1 /* $NetBSD: uipc_socket.c,v 1.213 2013/02/14 21:57:58 christos 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 #include <sys/cdefs.h> 66 __KERNEL_RCSID(0, "$NetBSD: uipc_socket.c,v 1.213 2013/02/14 21:57:58 christos Exp $"); 67 68 #include "opt_compat_netbsd.h" 69 #include "opt_sock_counters.h" 70 #include "opt_sosend_loan.h" 71 #include "opt_mbuftrace.h" 72 #include "opt_somaxkva.h" 73 #include "opt_multiprocessor.h" /* XXX */ 74 75 #include <sys/param.h> 76 #include <sys/systm.h> 77 #include <sys/proc.h> 78 #include <sys/file.h> 79 #include <sys/filedesc.h> 80 #include <sys/kmem.h> 81 #include <sys/mbuf.h> 82 #include <sys/domain.h> 83 #include <sys/kernel.h> 84 #include <sys/protosw.h> 85 #include <sys/socket.h> 86 #include <sys/socketvar.h> 87 #include <sys/signalvar.h> 88 #include <sys/resourcevar.h> 89 #include <sys/uidinfo.h> 90 #include <sys/event.h> 91 #include <sys/poll.h> 92 #include <sys/kauth.h> 93 #include <sys/mutex.h> 94 #include <sys/condvar.h> 95 #include <sys/kthread.h> 96 97 #ifdef COMPAT_50 98 #include <compat/sys/time.h> 99 #include <compat/sys/socket.h> 100 #endif 101 102 #include <uvm/uvm_extern.h> 103 #include <uvm/uvm_loan.h> 104 #include <uvm/uvm_page.h> 105 106 MALLOC_DEFINE(M_SOOPTS, "soopts", "socket options"); 107 MALLOC_DEFINE(M_SONAME, "soname", "socket name"); 108 109 extern const struct fileops socketops; 110 111 extern int somaxconn; /* patchable (XXX sysctl) */ 112 int somaxconn = SOMAXCONN; 113 kmutex_t *softnet_lock; 114 115 #ifdef SOSEND_COUNTERS 116 #include <sys/device.h> 117 118 static struct evcnt sosend_loan_big = EVCNT_INITIALIZER(EVCNT_TYPE_MISC, 119 NULL, "sosend", "loan big"); 120 static struct evcnt sosend_copy_big = EVCNT_INITIALIZER(EVCNT_TYPE_MISC, 121 NULL, "sosend", "copy big"); 122 static struct evcnt sosend_copy_small = EVCNT_INITIALIZER(EVCNT_TYPE_MISC, 123 NULL, "sosend", "copy small"); 124 static struct evcnt sosend_kvalimit = EVCNT_INITIALIZER(EVCNT_TYPE_MISC, 125 NULL, "sosend", "kva limit"); 126 127 #define SOSEND_COUNTER_INCR(ev) (ev)->ev_count++ 128 129 EVCNT_ATTACH_STATIC(sosend_loan_big); 130 EVCNT_ATTACH_STATIC(sosend_copy_big); 131 EVCNT_ATTACH_STATIC(sosend_copy_small); 132 EVCNT_ATTACH_STATIC(sosend_kvalimit); 133 #else 134 135 #define SOSEND_COUNTER_INCR(ev) /* nothing */ 136 137 #endif /* SOSEND_COUNTERS */ 138 139 #if defined(SOSEND_NO_LOAN) || defined(MULTIPROCESSOR) 140 int sock_loan_thresh = -1; 141 #else 142 int sock_loan_thresh = 4096; 143 #endif 144 145 static kmutex_t so_pendfree_lock; 146 static struct mbuf *so_pendfree = NULL; 147 148 #ifndef SOMAXKVA 149 #define SOMAXKVA (16 * 1024 * 1024) 150 #endif 151 int somaxkva = SOMAXKVA; 152 static int socurkva; 153 static kcondvar_t socurkva_cv; 154 155 static kauth_listener_t socket_listener; 156 157 #define SOCK_LOAN_CHUNK 65536 158 159 static void sopendfree_thread(void *); 160 static kcondvar_t pendfree_thread_cv; 161 static lwp_t *sopendfree_lwp; 162 163 static void sysctl_kern_socket_setup(void); 164 static struct sysctllog *socket_sysctllog; 165 166 static vsize_t 167 sokvareserve(struct socket *so, vsize_t len) 168 { 169 int error; 170 171 mutex_enter(&so_pendfree_lock); 172 while (socurkva + len > somaxkva) { 173 SOSEND_COUNTER_INCR(&sosend_kvalimit); 174 error = cv_wait_sig(&socurkva_cv, &so_pendfree_lock); 175 if (error) { 176 len = 0; 177 break; 178 } 179 } 180 socurkva += len; 181 mutex_exit(&so_pendfree_lock); 182 return len; 183 } 184 185 static void 186 sokvaunreserve(vsize_t len) 187 { 188 189 mutex_enter(&so_pendfree_lock); 190 socurkva -= len; 191 cv_broadcast(&socurkva_cv); 192 mutex_exit(&so_pendfree_lock); 193 } 194 195 /* 196 * sokvaalloc: allocate kva for loan. 197 */ 198 199 vaddr_t 200 sokvaalloc(vaddr_t sva, vsize_t len, struct socket *so) 201 { 202 vaddr_t lva; 203 204 /* 205 * reserve kva. 206 */ 207 208 if (sokvareserve(so, len) == 0) 209 return 0; 210 211 /* 212 * allocate kva. 213 */ 214 215 lva = uvm_km_alloc(kernel_map, len, atop(sva) & uvmexp.colormask, 216 UVM_KMF_COLORMATCH | UVM_KMF_VAONLY | UVM_KMF_WAITVA); 217 if (lva == 0) { 218 sokvaunreserve(len); 219 return (0); 220 } 221 222 return lva; 223 } 224 225 /* 226 * sokvafree: free kva for loan. 227 */ 228 229 void 230 sokvafree(vaddr_t sva, vsize_t len) 231 { 232 233 /* 234 * free kva. 235 */ 236 237 uvm_km_free(kernel_map, sva, len, UVM_KMF_VAONLY); 238 239 /* 240 * unreserve kva. 241 */ 242 243 sokvaunreserve(len); 244 } 245 246 static void 247 sodoloanfree(struct vm_page **pgs, void *buf, size_t size) 248 { 249 vaddr_t sva, eva; 250 vsize_t len; 251 int npgs; 252 253 KASSERT(pgs != NULL); 254 255 eva = round_page((vaddr_t) buf + size); 256 sva = trunc_page((vaddr_t) buf); 257 len = eva - sva; 258 npgs = len >> PAGE_SHIFT; 259 260 pmap_kremove(sva, len); 261 pmap_update(pmap_kernel()); 262 uvm_unloan(pgs, npgs, UVM_LOAN_TOPAGE); 263 sokvafree(sva, len); 264 } 265 266 /* 267 * sopendfree_thread: free mbufs on "pendfree" list. 268 * unlock and relock so_pendfree_lock when freeing mbufs. 269 */ 270 271 static void 272 sopendfree_thread(void *v) 273 { 274 struct mbuf *m, *next; 275 size_t rv; 276 277 mutex_enter(&so_pendfree_lock); 278 279 for (;;) { 280 rv = 0; 281 while (so_pendfree != NULL) { 282 m = so_pendfree; 283 so_pendfree = NULL; 284 mutex_exit(&so_pendfree_lock); 285 286 for (; m != NULL; m = next) { 287 next = m->m_next; 288 KASSERT((~m->m_flags & (M_EXT|M_EXT_PAGES)) == 0); 289 KASSERT(m->m_ext.ext_refcnt == 0); 290 291 rv += m->m_ext.ext_size; 292 sodoloanfree(m->m_ext.ext_pgs, m->m_ext.ext_buf, 293 m->m_ext.ext_size); 294 pool_cache_put(mb_cache, m); 295 } 296 297 mutex_enter(&so_pendfree_lock); 298 } 299 if (rv) 300 cv_broadcast(&socurkva_cv); 301 cv_wait(&pendfree_thread_cv, &so_pendfree_lock); 302 } 303 panic("sopendfree_thread"); 304 /* NOTREACHED */ 305 } 306 307 void 308 soloanfree(struct mbuf *m, void *buf, size_t size, void *arg) 309 { 310 311 KASSERT(m != NULL); 312 313 /* 314 * postpone freeing mbuf. 315 * 316 * we can't do it in interrupt context 317 * because we need to put kva back to kernel_map. 318 */ 319 320 mutex_enter(&so_pendfree_lock); 321 m->m_next = so_pendfree; 322 so_pendfree = m; 323 cv_signal(&pendfree_thread_cv); 324 mutex_exit(&so_pendfree_lock); 325 } 326 327 static long 328 sosend_loan(struct socket *so, struct uio *uio, struct mbuf *m, long space) 329 { 330 struct iovec *iov = uio->uio_iov; 331 vaddr_t sva, eva; 332 vsize_t len; 333 vaddr_t lva; 334 int npgs, error; 335 vaddr_t va; 336 int i; 337 338 if (VMSPACE_IS_KERNEL_P(uio->uio_vmspace)) 339 return (0); 340 341 if (iov->iov_len < (size_t) space) 342 space = iov->iov_len; 343 if (space > SOCK_LOAN_CHUNK) 344 space = SOCK_LOAN_CHUNK; 345 346 eva = round_page((vaddr_t) iov->iov_base + space); 347 sva = trunc_page((vaddr_t) iov->iov_base); 348 len = eva - sva; 349 npgs = len >> PAGE_SHIFT; 350 351 KASSERT(npgs <= M_EXT_MAXPAGES); 352 353 lva = sokvaalloc(sva, len, so); 354 if (lva == 0) 355 return 0; 356 357 error = uvm_loan(&uio->uio_vmspace->vm_map, sva, len, 358 m->m_ext.ext_pgs, UVM_LOAN_TOPAGE); 359 if (error) { 360 sokvafree(lva, len); 361 return (0); 362 } 363 364 for (i = 0, va = lva; i < npgs; i++, va += PAGE_SIZE) 365 pmap_kenter_pa(va, VM_PAGE_TO_PHYS(m->m_ext.ext_pgs[i]), 366 VM_PROT_READ, 0); 367 pmap_update(pmap_kernel()); 368 369 lva += (vaddr_t) iov->iov_base & PAGE_MASK; 370 371 MEXTADD(m, (void *) lva, space, M_MBUF, soloanfree, so); 372 m->m_flags |= M_EXT_PAGES | M_EXT_ROMAP; 373 374 uio->uio_resid -= space; 375 /* uio_offset not updated, not set/used for write(2) */ 376 uio->uio_iov->iov_base = (char *)uio->uio_iov->iov_base + space; 377 uio->uio_iov->iov_len -= space; 