, 2360 .f_event = filt_sowrite, 2361 }; 2362 2363 int 2364 soo_kqfilter(struct file *fp, struct knote *kn) 2365 { 2366 struct socket *so; 2367 struct sockbuf *sb; 2368 2369 so = ((file_t *)kn->kn_obj)->f_socket; 2370 solock(so); 2371 switch (kn->kn_filter) { 2372 case EVFILT_READ: 2373 if (so->so_options & SO_ACCEPTCONN) 2374 kn->kn_fop = &solisten_filtops; 2375 else 2376 kn->kn_fop = &soread_filtops; 2377 sb = &so->so_rcv; 2378 break; 2379 case EVFILT_WRITE: 2380 kn->kn_fop = &sowrite_filtops; 2381 sb = &so->so_snd; 2382 break; 2383 default: 2384 sounlock(so); 2385 return (EINVAL); 2386 } 2387 SLIST_INSERT_HEAD(&sb->sb_sel.sel_klist, kn, kn_selnext); 2388 sb->sb_flags |= SB_KNOTE; 2389 sounlock(so); 2390 return (0); 2391 } 2392 2393 static int 2394 sodopoll(struct socket *so, int events) 2395 { 2396 int revents; 2397 2398 revents = 0; 2399 2400 if (events & (POLLIN | POLLRDNORM)) 2401 if (soreadable(so)) 2402 revents |= events & (POLLIN | POLLRDNORM); 2403 2404 if (events & (POLLOUT | POLLWRNORM)) 2405 if (sowritable(so)) 2406 revents |= events & (POLLOUT | POLLWRNORM); 2407 2408 if (events & (POLLPRI | POLLRDBAND)) 2409 if (so->so_oobmark || (so->so_state & SS_RCVATMARK)) 2410 revents |= events & (POLLPRI | POLLRDBAND); 2411 2412 return revents; 2413 } 2414 2415 int 2416 sopoll(struct socket *so, int events) 2417 { 2418 int revents = 0; 2419 2420 #ifndef DIAGNOSTIC 2421 /* 2422 * Do a quick, unlocked check in expectation that the socket 2423 * will be ready for I/O. Don't do this check if DIAGNOSTIC, 2424 * as the solocked() assertions will fail. 2425 */ 2426 if ((revents = sodopoll(so, events)) != 0) 2427 return revents; 2428 #endif 2429 2430 solock(so); 2431 if ((revents = sodopoll(so, events)) == 0) { 2432 if (events & (POLLIN | POLLPRI | POLLRDNORM | POLLRDBAND)) { 2433 selrecord(curlwp, &so->so_rcv.sb_sel); 2434 so->so_rcv.sb_flags |= SB_NOTIFY; 2435 } 2436 2437 if (events & (POLLOUT | POLLWRNORM)) { 2438 selrecord(curlwp, &so->so_snd.sb_sel); 2439 so->so_snd.sb_flags |= SB_NOTIFY; 2440 } 2441 } 2442 sounlock(so); 2443 2444 return revents; 2445 } 2446 2447 struct mbuf ** 2448 sbsavetimestamp(int opt, struct mbuf **mp) 2449 { 2450 struct timeval tv; 2451 microtime(&tv); 2452 2453 #ifdef SO_OTIMESTAMP 2454 if (opt & SO_OTIMESTAMP) { 2455 struct timeval50 tv50; 2456 2457 timeval_to_timeval50(&tv, &tv50); 2458 *mp = sbcreatecontrol(&tv50, sizeof(tv50), 2459 SCM_OTIMESTAMP, SOL_SOCKET); 2460 if (*mp) 2461 mp = &(*mp)->m_next; 2462 } else 2463 #endif 2464 2465 if (opt & SO_TIMESTAMP) { 2466 *mp = sbcreatecontrol(&tv, sizeof(tv), 2467 SCM_TIMESTAMP, SOL_SOCKET); 2468 if (*mp) 2469 mp = &(*mp)->m_next; 2470 } 2471 return mp; 2472 } 2473 2474 2475 #include <sys/sysctl.h> 2476 2477 static int sysctl_kern_somaxkva(SYSCTLFN_PROTO); 2478 static int sysctl_kern_sbmax(SYSCTLFN_PROTO); 2479 2480 /* 2481 * sysctl helper routine for kern.somaxkva. ensures that the given 2482 * value is not too small. 