378 if (uio->uio_iov->iov_len == 0) { 379 uio->uio_iov++; 380 uio->uio_iovcnt--; 381 } 382 383 return (space); 384 } 385 386 struct mbuf * 387 getsombuf(struct socket *so, int type) 388 { 389 struct mbuf *m; 390 391 m = m_get(M_WAIT, type); 392 MCLAIM(m, so->so_mowner); 393 return m; 394 } 395 396 static int 397 socket_listener_cb(kauth_cred_t cred, kauth_action_t action, void *cookie, 398 void *arg0, void *arg1, void *arg2, void *arg3) 399 { 400 int result; 401 enum kauth_network_req req; 402 403 result = KAUTH_RESULT_DEFER; 404 req = (enum kauth_network_req)arg0; 405 406 if ((action != KAUTH_NETWORK_SOCKET) && 407 (action != KAUTH_NETWORK_BIND)) 408 return result; 409 410 switch (req) { 411 case KAUTH_REQ_NETWORK_BIND_PORT: 412 result = KAUTH_RESULT_ALLOW; 413 break; 414 415 case KAUTH_REQ_NETWORK_SOCKET_DROP: { 416 /* Normal users can only drop their own connections. */ 417 struct socket *so = (struct socket *)arg1; 418 419 if (proc_uidmatch(cred, so->so_cred)) 420 result = KAUTH_RESULT_ALLOW; 421 422 break; 423 } 424 425 case KAUTH_REQ_NETWORK_SOCKET_OPEN: 426 /* We allow "raw" routing/bluetooth sockets to anyone. */ 427 if ((u_long)arg1 == PF_ROUTE || (u_long)arg1 == PF_OROUTE 428 || (u_long)arg1 == PF_BLUETOOTH) { 429 result = KAUTH_RESULT_ALLOW; 430 } else { 431 /* Privileged, let secmodel handle this. */ 432 if ((u_long)arg2 == SOCK_RAW) 433 break; 434 } 435 436 result = KAUTH_RESULT_ALLOW; 437 438 break; 439 440 case KAUTH_REQ_NETWORK_SOCKET_CANSEE: 441 result = KAUTH_RESULT_ALLOW; 442 443 break; 444 445 default: 446 break; 447 } 448 449 return result; 450 } 451 452 void 453 soinit(void) 454 { 455 456 sysctl_kern_socket_setup(); 457 458 mutex_init(&so_pendfree_lock, MUTEX_DEFAULT, IPL_VM); 459 softnet_lock = mutex_obj_alloc(MUTEX_DEFAULT, IPL_NONE); 460 cv_init(&socurkva_cv, "sokva"); 461 cv_init(&pendfree_thread_cv, "sopendfr"); 462 soinit2(); 463 464 /* Set the initial adjusted socket buffer size. */ 465 if (sb_max_set(sb_max)) 466 panic("bad initial sb_max value: %lu", sb_max); 467 468 socket_listener = kauth_listen_scope(KAUTH_SCOPE_NETWORK, 469 socket_listener_cb, NULL); 470 } 471 472 void 473 soinit1(void) 474 { 475 int error = kthread_create(PRI_NONE, KTHREAD_MPSAFE, NULL, 476 sopendfree_thread, NULL, &sopendfree_lwp, "sopendfree"); 477 if (error) 478 panic("soinit1 %d", error); 479 } 480 481 /* 482 * Socket operation routines. 483 * These routines are called by the routines in 484 * sys_socket.c or from a system process, and 485 * implement the semantics of socket operations by 486 * switching out to the protocol specific routines. 487 */ 488 /*ARGSUSED*/ 489 int 490 socreate(int dom, struct socket **aso, int type, int proto, struct lwp *l, 491 struct socket *lockso) 492 { 493 const struct protosw *prp; 494 struct socket *so; 495 uid_t uid; 496 int error; 497 kmutex_t *lock; 498 499 error = kauth_authorize_network(l->l_cred, KAUTH_NETWORK_SOCKET, 500 KAUTH_REQ_NETWORK_SOCKET_OPEN, KAUTH_ARG(dom), KAUTH_ARG(type), 501 KAUTH_ARG(proto)); 502 if (error != 0) 503 return error; 504 505 if (proto) 506 prp = pffindproto(dom, proto, type); 507 else 508 prp = pffindtype(dom, type); 509 if (prp == NULL) { 510 /* no support for domain */ 511 if (pffinddomain(dom) == 0) 512 return EAFNOSUPPORT; 513 /* no support for socket type */ 514 if (proto == 0 && type != 0) 515 return EPROTOTYPE; 516 return EPROTONOSUPPORT; 517 } 518 if (prp->pr_usrreq == NULL) 519 return EPROTONOSUPPORT; 520 if (prp->pr_type != type) 521 return EPROTOTYPE; 522 523 so = soget(true); 524 so->so_type = type; 525 so->so_proto = prp; 526 so->so_send = sosend; 527 so->so_receive = soreceive; 528 #ifdef MBUFTRACE 529 so->so_rcv.sb_mowner = &prp->pr_domain->dom_mowner; 530 so->so_snd.sb_mowner = &prp->pr_domain->dom_mowner; 531 so->so_mowner = &prp->pr_domain->dom_mowner; 532 #endif 533 uid = kauth_cred_geteuid(l->l_cred); 534 so->so_uidinfo = uid_find(uid); 535 so->so_cpid = l->l_proc->p_pid; 536 if (lockso != NULL) { 537 /* Caller wants us to share a lock. */ 538 lock = lockso->so_lock; 539 so->so_lock = lock; 540 mutex_obj_hold(lock); 541 mutex_enter(lock); 542 } else { 543 /* Lock assigned and taken during PRU_ATTACH. */ 544 } 545 error = (*prp->pr_usrreq)(so, PRU_ATTACH, NULL, 546 (struct mbuf *)(long)proto, NULL, l); 547 KASSERT(solocked(so)); 548 if (error != 0) { 549 so->so_state |= SS_NOFDREF; 550 sofree(so); 551 return error; 552 } 553 so->so_cred = kauth_cred_dup(l->l_cred); 554 sounlock(so); 555 *aso = so; 556 return 0; 557 } 558 559 /* On success, write file descriptor to fdout and return zero. On 560 * failure, return non-zero; *fdout will be undefined. 561 */ 562 int 563 fsocreate(int domain, struct socket **sop, int type, int protocol, 564 struct lwp *l, int *fdout) 565 { 566 struct socket *so; 567 struct file *fp; 568 int fd, error; 569 int flags = type & SOCK_FLAGS_MASK; 570 571 type &= ~SOCK_FLAGS_MASK; 572 if ((error = fd_allocfile(&fp, &fd)) != 0) 573 return error; 574 fd_set_exclose(l, fd, (flags & SOCK_CLOEXEC) != 0); 575 fp->f_flag = FREAD|FWRITE|((flags & SOCK_NONBLOCK) ? FNONBLOCK : 0)| 576 ((flags & SOCK_NOSIGPIPE) ? FNOSIGPIPE : 0); 577 fp->f_type = DTYPE_SOCKET; 578 fp->f_ops = &socketops; 579 error = socreate(domain, &so, type, protocol, l, NULL); 580 if (error != 0) { 581 fd_abort(curproc, fp, fd); 582 } else { 583 if (sop != NULL) 584 *sop = so; 585 fp->f_data = so; 586 fd_affix(curproc, fp, fd); 587 *fdout = fd; 588 if (flags & SOCK_NONBLOCK) 589 so->so_state |= SS_NBIO; 590 } 591 return error; 592 } 593 594 int 595 sofamily(const struct socket *so) 596 { 597 const struct protosw *pr; 598 const struct domain *dom; 599 600 if ((pr = so->so_proto) == NULL) 601 return AF_UNSPEC; 602 if ((dom = pr->pr_domain) == NULL) 603 return AF_UNSPEC; 604 return dom->dom_family; 605 } 606 607 int 608 sobind(struct socket *so, struct mbuf *nam, struct lwp *l) 609 { 610 int error; 611 612 solock(so); 613 error = (*so->so_proto->pr_usrreq)(so, PRU_BIND, NULL, nam, NULL, l); 614 sounlock(so); 615 return error; 616 } 617 618 int 619 solisten(struct socket *so, int backlog, struct lwp *l) 620 { 621 int error; 622 623 solock(so); 624 if ((so->so_state & (SS_ISCONNECTED | SS_ISCONNECTING | 625 SS_ISDISCONNECTING)) != 0) { 626 sounlock(so); 627 return (EINVAL); 628 } 629 error = (*so->so_proto->pr_usrreq)(so, PRU_LISTEN, NULL, 630 NULL, NULL, l); 631 if (error != 0) { 632 sounlock(so); 633 return error; 634 } 635 if (TAILQ_EMPTY(&so->so_q)) 636 so->so_options |= SO_ACCEPTCONN; 637 if (backlog < 0) 638 backlog = 0; 639 so->so_qlimit = min(backlog, somaxconn); 640 sounlock(so); 641 return 0; 642 } 643 644 void 645 sofree(struct socket *so) 646 { 647 u_int refs; 648 649 KASSERT(solocked(so)); 650 651 if (so->so_pcb || (so->so_state & SS_NOFDREF) == 0) { 652 sounlock(so); 653 return; 654 } 655 if (so->so_head) { 656 /* 657 * We must not decommission a socket that's on the accept(2) 658 * queue. If we do, then accept(2) may hang after select(2) 659 * indicated that the listening socket was ready. 660 */ 661 if (!soqremque(so, 0)) { 662 sounlock(so); 663 return; 664 } 665 } 666 if (so->so_rcv.sb_hiwat) 667 (void)chgsbsize(so->so_uidinfo, &so->so_rcv.sb_hiwat, 0, 668 RLIM_INFINITY); 669 if (so->so_snd.sb_hiwat) 670 (void)chgsbsize(so->so_uidinfo, &so->so_snd.sb_hiwat, 0, 671 RLIM_INFINITY); 672 sbrelease(&so->so_snd, so); 673 KASSERT(!cv_has_waiters(&so->so_cv)); 674 KASSERT(!cv_has_waiters(&so->so_rcv.sb_cv)); 675 KASSERT(!cv_has_waiters(&so->so_snd.sb_cv)); 676 sorflush(so); 677 refs = so->so_aborting; /* XXX */ 678 /* Remove acccept filter if one is present. */ 679 if (so->so_accf != NULL) 680 (void)accept_filt_clear(so); 681 sounlock(so); 682 if (refs == 0) /* XXX */ 683 soput(so); 684 } 685 686 /* 687 * Close a socket on last file table reference removal. 688 * Initiate disconnect if connected. 689 * Free socket when disconnect complete. 690 */ 691 int 692 soclose(struct socket *so) 693 { 694 struct socket *so2; 695 int error; 696 int error2; 697 698 error = 0; 699 solock(so); 700 if (so->so_options & SO_ACCEPTCONN) { 701 for (;;) { 702 if ((so2 = TAILQ_FIRST(&so->so_q0)) != 0) { 703 KASSERT(solocked2(so, so2)); 704 (void) soqremque(so2, 0); 705 /* soabort drops the lock. */ 706 (void) soabort(so2); 707 solock(so); 708 continue; 709 } 710 if ((so2 = TAILQ_FIRST(&so->so_q)) != 0) { 711 KASSERT(solocked2(so, so2)); 712 (void) soqremque(so2, 1); 713 /* soabort drops the lock. */ 714 (void) soabort(so2); 715 solock(so); 716 continue; 717 } 718 break; 719 } 720 } 721 if (so->so_pcb == 0) 722 goto discard; 723 if (so->so_state & SS_ISCONNECTED) { 724 if ((so->so_state & SS_ISDISCONNECTING) == 0) { 725 error = sodisconnect(so); 726 if (error) 727 goto drop; 728 } 729 if (so->so_options & SO_LINGER) { 730 if ((so->so_state & (SS_ISDISCONNECTING|SS_NBIO)) == 731 (SS_ISDISCONNECTING|SS_NBIO)) 732 goto drop; 733 while (so->so_state & SS_ISCONNECTED) { 734 error = sowait(so, true, so->so_linger * hz); 735 if (error) 736 break; 737 } 738 } 739 } 740 drop: 741 if (so->so_pcb) { 742 error2 = (*so->so_proto->pr_usrreq)(so, PRU_DETACH, 743 NULL, NULL, NULL, NULL); 744 if (error == 0) 745 error = error2; 746 } 747 discard: 748 if (so->so_state & SS_NOFDREF) 749 panic("soclose: NOFDREF"); 750 kauth_cred_free(so->so_cred); 751 so->so_state |= SS_NOFDREF; 752 sofree(so); 753 return (error); 754 } 755 756 /* 757 * Must be called with the socket locked.. Will return with it unlocked. 758 */ 759 int 760 soabort(struct socket *so) 761 { 762 u_int refs; 763 int error; 764 765 KASSERT(solocked(so)); 766 KASSERT(so->so_head == NULL); 767 768 so->so_aborting++; /* XXX */ 769 error = (*so->so_proto->pr_usrreq)(so, PRU_ABORT, NULL, 770 NULL, NULL, NULL); 771 refs = --so->so_aborting; /* XXX */ 772 if (error || (refs == 0)) { 773 sofree(so); 774 } else { 775 sounlock(so); 776 } 777 return error; 778 } 779 780 int 781 soaccept(struct socket *so, struct mbuf *nam) 782 { 783 int error; 784 785 KASSERT(solocked(so)); 786 787 error = 0; 788 if ((so->so_state & SS_NOFDREF) == 0) 789 panic("soaccept: !NOFDREF"); 790 so->so_state &= ~SS_NOFDREF; 791 if ((so->so_state & SS_ISDISCONNECTED) == 0 || 792 (so->so_proto->pr_flags & PR_ABRTACPTDIS) == 0) 793 error = (*so->so_proto->pr_usrreq)(so, PRU_ACCEPT, 794 NULL, nam, NULL, NULL); 795 else 796 error = ECONNABORTED; 797 798 return (error); 799 } 800 801 int 802 soconnect(struct socket *so, struct mbuf *nam, struct lwp *l) 803 { 804 int error; 805 806 KASSERT(solocked(so)); 807 808 if (so->so_options & SO_ACCEPTCONN) 809 return (EOPNOTSUPP); 810 /* 811 * If protocol is connection-based, can only connect once. 812 * Otherwise, if connected, try to disconnect first. 813 * This allows user to disconnect by connecting to, e.g., 814 * a null address. 815 */ 816 if (so->so_state & (SS_ISCONNECTED|SS_ISCONNECTING) && 817 ((so->so_proto->pr_flags & PR_CONNREQUIRED) || 818 (error = sodisconnect(so)))) 819 error = EISCONN; 820 else 821 error = (*so->so_proto->pr_usrreq)(so, PRU_CONNECT, 822 NULL, nam, NULL, l); 823 return (error); 824 } 825 826 int 827 soconnect2(struct socket *so1, struct socket *so2) 828 { 829 int error; 830 831 KASSERT(solocked2(so1, so2)); 832 833 error = (*so1->so_proto->pr_usrreq)(so1, PRU_CONNECT2, 834 NULL, (struct mbuf *)so2, NULL, NULL); 835 return (error); 836 } 837 838 int 839 sodisconnect(struct socket *so) 840 { 841 int error; 842 843 KASSERT(solocked(so)); 844 845 if ((so->so_state & SS_ISCONNECTED) == 0) { 846 error = ENOTCONN; 847 } else if (so->so_state & SS_ISDISCONNECTING) { 848 error = EALREADY; 849 } else { 850 error = (*so->so_proto->pr_usrreq)(so, PRU_DISCONNECT, 851 NULL, NULL, NULL, NULL); 852 } 853 return (error); 854 } 855 856 #define SBLOCKWAIT(f) (((f) & MSG_DONTWAIT) ? M_NOWAIT : M_WAITOK) 857 /* 858 * Send on a socket. 859 * If send must go all at once and message is larger than 860 * send buffering, then hard error. 861 * Lock against other senders. 862 * If must go all at once and not enough room now, then 863 * inform user that this would block and do nothing. 864 * Otherwise, if nonblocking, send as much as possible. 865 * The data to be sent is described by "uio" if nonzero, 866 * otherwise by the mbuf chain "top" (which must be null 867 * if uio is not). Data provided in mbuf chain must be small 868 * enough to send all at once. 869 * 870 * Returns nonzero on error, timeout or signal; callers 871 * must check for short counts if EINTR/ERESTART are returned. 872 * Data and control buffers are freed on return. 873 */ 874 int 875 sosend(struct socket *so, struct mbuf *addr, struct uio *uio, struct mbuf *top, 876 struct mbuf *control, int flags, struct lwp *l) 877 { 878 struct mbuf **mp, *m; 879 struct proc *p; 880 long space, len, resid, clen, mlen; 881 int error, s, dontroute, atomic; 882 short wakeup_state = 0; 883 884 p = l->l_proc; 885 clen = 0; 886 887 /* 888 * solock() provides atomicity of access. splsoftnet() prevents 889 * protocol processing soft interrupts from interrupting us and 890 * blocking (expensive). 891 */ 892 s = splsoftnet(); 893 solock(so); 894 atomic = sosendallatonce(so) || top; 895 if (uio) 896 resid = uio->uio_resid; 897 else 898 resid = top->m_pkthdr.len; 899 /* 900 * In theory resid should be unsigned. 901 * However, space must be signed, as it might be less than 0 902 * if we over-committed, and we must use a signed comparison 903 * of space and resid. On the other hand, a negative resid 904 * causes us to loop sending 0-length segments to the protocol. 905 */ 906 if (resid < 0) { 907 error = EINVAL; 908 goto out; 909 } 910 dontroute = 911 (flags & MSG_DONTROUTE) && (so->so_options & SO_DONTROUTE) == 0 && 912 (so->so_proto->pr_flags & PR_ATOMIC); 913 l->l_ru.ru_msgsnd++; 914 if (control) 915 clen = control->m_len; 916 restart: 917 if ((error = sblock(&so->so_snd, SBLOCKWAIT(flags))) != 0) 918 goto out; 919 do { 920 if (so->so_state & SS_CANTSENDMORE) { 921 error = EPIPE; 922 goto release; 923 } 924 if (so->so_error) { 925 error = so->so_error; 926 so->so_error = 0; 927 goto release; 928 } 929 if ((so->so_state & SS_ISCONNECTED) == 0) { 930 if (so->so_proto->pr_flags & PR_CONNREQUIRED) { 931 if ((so->so_state & SS_ISCONFIRMING) == 0 && 932 !(resid == 0 && clen != 0)) { 933 error = ENOTCONN; 934 goto release; 935 } 936 } else if (addr == 0) { 937 error = EDESTADDRREQ; 938 goto release; 939 } 940 } 941 space = sbspace(&so->so_snd); 942 if (flags & MSG_OOB) 943 space += 1024; 944 if ((atomic && resid > so->so_snd.sb_hiwat) || 945 clen > so->so_snd.sb_hiwat) { 946 error = EMSGSIZE; 947 goto release; 948 } 949 if (space < resid + clen && 950 (atomic || space < so->so_snd.sb_lowat || space < clen)) { 951 if ((so->so_state & SS_NBIO) || (flags & MSG_NBIO)) { 952 error = EWOULDBLOCK; 953 goto release; 954 } 955 sbunlock(&so->so_snd); 956 if (wakeup_state & SS_RESTARTSYS) { 957 error = ERESTART; 958 goto out; 959 } 960 error = sbwait(&so->so_snd); 961 if (error) 962 goto out; 963 wakeup_state = so->so_state; 964 goto restart; 965 } 966 wakeup_state = 0; 967 mp = ⊤ 968 space -= clen; 969 do { 970 if (uio == NULL) { 971 /* 972 * Data is prepackaged in "top". 