2483 * (XXX should we maybe make sure it's not too large as well?) 2484 */ 2485 static int 2486 sysctl_kern_somaxkva(SYSCTLFN_ARGS) 2487 { 2488 int error, new_somaxkva; 2489 struct sysctlnode node; 2490 2491 new_somaxkva = somaxkva; 2492 node = *rnode; 2493 node.sysctl_data = &new_somaxkva; 2494 error = sysctl_lookup(SYSCTLFN_CALL(&node)); 2495 if (error || newp == NULL) 2496 return (error); 2497 2498 if (new_somaxkva < (16 * 1024 * 1024)) /* sanity */ 2499 return (EINVAL); 2500 2501 mutex_enter(&so_pendfree_lock); 2502 somaxkva = new_somaxkva; 2503 cv_broadcast(&socurkva_cv); 2504 mutex_exit(&so_pendfree_lock); 2505 2506 return (error); 2507 } 2508 2509 /* 2510 * sysctl helper routine for kern.sbmax. Basically just ensures that 2511 * any new value is not too small. 2512 */ 2513 static int 2514 sysctl_kern_sbmax(SYSCTLFN_ARGS) 2515 { 2516 int error, new_sbmax; 2517 struct sysctlnode node; 2518 2519 new_sbmax = sb_max; 2520 node = *rnode; 2521 node.sysctl_data = &new_sbmax; 2522 error = sysctl_lookup(SYSCTLFN_CALL(&node)); 2523 if (error || newp == NULL) 2524 return (error); 2525 2526 KERNEL_LOCK(1, NULL); 2527 error = sb_max_set(new_sbmax); 2528 KERNEL_UNLOCK_ONE(NULL); 2529 2530 return (error); 2531 } 2532 2533 /* 2534 * sysctl helper routine for kern.sooptions. Ensures that only allowed 2535 * options can be set. 2536 */ 2537 static int 2538 sysctl_kern_sooptions(SYSCTLFN_ARGS) 2539 { 2540 int error, new_options; 2541 struct sysctlnode node; 2542 2543 new_options = sooptions; 2544 node = *rnode; 2545 node.sysctl_data = &new_options; 2546 error = sysctl_lookup(SYSCTLFN_CALL(&node)); 2547 if (error || newp == NULL) 2548 return error; 2549 2550 if (new_options & ~SO_DEFOPTS) 2551 return EINVAL; 2552 2553 sooptions = new_options; 2554 2555 return 0; 2556 } 2557 2558 static void 2559 sysctl_kern_socket_setup(void) 2560 { 2561 2562 KASSERT(socket_sysctllog == NULL); 2563 2564 sysctl_createv(&socket_sysctllog, 0, NULL, NULL, 2565 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2566 CTLTYPE_INT, "somaxkva", 2567 SYSCTL_DESCR("Maximum amount of kernel memory to be " 2568 "used for socket buffers"), 2569 sysctl_kern_somaxkva, 0, NULL, 0, 2570 CTL_KERN, KERN_SOMAXKVA, CTL_EOL); 2571 2572 sysctl_createv(&socket_sysctllog, 0, NULL, NULL, 2573 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2574 CTLTYPE_INT, "sbmax", 2575 SYSCTL_DESCR("Maximum socket buffer size"), 2576 sysctl_kern_sbmax, 0, NULL, 0, 2577 CTL_KERN, KERN_SBMAX, CTL_EOL); 2578 2579 sysctl_createv(&socket_sysctllog, 0, NULL, NULL, 2580 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2581 CTLTYPE_INT, "sooptions", 2582 SYSCTL_DESCR("Default socket options"), 2583 sysctl_kern_sooptions, 0, NULL, 0, 2584 CTL_KERN, CTL_CREATE, CTL_EOL); 2585 } 2586
Indexes created Sat Sep 20 22:09:52 GMT 2025