973 */ 974 resid = 0; 975 if (flags & MSG_EOR) 976 top->m_flags |= M_EOR; 977 } else do { 978 sounlock(so); 979 splx(s); 980 if (top == NULL) { 981 m = m_gethdr(M_WAIT, MT_DATA); 982 mlen = MHLEN; 983 m->m_pkthdr.len = 0; 984 m->m_pkthdr.rcvif = NULL; 985 } else { 986 m = m_get(M_WAIT, MT_DATA); 987 mlen = MLEN; 988 } 989 MCLAIM(m, so->so_snd.sb_mowner); 990 if (sock_loan_thresh >= 0 && 991 uio->uio_iov->iov_len >= sock_loan_thresh && 992 space >= sock_loan_thresh && 993 (len = sosend_loan(so, uio, m, 994 space)) != 0) { 995 SOSEND_COUNTER_INCR(&sosend_loan_big); 996 space -= len; 997 goto have_data; 998 } 999 if (resid >= MINCLSIZE && space >= MCLBYTES) { 1000 SOSEND_COUNTER_INCR(&sosend_copy_big); 1001 m_clget(m, M_DONTWAIT); 1002 if ((m->m_flags & M_EXT) == 0) 1003 goto nopages; 1004 mlen = MCLBYTES; 1005 if (atomic && top == 0) { 1006 len = lmin(MCLBYTES - max_hdr, 1007 resid); 1008 m->m_data += max_hdr; 1009 } else 1010 len = lmin(MCLBYTES, resid); 1011 space -= len; 1012 } else { 1013 nopages: 1014 SOSEND_COUNTER_INCR(&sosend_copy_small); 1015 len = lmin(lmin(mlen, resid), space); 1016 space -= len; 1017 /* 1018 * For datagram protocols, leave room 1019 * for protocol headers in first mbuf. 1020 */ 1021 if (atomic && top == 0 && len < mlen) 1022 MH_ALIGN(m, len); 1023 } 1024 error = uiomove(mtod(m, void *), (int)len, uio); 1025 have_data: 1026 resid = uio->uio_resid; 1027 m->m_len = len; 1028 *mp = m; 1029 top->m_pkthdr.len += len; 1030 s = splsoftnet(); 1031 solock(so); 1032 if (error != 0) 1033 goto release; 1034 mp = &m->m_next; 1035 if (resid <= 0) { 1036 if (flags & MSG_EOR) 1037 top->m_flags |= M_EOR; 1038 break; 1039 } 1040 } while (space > 0 && atomic); 1041 1042 if (so->so_state & SS_CANTSENDMORE) { 1043 error = EPIPE; 1044 goto release; 1045 } 1046 if (dontroute) 1047 so->so_options |= SO_DONTROUTE; 1048 if (resid > 0) 1049 so->so_state |= SS_MORETOCOME; 1050 error = (*so->so_proto->pr_usrreq)(so, 1051 (flags & MSG_OOB) ? PRU_SENDOOB : PRU_SEND, 1052 top, addr, control, curlwp); 1053 if (dontroute) 1054 so->so_options &= ~SO_DONTROUTE; 1055 if (resid > 0) 1056 so->so_state &= ~SS_MORETOCOME; 1057 clen = 0; 1058 control = NULL; 1059 top = NULL; 1060 mp = ⊤ 1061 if (error != 0) 1062 goto release; 1063 } while (resid && space > 0); 1064 } while (resid); 1065 1066 release: 1067 sbunlock(&so->so_snd); 1068 out: 1069 sounlock(so); 1070 splx(s); 1071 if (top) 1072 m_freem(top); 1073 if (control) 1074 m_freem(control); 1075 return (error); 1076 } 1077 1078 /* 1079 * Following replacement or removal of the first mbuf on the first 1080 * mbuf chain of a socket buffer, push necessary state changes back 1081 * into the socket buffer so that other consumers see the values 1082 * consistently. 'nextrecord' is the callers locally stored value of 1083 * the original value of sb->sb_mb->m_nextpkt which must be restored 1084 * when the lead mbuf changes. NOTE: 'nextrecord' may be NULL. 1085 */ 1086 static void 1087 sbsync(struct sockbuf *sb, struct mbuf *nextrecord) 1088 { 1089 1090 KASSERT(solocked(sb->sb_so)); 1091 1092 /* 1093 * First, update for the new value of nextrecord. If necessary, 1094 * make it the first record. 1095 */ 1096 if (sb->sb_mb != NULL) 1097 sb->sb_mb->m_nextpkt = nextrecord; 1098 else 1099 sb->sb_mb = nextrecord; 1100 1101 /* 1102 * Now update any dependent socket buffer fields to reflect 1103 * the new state. This is an inline of SB_EMPTY_FIXUP, with 1104 * the addition of a second clause that takes care of the 1105 * case where sb_mb has been updated, but remains the last 1106 * record. 1107 */ 1108 if (sb->sb_mb == NULL) { 1109 sb->sb_mbtail = NULL; 1110 sb->sb_lastrecord = NULL; 1111 } else if (sb->sb_mb->m_nextpkt == NULL) 1112 sb->sb_lastrecord = sb->sb_mb; 1113 } 1114 1115 /* 1116 * Implement receive operations on a socket. 1117 * We depend on the way that records are added to the sockbuf 1118 * by sbappend*. In particular, each record (mbufs linked through m_next) 1119 * must begin with an address if the protocol so specifies, 1120 * followed by an optional mbuf or mbufs containing ancillary data, 1121 * and then zero or more mbufs of data. 1122 * In order to avoid blocking network interrupts for the entire time here, 1123 * we splx() while doing the actual copy to user space. 1124 * Although the sockbuf is locked, new data may still be appended, 1125 * and thus we must maintain consistency of the sockbuf during that time. 1126 * 1127 * The caller may receive the data as a single mbuf chain by supplying 1128 * an mbuf **mp0 for use in returning the chain. The uio is then used 1129 * only for the count in uio_resid. 1130 */ 1131 int 1132 soreceive(struct socket *so, struct mbuf **paddr, struct uio *uio, 1133 struct mbuf **mp0, struct mbuf **controlp, int *flagsp) 1134 { 1135 struct lwp *l = curlwp; 1136 struct mbuf *m, **mp, *mt; 1137 size_t len, offset, moff, orig_resid; 1138 int atomic, flags, error, s, type; 1139 const struct protosw *pr; 1140 struct mbuf *nextrecord; 1141 int mbuf_removed = 0; 1142 const struct domain *dom; 1143 short wakeup_state = 0; 1144 1145 pr = so->so_proto; 1146 atomic = pr->pr_flags & PR_ATOMIC; 1147 dom = pr->pr_domain; 1148 mp = mp0; 1149 type = 0; 1150 orig_resid = uio->uio_resid; 1151 1152 if (paddr != NULL) 1153 *paddr = NULL; 1154 if (controlp != NULL) 1155 *controlp = NULL; 1156 if (flagsp != NULL) 1157 flags = *flagsp &~ MSG_EOR; 1158 else 1159 flags = 0; 1160 1161 if (flags & MSG_OOB) { 1162 m = m_get(M_WAIT, MT_DATA); 1163 solock(so); 1164 error = (*pr->pr_usrreq)(so, PRU_RCVOOB, m, 1165 (struct mbuf *)(long)(flags & MSG_PEEK), NULL, l); 1166 sounlock(so); 1167 if (error) 1168 goto bad; 1169 do { 1170 error = uiomove(mtod(m, void *), 1171 MIN(uio->uio_resid, m->m_len), uio); 1172 m = m_free(m); 1173 } while (uio->uio_resid > 0 && error == 0 && m); 1174 bad: 1175 if (m != NULL) 1176 m_freem(m); 1177 return error; 1178 } 1179 if (mp != NULL) 1180 *mp = NULL; 1181 1182 /* 1183 * solock() provides atomicity of access. splsoftnet() prevents 1184 * protocol processing soft interrupts from interrupting us and 1185 * blocking (expensive). 1186 */ 1187 s = splsoftnet(); 1188 solock(so); 1189 if (so->so_state & SS_ISCONFIRMING && uio->uio_resid) 1190 (*pr->pr_usrreq)(so, PRU_RCVD, NULL, NULL, NULL, l); 1191 1192 restart: 1193 if ((error = sblock(&so->so_rcv, SBLOCKWAIT(flags))) != 0) { 1194 sounlock(so); 1195 splx(s); 1196 return error; 1197 } 1198 1199 m = so->so_rcv.sb_mb; 1200 /* 1201 * If we have less data than requested, block awaiting more 1202 * (subject to any timeout) if: 1203 * 1. the current count is less than the low water mark, 1204 * 2. MSG_WAITALL is set, and it is possible to do the entire 1205 * receive operation at once if we block (resid <= hiwat), or 1206 * 3. MSG_DONTWAIT is not set. 1207 * If MSG_WAITALL is set but resid is larger than the receive buffer, 1208 * we have to do the receive in sections, and thus risk returning 1209 * a short count if a timeout or signal occurs after we start. 1210 */ 1211 if (m == NULL || 1212 ((flags & MSG_DONTWAIT) == 0 && 1213 so->so_rcv.sb_cc < uio->uio_resid && 1214 (so->so_rcv.sb_cc < so->so_rcv.sb_lowat || 1215 ((flags & MSG_WAITALL) && 1216 uio->uio_resid <= so->so_rcv.sb_hiwat)) && 1217 m->m_nextpkt == NULL && !atomic)) { 1218 #ifdef DIAGNOSTIC 1219 if (m == NULL && so->so_rcv.sb_cc) 1220 panic("receive 1"); 1221 #endif 1222 if (so->so_error) { 1223 if (m != NULL) 1224 goto dontblock; 1225 error = so->so_error; 1226 if ((flags & MSG_PEEK) == 0) 1227 so->so_error = 0; 1228 goto release; 1229 } 1230 if (so->so_state & SS_CANTRCVMORE) { 1231 if (m != NULL) 1232 goto dontblock; 1233 else 1234 goto release; 1235 } 1236 for (; m != NULL; m = m->m_next) 1237 if (m->m_type == MT_OOBDATA || (m->m_flags & M_EOR)) { 1238 m = so->so_rcv.sb_mb; 1239 goto dontblock; 1240 } 1241 if ((so->so_state & (SS_ISCONNECTED|SS_ISCONNECTING)) == 0 && 1242 (so->so_proto->pr_flags & PR_CONNREQUIRED)) { 1243 error = ENOTCONN; 1244 goto release; 1245 } 1246 if (uio->uio_resid == 0) 1247 goto release; 1248 if ((so->so_state & SS_NBIO) || 1249 (flags & (MSG_DONTWAIT|MSG_NBIO))) { 1250 error = EWOULDBLOCK; 1251 goto release; 1252 } 1253 SBLASTRECORDCHK(&so->so_rcv, "soreceive sbwait 1"); 1254 SBLASTMBUFCHK(&so->so_rcv, "soreceive sbwait 1"); 1255 sbunlock(&so->so_rcv); 1256 if (wakeup_state & SS_RESTARTSYS) 1257 error = ERESTART; 1258 else 1259 error = sbwait(&so->so_rcv); 1260 if (error != 0) { 1261 sounlock(so); 1262 splx(s); 1263 return error; 1264 } 1265 wakeup_state = so->so_state; 1266 goto restart; 1267 } 1268 dontblock: 1269 /* 1270 * On entry here, m points to the first record of the socket buffer. 1271 * From this point onward, we maintain 'nextrecord' as a cache of the 1272 * pointer to the next record in the socket buffer. We must keep the 1273 * various socket buffer pointers and local stack versions of the 1274 * pointers in sync, pushing out modifications before dropping the 1275 * socket lock, and re-reading them when picking it up. 1276 * 1277 * Otherwise, we will race with the network stack appending new data 1278 * or records onto the socket buffer by using inconsistent/stale 1279 * versions of the field, possibly resulting in socket buffer 1280 * corruption. 1281 * 1282 * By holding the high-level sblock(), we prevent simultaneous 1283 * readers from pulling off the front of the socket buffer. 1284 */ 1285 if (l != NULL) 1286 l->l_ru.ru_msgrcv++; 1287 KASSERT(m == so->so_rcv.sb_mb); 1288 SBLASTRECORDCHK(&so->so_rcv, "soreceive 1"); 1289 SBLASTMBUFCHK(&so->so_rcv, "soreceive 1"); 1290 nextrecord = m->m_nextpkt; 1291 if (pr->pr_flags & PR_ADDR) { 1292 #ifdef DIAGNOSTIC 1293 if (m->m_type != MT_SONAME) 1294 panic("receive 1a"); 1295 #endif 1296 orig_resid = 0; 1297 if (flags & MSG_PEEK) { 1298 if (paddr) 1299 *paddr = m_copy(m, 0, m->m_len); 1300 m = m->m_next; 1301 } else { 1302 sbfree(&so->so_rcv, m); 1303 mbuf_removed = 1; 1304 if (paddr != NULL) { 1305 *paddr = m; 1306 so->so_rcv.sb_mb = m->m_next; 1307 m->m_next = NULL; 1308 m = so->so_rcv.sb_mb; 1309 } else { 1310 MFREE(m, so->so_rcv.sb_mb); 1311 m = so->so_rcv.sb_mb; 1312 } 1313 sbsync(&so->so_rcv, nextrecord); 1314 } 1315 } 1316 1317 /* 1318 * Process one or more MT_CONTROL mbufs present before any data mbufs 1319 * in the first mbuf chain on the socket buffer. If MSG_PEEK, we 1320 * just copy the data; if !MSG_PEEK, we call into the protocol to 1321 * perform externalization (or freeing if controlp == NULL). 1322 */ 1323 if (__predict_false(m != NULL && m->m_type == MT_CONTROL)) { 1324 struct mbuf *cm = NULL, *cmn; 1325 struct mbuf **cme = &cm; 1326 1327 do { 1328 if (flags & MSG_PEEK) { 1329 if (controlp != NULL) { 1330 *controlp = m_copy(m, 0, m->m_len); 1331 controlp = &(*controlp)->m_next; 1332 } 1333 m = m->m_next; 1334 } else { 1335 sbfree(&so->so_rcv, m); 1336 so->so_rcv.sb_mb = m->m_next; 1337 m->m_next = NULL; 1338 *cme = m; 1339 cme = &(*cme)->m_next; 1340 m = so->so_rcv.sb_mb; 1341 } 1342 } while (m != NULL && m->m_type == MT_CONTROL); 1343 if ((flags & MSG_PEEK) == 0) 1344 sbsync(&so->so_rcv, nextrecord); 1345 for (; cm != NULL; cm = cmn) { 1346 cmn = cm->m_next; 1347 cm->m_next = NULL; 1348 type = mtod(cm, struct cmsghdr *)->cmsg_type; 1349 if (controlp != NULL) { 1350 if (dom->dom_externalize != NULL && 1351 type == SCM_RIGHTS) { 1352 sounlock(so); 1353 splx(s); 1354 error = (*dom->dom_externalize)(cm, l, 1355 (flags & MSG_CMSG_CLOEXEC) ? 1356 O_CLOEXEC : 0); 1357 s = splsoftnet(); 1358 solock(so); 1359 } 1360 *controlp = cm; 1361 while (*controlp != NULL) 1362 controlp = &(*controlp)->m_next; 1363 } else { 1364 /* 1365 * Dispose of any SCM_RIGHTS message that went 1366 * through the read path rather than recv. 1367 */ 1368 if (dom->dom_dispose != NULL && 1369 type == SCM_RIGHTS) { 1370 sounlock(so); 1371 (*dom->dom_dispose)(cm); 1372 solock(so); 1373 } 1374 m_freem(cm); 1375 } 1376 } 1377 if (m != NULL) 1378 nextrecord = so->so_rcv.sb_mb->m_nextpkt; 1379 else 1380 nextrecord = so->so_rcv.sb_mb; 1381 orig_resid = 0; 1382 } 1383 1384 /* If m is non-NULL, we have some data to read. */ 1385 if (__predict_true(m != NULL)) { 1386 type = m->m_type; 1387 if (type == MT_OOBDATA) 1388 flags |= MSG_OOB; 1389 } 1390 SBLASTRECORDCHK(&so->so_rcv, "soreceive 2"); 1391 SBLASTMBUFCHK(&so->so_rcv, "soreceive 2"); 1392 1393 moff = 0; 1394 offset = 0; 1395 while (m != NULL && uio->uio_resid > 0 && error == 0) { 1396 if (m->m_type == MT_OOBDATA) { 1397 if (type != MT_OOBDATA) 1398 break; 1399 } else if (type == MT_OOBDATA) 1400 break; 1401 #ifdef DIAGNOSTIC 1402 else if (m->m_type != MT_DATA && m->m_type != MT_HEADER) 1403 panic("receive 3"); 1404 #endif 1405 so->so_state &= ~SS_RCVATMARK; 1406 wakeup_state = 0; 1407 len = uio->uio_resid; 1408 if (so->so_oobmark && len > so->so_oobmark - offset) 1409 len = so->so_oobmark - offset; 1410 if (len > m->m_len - moff) 1411 len = m->m_len - moff; 1412 /* 1413 * If mp is set, just pass back the mbufs. 1414 * Otherwise copy them out via the uio, then free. 1415 * Sockbuf must be consistent here (points to current mbuf, 1416 * it points to next record) when we drop priority; 1417 * we must note any additions to the sockbuf when we 1418 * block interrupts again. 1419 */ 1420 if (mp == NULL) { 1421 SBLASTRECORDCHK(&so->so_rcv, "soreceive uiomove"); 1422 SBLASTMBUFCHK(&so->so_rcv, "soreceive uiomove"); 1423 sounlock(so); 1424 splx(s); 1425 error = uiomove(mtod(m, char *) + moff, len, uio); 1426 s = splsoftnet(); 1427 solock(so); 1428 if (error != 0) { 1429 /* 1430 * If any part of the record has been removed 1431 * (such as the MT_SONAME mbuf, which will 1432 * happen when PR_ADDR, and thus also 1433 * PR_ATOMIC, is set), then drop the entire 1434 * record to maintain the atomicity of the 1435 * receive operation. 1436 * 1437 * This avoids a later panic("receive 1a") 1438 * when compiled with DIAGNOSTIC. 1439 */ 1440 if (m && mbuf_removed && atomic) 1441 (void) sbdroprecord(&so->so_rcv); 1442 1443 goto release; 1444 } 1445 } else 1446 uio->uio_resid -= len; 1447 if (len == m->m_len - moff) { 1448 if (m->m_flags & M_EOR) 1449 flags |= MSG_EOR; 1450 if (flags & MSG_PEEK) { 1451 m = m->m_next; 1452 moff = 0; 1453 } else { 1454 nextrecord = m->m_nextpkt; 1455 sbfree(&so->so_rcv, m); 1456 if (mp) { 1457 *mp = m; 1458 mp = &m->m_next; 1459 so->so_rcv.sb_mb = m = m->m_next; 1460 *mp = NULL; 1461 } else { 1462 MFREE(m, so->so_rcv.sb_mb); 1463 m = so->so_rcv.sb_mb; 1464 } 1465 /* 1466 * If m != NULL, we also know that 1467 * so->so_rcv.sb_mb != NULL. 1468 */ 1469 KASSERT(so->so_rcv.sb_mb == m); 1470 if (m) { 1471 m->m_nextpkt = nextrecord; 1472 if (nextrecord == NULL) 1473 so->so_rcv.sb_lastrecord = m; 1474 } else { 1475 so->so_rcv.sb_mb = nextrecord; 1476 SB_EMPTY_FIXUP(&so->so_rcv); 1477 } 1478 SBLASTRECORDCHK(&so->so_rcv, "soreceive 3"); 1479 SBLASTMBUFCHK(&so->so_rcv, "soreceive 3"); 1480 } 1481 } else if (flags & MSG_PEEK) 1482 moff += len; 1483 else { 1484 if (mp != NULL) { 1485 mt = m_copym(m, 0, len, M_NOWAIT); 1486 if (__predict_false(mt == NULL)) { 1487 sounlock(so); 1488 mt = m_copym(m, 0, len, M_WAIT); 1489 solock(so); 1490 } 1491 *mp = mt; 1492 } 1493 m->m_data += len; 1494 m->m_len -= len; 1495 so->so_rcv.sb_cc -= len; 1496 } 1497 if (so->so_oobmark) { 1498 if ((flags & MSG_PEEK) == 0) { 1499 so->so_oobmark -= len; 1500 if (so->so_oobmark == 0) { 1501 so->so_state |= SS_RCVATMARK; 1502 break; 1503 } 1504 } else { 1505 offset += len; 1506 if (offset == so->so_oobmark) 1507 break; 1508 } 1509 } 1510 if (flags & MSG_EOR) 1511 break; 1512 /* 1513 * If the MSG_WAITALL flag is set (for non-atomic socket), 1514 * we must not quit until "uio->uio_resid == 0" or an error 1515 * termination. If a signal/timeout occurs, return 1516 * with a short count but without error. 1517 * Keep sockbuf locked against other readers. 1518 */ 1519 while (flags & MSG_WAITALL && m == NULL && uio->uio_resid > 0 && 1520 !sosendallatonce(so) && !nextrecord) { 1521 if (so->so_error || so->so_state & SS_CANTRCVMORE) 1522 break; 1523 /* 1524 * If we are peeking and the socket receive buffer is 1525 * full, stop since we can't get more data to peek at. 1526 */ 1527 if ((flags & MSG_PEEK) && sbspace(&so->so_rcv) <= 0) 1528 break; 1529 /* 1530 * If we've drained the socket buffer, tell the 1531 * protocol in case it needs to do something to 1532 * get it filled again. 1533 */ 1534 if ((pr->pr_flags & PR_WANTRCVD) && so->so_pcb) 1535 (*pr->pr_usrreq)(so, PRU_RCVD, 1536 NULL, (struct mbuf *)(long)flags, NULL, l); 1537 SBLASTRECORDCHK(&so->so_rcv, "soreceive sbwait 2"); 1538 SBLASTMBUFCHK(&so->so_rcv, "soreceive sbwait 2"); 1539 if (wakeup_state & SS_RESTARTSYS) 1540 error = ERESTART; 1541 else 1542 error = sbwait(&so->so_rcv); 1543 if (error != 0) { 1544 sbunlock(&so->so_rcv); 1545 sounlock(so); 1546 splx(s); 1547 return 0; 1548 } 1549 if ((m = so->so_rcv.sb_mb) != NULL) 1550 nextrecord = m->m_nextpkt; 1551 wakeup_state = so->so_state; 1552 } 1553 } 1554 1555 if (m && atomic) { 1556 flags |= MSG_TRUNC; 1557 if ((flags & MSG_PEEK) == 0) 1558 (void) sbdroprecord(&so->so_rcv); 1559 } 1560 if ((flags & MSG_PEEK) == 0) { 1561 if (m == NULL) { 1562 /* 1563 * First part is an inline SB_EMPTY_FIXUP(). Second 1564 * part makes sure sb_lastrecord is up-to-date if 1565 * there is still data in the socket buffer. 1566 */ 1567 so->so_rcv.sb_mb = nextrecord; 1568 if (so->so_rcv.sb_mb == NULL) { 1569 so->so_rcv.sb_mbtail = NULL; 1570 so->so_rcv.sb_lastrecord = NULL; 1571 } else if (nextrecord->m_nextpkt == NULL) 1572 so->so_rcv.sb_lastrecord = nextrecord; 1573 } 1574 SBLASTRECORDCHK(&so->so_rcv, "soreceive 4"); 1575 SBLASTMBUFCHK(&so->so_rcv, "soreceive 4"); 1576 if (pr->pr_flags & PR_WANTRCVD && so->so_pcb) 1577 (*pr->pr_usrreq)(so, PRU_RCVD, NULL, 1578 (struct mbuf *)(long)flags, NULL, l); 1579 } 1580 if (orig_resid == uio->uio_resid && orig_resid && 1581 (flags & MSG_EOR) == 0 && (so->so_state & SS_CANTRCVMORE) == 0) { 1582 sbunlock(&so->so_rcv); 1583 goto restart; 1584 } 1585 1586 if (flagsp != NULL) 1587 *flagsp |= flags; 1588 release: 1589 sbunlock(&so->so_rcv); 1590 sounlock(so); 1591 splx(s); 1592 return error; 1593 } 1594 1595 int 1596 soshutdown(struct socket *so, int how) 1597 { 1598 const struct protosw *pr; 1599 int error; 1600 1601 KASSERT(solocked(so)); 1602 1603 pr = so->so_proto; 1604 if (!(how == SHUT_RD || how == SHUT_WR || how == SHUT_RDWR)) 1605 return (EINVAL); 1606 1607 if (how == SHUT_RD || how == SHUT_RDWR) { 1608 sorflush(so); 1609 error = 0; 1610 } 1611 if (how == SHUT_WR || how == SHUT_RDWR) 1612 error = (*pr->pr_usrreq)(so, PRU_SHUTDOWN, NULL, 1613 NULL, NULL, NULL); 1614 1615 return error; 1616 } 1617 1618 void 1619 sorestart(struct socket *so) 1620 { 1621 /* 1622 * An application has called close() on an fd on which another 1623 * of its threads has called a socket system call. 1624 * Mark this and wake everyone up, and code that would block again 1625 * instead returns ERESTART. 1626 * On system call re-entry the fd is validated and EBADF returned. 1627 * Any other fd will block again on the 2nd syscall. 1628 */ 1629 solock(so); 1630 so->so_state |= SS_RESTARTSYS; 1631 cv_broadcast(&so->so_cv); 1632 cv_broadcast(&so->so_snd.sb_cv); 1633 cv_broadcast(&so->so_rcv.sb_cv); 1634 sounlock(so); 1635 } 1636 1637 void 1638 sorflush(struct socket *so) 1639 { 1640 struct sockbuf *sb, asb; 1641 const struct protosw *pr; 1642 1643 KASSERT(solocked(so)); 1644 1645 sb = &so->so_rcv; 1646 pr = so->so_proto; 1647 socantrcvmore(so); 1648 sb->sb_flags |= SB_NOINTR; 1649 (void )sblock(sb, M_WAITOK); 1650 sbunlock(sb); 1651 asb = *sb; 1652 /* 1653 * Clear most of the sockbuf structure, but leave some of the 1654 * fields valid. 1655 */ 1656 memset(&sb->sb_startzero, 0, 1657 sizeof(*sb) - offsetof(struct sockbuf, sb_startzero)); 1658 if (pr->pr_flags & PR_RIGHTS && pr->pr_domain->dom_dispose) { 1659 sounlock(so); 1660 (*pr->pr_domain->dom_dispose)(asb.sb_mb); 1661 solock(so); 1662 } 1663 sbrelease(&asb, so); 1664 } 1665 1666 /* 1667 * internal set SOL_SOCKET options 1668 */ 1669 static int 1670 sosetopt1(struct socket *so, const struct sockopt *sopt) 1671 { 1672 int error = EINVAL, optval, opt; 1673 struct linger l; 1674 struct timeval tv; 1675 1676 switch ((opt = sopt->sopt_name)) { 1677 1678 case SO_ACCEPTFILTER: 1679 error = accept_filt_setopt(so, sopt); 1680 KASSERT(solocked(so)); 1681 break; 1682 1683 case SO_LINGER: 1684 error = sockopt_get(sopt, &l, sizeof(l)); 1685 solock(so); 1686 if (error) 1687 break; 1688 if (l.l_linger < 0 || l.l_linger > USHRT_MAX || 1689 l.l_linger > (INT_MAX / hz)) { 1690 error = EDOM; 1691 break; 1692 } 1693 so->so_linger = l.l_linger; 1694 if (l.l_onoff) 1695 so->so_options |= SO_LINGER; 1696 else 1697 so->so_options &= ~SO_LINGER; 1698 break; 1699 1700 case SO_DEBUG: 1701 case SO_KEEPALIVE: 1702 case SO_DONTROUTE: 1703 case SO_USELOOPBACK: 1704 case SO_BROADCAST: 1705 case SO_REUSEADDR: 1706 case SO_REUSEPORT: 1707 case SO_OOBINLINE: 1708 case SO_TIMESTAMP: 1709 case SO_NOSIGPIPE: 1710 #ifdef SO_OTIMESTAMP 1711 case SO_OTIMESTAMP: 1712 #endif 1713 error = sockopt_getint(sopt, &optval); 1714 solock(so); 1715 if (error) 1716 break; 1717 if (optval) 1718 so->so_options |= opt; 1719 else 1720 so->so_options &= ~opt; 1721 break; 1722 1723 case SO_SNDBUF: 1724 case SO_RCVBUF: 1725 case SO_SNDLOWAT: 1726 case SO_RCVLOWAT: 1727 error = sockopt_getint(sopt, &optval); 1728 solock(so); 1729 if (error) 1730 break; 1731 1732 /* 1733 * Values < 1 make no sense for any of these 1734 * options, so disallow them. 1735 */ 1736 if (optval < 1) { 1737 error = EINVAL; 1738 break; 1739 } 1740 1741 switch (opt) { 1742 case SO_SNDBUF: 1743 if (sbreserve(&so->so_snd, (u_long)optval, so) == 0) { 1744 error = ENOBUFS; 1745 break; 1746 } 1747 so->so_snd.sb_flags &= ~SB_AUTOSIZE; 1748 break; 1749 1750 case SO_RCVBUF: 1751 if (sbreserve(&so->so_rcv, (u_long)optval, so) == 0) { 1752 error = ENOBUFS; 1753 break; 1754 } 1755 so->so_rcv.sb_flags &= ~SB_AUTOSIZE; 1756 break; 1757 1758 /* 1759 * Make sure the low-water is never greater than 1760 * the high-water. 1761 */ 1762 case SO_SNDLOWAT: 1763 if (optval > so->so_snd.sb_hiwat) 1764 optval = so->so_snd.sb_hiwat; 1765 1766 so->so_snd.sb_lowat = optval; 1767 break; 1768 1769 case SO_RCVLOWAT: 1770 if (optval > so->so_rcv.sb_hiwat) 1771 optval = so->so_rcv.sb_hiwat; 1772 1773 so->so_rcv.sb_lowat = optval; 1774 break; 1775 } 1776 break; 1777 1778 #ifdef COMPAT_50 1779 case SO_OSNDTIMEO: 1780 case SO_ORCVTIMEO: { 1781 struct timeval50 otv; 1782 error = sockopt_get(sopt, &otv, sizeof(otv)); 1783 if (error) { 1784 solock(so); 1785 break; 1786 } 1787 timeval50_to_timeval(&otv, &tv); 1788 opt = opt == SO_OSNDTIMEO ? SO_SNDTIMEO : SO_RCVTIMEO; 1789 error = 0; 1790 /*FALLTHROUGH*/ 1791 } 1792 #endif /* COMPAT_50 */ 1793 1794 case SO_SNDTIMEO: 1795 case SO_RCVTIMEO: 1796 if (error) 1797 error = sockopt_get(sopt, &tv, sizeof(tv)); 1798 solock(so); 1799 if (error) 1800 break; 1801 1802 if (tv.tv_sec > (INT_MAX - tv.tv_usec / tick) / hz) { 1803 error = EDOM; 1804 break; 1805 } 1806 1807 optval = tv.tv_sec * hz + tv.tv_usec / tick; 1808 if (optval == 0 && tv.tv_usec != 0) 1809 optval = 1; 1810 1811 switch (opt) { 1812 case SO_SNDTIMEO: 1813 so->so_snd.sb_timeo = optval; 1814 break; 1815 case SO_RCVTIMEO: 1816 so->so_rcv.sb_timeo = optval; 1817 break; 1818 } 1819 break; 1820 1821 default: 1822 solock(so); 1823 error = ENOPROTOOPT; 1824 break; 1825 } 1826 KASSERT(solocked(so)); 1827 return error; 1828 } 1829 1830 int 1831 sosetopt(struct socket *so, struct sockopt *sopt) 1832 { 1833 int error, prerr; 1834 1835 if (sopt->sopt_level == SOL_SOCKET) { 1836 error = sosetopt1(so, sopt); 1837 KASSERT(solocked(so)); 1838 } else { 1839 error = ENOPROTOOPT; 1840 solock(so); 1841 } 1842 1843 if ((error == 0 || error == ENOPROTOOPT) && 1844 so->so_proto != NULL && so->so_proto->pr_ctloutput != NULL) { 1845 /* give the protocol stack a shot */ 1846 prerr = (*so->so_proto->pr_ctloutput)(PRCO_SETOPT, so, sopt); 1847 if (prerr == 0) 1848 error = 0; 1849 else if (prerr != ENOPROTOOPT) 1850 error = prerr; 1851 } 1852 sounlock(so); 1853 return error; 1854 } 1855 1856 /* 1857 * so_setsockopt() is a wrapper providing a sockopt structure for sosetopt() 1858 */ 1859 int 1860 so_setsockopt(struct lwp *l, struct socket *so, int level, int name, 1861 const void *val, size_t valsize) 1862 { 1863 struct sockopt sopt; 1864 int error; 1865 1866 KASSERT(valsize == 0 || val != NULL); 1867 1868 sockopt_init(&sopt, level, name, valsize); 1869 sockopt_set(&sopt, val, valsize); 1870 1871 error = sosetopt(so, &sopt); 1872 1873 sockopt_destroy(&sopt); 1874 1875 return error; 1876 } 1877 1878 /* 1879 * internal get SOL_SOCKET options 1880 */ 1881 static int 1882 sogetopt1(struct socket *so, struct sockopt *sopt) 1883 { 1884 int error, optval, opt; 1885 struct linger l; 1886 struct timeval tv; 1887 1888 switch ((opt = sopt->sopt_name)) { 1889 1890 case SO_ACCEPTFILTER: 1891 error = accept_filt_getopt(so, sopt); 1892 break; 1893 1894 case SO_LINGER: 1895 l.l_onoff = (so->so_options & SO_LINGER) ? 1 : 0; 1896 l.l_linger = so->so_linger; 1897 1898 error = sockopt_set(sopt, &l, sizeof(l)); 1899 break; 1900 1901 case SO_USELOOPBACK: 1902 case SO_DONTROUTE: 1903 case SO_DEBUG: 1904 case SO_KEEPALIVE: 1905 case SO_REUSEADDR: 1906 case SO_REUSEPORT: 1907 case SO_BROADCAST: 1908 case SO_OOBINLINE: 1909 case SO_TIMESTAMP: 1910 case SO_NOSIGPIPE: 1911 #ifdef SO_OTIMESTAMP 1912 case SO_OTIMESTAMP: 1913 #endif 1914 error = sockopt_setint(sopt, (so->so_options & opt) ? 1 : 0); 1915 break; 1916 1917 case SO_TYPE: 1918 error = sockopt_setint(sopt, so->so_type); 1919 break; 1920 1921 case SO_ERROR: 1922 error = sockopt_setint(sopt, so->so_error); 1923 so->so_error = 0; 1924 break; 1925 1926 case SO_SNDBUF: 1927 error = sockopt_setint(sopt, so->so_snd.sb_hiwat); 1928 break; 1929 1930 case SO_RCVBUF: 1931 error = sockopt_setint(sopt, so->so_rcv.sb_hiwat); 1932 break; 1933 1934 case SO_SNDLOWAT: 1935 error = sockopt_setint(sopt, so->so_snd.sb_lowat); 1936 break; 1937 1938 case SO_RCVLOWAT: 1939 error = sockopt_setint(sopt, so->so_rcv.sb_lowat); 1940 break; 1941 1942 #ifdef COMPAT_50 1943 case SO_OSNDTIMEO: 1944 case SO_ORCVTIMEO: { 1945 struct timeval50 otv; 1946 1947 optval = (opt == SO_OSNDTIMEO ? 1948 so->so_snd.sb_timeo : so->so_rcv.sb_timeo); 1949 1950 otv.tv_sec = optval / hz; 1951 otv.tv_usec = (optval % hz) * tick; 1952 1953 error = sockopt_set(sopt, &otv, sizeof(otv)); 1954 break; 1955 } 1956 #endif /* COMPAT_50 */ 1957 1958 case SO_SNDTIMEO: 1959 case SO_RCVTIMEO: 1960 optval = (opt == SO_SNDTIMEO ? 1961 so->so_snd.sb_timeo : so->so_rcv.sb_timeo); 1962 1963 tv.tv_sec = optval / hz; 1964 tv.tv_usec = (optval % hz) * tick; 1965 1966 error = sockopt_set(sopt, &tv, sizeof(tv)); 1967 break; 1968 1969 case SO_OVERFLOWED: 1970 error = sockopt_setint(sopt, so->so_rcv.sb_overflowed); 1971 break; 1972 1973 default: 1974 error = ENOPROTOOPT; 1975 break; 1976 } 1977 1978 return (error); 1979 } 1980 1981 int 1982 sogetopt(struct socket *so, struct sockopt *sopt) 1983 { 1984 int error; 1985 1986 solock(so); 1987 if (sopt->sopt_level != SOL_SOCKET) { 1988 if (so->so_proto && so->so_proto->pr_ctloutput) { 1989 error = ((*so->so_proto->pr_ctloutput) 1990 (PRCO_GETOPT, so, sopt)); 1991 } else 1992 error = (ENOPROTOOPT); 1993 } else { 1994 error = sogetopt1(so, sopt); 1995 } 1996 sounlock(so); 1997 return (error); 1998 } 1999 2000 /* 2001 * alloc sockopt data buffer buffer 2002 * - will be released at destroy 2003 */ 2004 static int 2005 sockopt_alloc(struct sockopt *sopt, size_t len, km_flag_t kmflag) 2006 { 2007 2008 KASSERT(sopt->sopt_size == 0); 2009 2010 if (len > sizeof(sopt->sopt_buf)) { 2011 sopt->sopt_data = kmem_zalloc(len, kmflag); 2012 if (sopt->sopt_data == NULL) 2013 return ENOMEM; 2014 } else 2015 sopt->sopt_data = sopt->sopt_buf; 2016 2017 sopt->sopt_size = len; 2018 return 0; 2019 } 2020 2021 /* 2022 * initialise sockopt storage 2023 * - MAY sleep during allocation 2024 */ 2025 void 2026 sockopt_init(struct sockopt *sopt, int level, int name, size_t size) 2027 { 2028 2029 memset(sopt, 0, sizeof(*sopt)); 2030 2031 sopt->sopt_level = level; 2032 sopt->sopt_name = name; 2033 (void)sockopt_alloc(sopt, size, KM_SLEEP); 2034 } 2035 2036 /* 2037 * destroy sockopt storage 2038 * - will release any held memory references 2039 */ 2040 void 2041 sockopt_destroy(struct sockopt *sopt) 2042 { 2043 2044 if (sopt->sopt_data != sopt->sopt_buf) 2045 kmem_free(sopt->sopt_data, sopt->sopt_size); 2046 2047 memset(sopt, 0, sizeof(*sopt)); 2048 } 2049 2050 /* 2051 * set sockopt value 2052 * - value is copied into sockopt 2053 * - memory is allocated when necessary, will not sleep 2054 */ 2055 int 2056 sockopt_set(struct sockopt *sopt, const void *buf, size_t len) 2057 { 2058 int error; 2059 2060 if (sopt->sopt_size == 0) { 2061 error = sockopt_alloc(sopt, len, KM_NOSLEEP); 2062 if (error) 2063 return error; 2064 } 2065 2066 KASSERT(sopt->sopt_size == len); 2067 memcpy(sopt->sopt_data, buf, len); 2068 return 0; 2069 } 2070 2071 /* 2072 * common case of set sockopt integer value 2073 */ 2074 int 2075 sockopt_setint(struct sockopt *sopt, int val) 2076 { 2077 2078 return sockopt_set(sopt, &val, sizeof(int)); 2079 } 2080 2081 /* 2082 * get sockopt value 2083 * - correct size must be given 2084 */ 2085 int 2086 sockopt_get(const struct sockopt *sopt, void *buf, size_t len) 2087 { 2088 2089 if (sopt->sopt_size != len) 2090 return EINVAL; 2091 2092 memcpy(buf, sopt->sopt_data, len); 2093 return 0; 2094 } 2095 2096 /* 2097 * common case of get sockopt integer value 2098 */ 2099 int 2100 sockopt_getint(const struct sockopt *sopt, int *valp) 2101 { 2102 2103 return sockopt_get(sopt, valp, sizeof(int)); 2104 } 2105 2106 /* 2107 * set sockopt value from mbuf 2108 * - ONLY for legacy code 2109 * - mbuf is released by sockopt 2110 * - will not sleep 2111 */ 2112 int 2113 sockopt_setmbuf(struct sockopt *sopt, struct mbuf *m) 2114 { 2115 size_t len; 2116 int error; 2117 2118 len = m_length(m); 2119 2120 if (sopt->sopt_size == 0) { 2121 error = sockopt_alloc(sopt, len, KM_NOSLEEP); 2122 if (error) 2123 return error; 2124 } 2125 2126 KASSERT(sopt->sopt_size == len); 2127 m_copydata(m, 0, len, sopt->sopt_data); 2128 m_freem(m); 2129 2130 return 0; 2131 } 2132 2133 /* 2134 * get sockopt value into mbuf 2135 * - ONLY for legacy code 2136 * - mbuf to be released by the caller 2137 * - will not sleep 2138 */ 2139 struct mbuf * 2140 sockopt_getmbuf(const struct sockopt *sopt) 2141 { 2142 struct mbuf *m; 2143 2144 if (sopt->sopt_size > MCLBYTES) 2145 return NULL; 2146 2147 m = m_get(M_DONTWAIT, MT_SOOPTS); 2148 if (m == NULL) 2149 return NULL; 2150 2151 if (sopt->sopt_size > MLEN) { 2152 MCLGET(m, M_DONTWAIT); 2153 if ((m->m_flags & M_EXT) == 0) { 2154 m_free(m); 2155 return NULL; 2156 } 2157 } 2158 2159 memcpy(mtod(m, void *), sopt->sopt_data, sopt->sopt_size); 2160 m->m_len = sopt->sopt_size; 2161 2162 return m; 2163 } 2164 2165 void 2166 sohasoutofband(struct socket *so) 2167 { 2168 2169 fownsignal(so->so_pgid, SIGURG, POLL_PRI, POLLPRI|POLLRDBAND, so); 2170 selnotify(&so->so_rcv.sb_sel, POLLPRI | POLLRDBAND, NOTE_SUBMIT); 2171 } 2172 2173 static void 2174 filt_sordetach(struct knote *kn) 2175 { 2176 struct socket *so; 2177 2178 so = ((file_t *)kn->kn_obj)->f_data; 2179 solock(so); 2180 SLIST_REMOVE(&so->so_rcv.sb_sel.sel_klist, kn, knote, kn_selnext); 2181 if (SLIST_EMPTY(&so->so_rcv.sb_sel.sel_klist)) 2182 so->so_rcv.sb_flags &= ~SB_KNOTE; 2183 sounlock(so); 2184 } 2185 2186 /*ARGSUSED*/ 2187 static int 2188 filt_soread(struct knote *kn, long hint) 2189 { 2190 struct socket *so; 2191 int rv; 2192 2193 so = ((file_t *)kn->kn_obj)->f_data; 2194 if (hint != NOTE_SUBMIT) 2195 solock(so); 2196 kn->kn_data = so->so_rcv.sb_cc; 2197 if (so->so_state & SS_CANTRCVMORE) { 2198 kn->kn_flags |= EV_EOF; 2199 kn->kn_fflags = so->so_error; 2200 rv = 1; 2201 } else if (so->so_error) /* temporary udp error */ 2202 rv = 1; 2203 else if (kn->kn_sfflags & NOTE_LOWAT) 2204 rv = (kn->kn_data >= kn->kn_sdata); 2205 else 2206 rv = (kn->kn_data >= so->so_rcv.sb_lowat); 2207 if (hint != NOTE_SUBMIT) 2208 sounlock(so); 2209 return rv; 2210 } 2211 2212 static void 2213 filt_sowdetach(struct knote *kn) 2214 { 2215 struct socket *so; 2216 2217 so = ((file_t *)kn->kn_obj)->f_data; 2218 solock(so); 2219 SLIST_REMOVE(&so->so_snd.sb_sel.sel_klist, kn, knote, kn_selnext); 2220 if (SLIST_EMPTY(&so->so_snd.sb_sel.sel_klist)) 2221 so->so_snd.sb_flags &= ~SB_KNOTE; 2222 sounlock(so); 2223 } 2224 2225 /*ARGSUSED*/ 2226 static int 2227 filt_sowrite(struct knote *kn, long hint) 2228 { 2229 struct socket *so; 2230 int rv; 2231 2232 so = ((file_t *)kn->kn_obj)->f_data; 2233 if (hint != NOTE_SUBMIT) 2234 solock(so); 2235 kn->kn_data = sbspace(&so->so_snd); 2236 if (so->so_state & SS_CANTSENDMORE) { 2237 kn->kn_flags |= EV_EOF; 2238 kn->kn_fflags = so->so_error; 2239 rv = 1; 2240 } else if (so->so_error) /* temporary udp error */ 2241 rv = 1; 2242 else if (((so->so_state & SS_ISCONNECTED) == 0) && 2243 (so->so_proto->pr_flags & PR_CONNREQUIRED)) 2244 rv = 0; 2245 else if (kn->kn_sfflags & NOTE_LOWAT) 2246 rv = (kn->kn_data >= kn->kn_sdata); 2247 else 2248 rv = (kn->kn_data >= so->so_snd.sb_lowat); 2249 if (hint != NOTE_SUBMIT) 2250 sounlock(so); 2251 return rv; 2252 } 2253 2254 /*ARGSUSED*/ 2255 static int 2256 filt_solisten(struct knote *kn, long hint) 2257 { 2258 struct socket *so; 2259 int rv; 2260 2261 so = ((file_t *)kn->kn_obj)->f_data; 2262 2263 /* 2264 * Set kn_data to number of incoming connections, not 2265 * counting partial (incomplete) connections. 2266 */ 2267 if (hint != NOTE_SUBMIT) 2268 solock(so); 2269 kn->kn_data = so->so_qlen; 2270 rv = (kn->kn_data > 0); 2271 if (hint != NOTE_SUBMIT) 2272 sounlock(so); 2273 return rv; 2274 } 2275 2276 static const struct filterops solisten_filtops = 2277 { 1, NULL, filt_sordetach, filt_solisten }; 2278 static const struct filterops soread_filtops = 2279 { 1, NULL, filt_sordetach, filt_soread }; 2280 static const struct filterops sowrite_filtops = 2281 { 1, NULL, filt_sowdetach, filt_sowrite }; 2282 2283 int 2284 soo_kqfilter(struct file *fp, struct knote *kn) 2285 { 2286 struct socket *so; 2287 struct sockbuf *sb; 2288 2289 so = ((file_t *)kn->kn_obj)->f_data; 2290 solock(so); 2291 switch (kn->kn_filter) { 2292 case EVFILT_READ: 2293 if (so->so_options & SO_ACCEPTCONN) 2294 kn->kn_fop = &solisten_filtops; 2295 else 2296 kn->kn_fop = &soread_filtops; 2297 sb = &so->so_rcv; 2298 break; 2299 case EVFILT_WRITE: 2300 kn->kn_fop = &sowrite_filtops; 2301 sb = &so->so_snd; 2302 break; 2303 default: 2304 sounlock(so); 2305 return (EINVAL); 2306 } 2307 SLIST_INSERT_HEAD(&sb->sb_sel.sel_klist, kn, kn_selnext); 2308 sb->sb_flags |= SB_KNOTE; 2309 sounlock(so); 2310 return (0); 2311 } 2312 2313 static int 2314 sodopoll(struct socket *so, int events) 2315 { 2316 int revents; 2317 2318 revents = 0; 2319 2320 if (events & (POLLIN | POLLRDNORM)) 2321 if (soreadable(so)) 2322 revents |= events & (POLLIN | POLLRDNORM); 2323 2324 if (events & (POLLOUT | POLLWRNORM)) 2325 if (sowritable(so)) 2326 revents |= events & (POLLOUT | POLLWRNORM); 2327 2328 if (events & (POLLPRI | POLLRDBAND)) 2329 if (so->so_oobmark || (so->so_state & SS_RCVATMARK)) 2330 revents |= events & (POLLPRI | POLLRDBAND); 2331 2332 return revents; 2333 } 2334 2335 int 2336 sopoll(struct socket *so, int events) 2337 { 2338 int revents = 0; 2339 2340 #ifndef DIAGNOSTIC 2341 /* 2342 * Do a quick, unlocked check in expectation that the socket 2343 * will be ready for I/O. Don't do this check if DIAGNOSTIC, 2344 * as the solocked() assertions will fail. 2345 */ 2346 if ((revents = sodopoll(so, events)) != 0) 2347 return revents; 2348 #endif 2349 2350 solock(so); 2351 if ((revents = sodopoll(so, events)) == 0) { 2352 if (events & (POLLIN | POLLPRI | POLLRDNORM | POLLRDBAND)) { 2353 selrecord(curlwp, &so->so_rcv.sb_sel); 2354 so->so_rcv.sb_flags |= SB_NOTIFY; 2355 } 2356 2357 if (events & (POLLOUT | POLLWRNORM)) { 2358 selrecord(curlwp, &so->so_snd.sb_sel); 2359 so->so_snd.sb_flags |= SB_NOTIFY; 2360 } 2361 } 2362 sounlock(so); 2363 2364 return revents; 2365 } 2366 2367 2368 #include <sys/sysctl.h> 2369 2370 static int sysctl_kern_somaxkva(SYSCTLFN_PROTO); 2371 static int sysctl_kern_sbmax(SYSCTLFN_PROTO); 2372 2373 /* 2374 * sysctl helper routine for kern.somaxkva. ensures that the given 2375 * value is not too small. 2376 * (XXX should we maybe make sure it's not too large as well?) 2377 */ 2378 static int 2379 sysctl_kern_somaxkva(SYSCTLFN_ARGS) 2380 { 2381 int error, new_somaxkva; 2382 struct sysctlnode node; 2383 2384 new_somaxkva = somaxkva; 2385 node = *rnode; 2386 node.sysctl_data = &new_somaxkva; 2387 error = sysctl_lookup(SYSCTLFN_CALL(&node)); 2388 if (error || newp == NULL) 2389 return (error); 2390 2391 if (new_somaxkva < (16 * 1024 * 1024)) /* sanity */ 2392 return (EINVAL); 2393 2394 mutex_enter(&so_pendfree_lock); 2395 somaxkva = new_somaxkva; 2396 cv_broadcast(&socurkva_cv); 2397 mutex_exit(&so_pendfree_lock); 2398 2399 return (error); 2400 } 2401 2402 /* 2403 * sysctl helper routine for kern.sbmax. Basically just ensures that 2404 * any new value is not too small. 2405 */ 2406 static int 2407 sysctl_kern_sbmax(SYSCTLFN_ARGS) 2408 { 2409 int error, new_sbmax; 2410 struct sysctlnode node; 2411 2412 new_sbmax = sb_max; 2413 node = *rnode; 2414 node.sysctl_data = &new_sbmax; 2415 error = sysctl_lookup(SYSCTLFN_CALL(&node)); 2416 if (error || newp == NULL) 2417 return (error); 2418 2419 KERNEL_LOCK(1, NULL); 2420 error = sb_max_set(new_sbmax); 2421 KERNEL_UNLOCK_ONE(NULL); 2422 2423 return (error); 2424 } 2425 2426 static void 2427 sysctl_kern_socket_setup(void) 2428 { 2429 2430 KASSERT(socket_sysctllog == NULL); 2431 sysctl_createv(&socket_sysctllog, 0, NULL, NULL, 2432 CTLFLAG_PERMANENT, 2433 CTLTYPE_NODE, "kern", NULL, 2434 NULL, 0, NULL, 0, 2435 CTL_KERN, CTL_EOL); 2436 2437 sysctl_createv(&socket_sysctllog, 0, NULL, NULL, 2438 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2439 CTLTYPE_INT, "somaxkva", 2440 SYSCTL_DESCR("Maximum amount of kernel memory to be " 2441 "used for socket buffers"), 2442 sysctl_kern_somaxkva, 0, NULL, 0, 2443 CTL_KERN, KERN_SOMAXKVA, CTL_EOL); 2444 2445 sysctl_createv(&socket_sysctllog, 0, NULL, NULL, 2446 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 2447 CTLTYPE_INT, "sbmax", 2448 SYSCTL_DESCR("Maximum socket buffer size"), 2449 sysctl_kern_sbmax, 0, NULL, 0, 2450 CTL_KERN, KERN_SBMAX, CTL_EOL); 2451 } 2452