uipc_socket.c revision 1.225
1 /* $NetBSD: uipc_socket.c,v 1.225 2014/07/09 14:41:42 rtr 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 1.225 2014/07/09 14:41:42 rtr Exp $"); 75 76 #include "opt_compat_netbsd.h" 77 #include "opt_sock_counters.h" 78 #include "opt_sosend_loan.h" 79 #include "opt_mbuftrace.h" 80 #include "opt_somaxkva.h" 81 #include "opt_multiprocessor.h" /* XXX */ 82 83 #include <sys/param.h> 84 #include <sys/systm.h> 85 #include <sys/proc.h> 86 #include <sys/file.h> 87 #include <sys/filedesc.h> 88 #include <sys/kmem.h> 89 #include <sys/mbuf.h> 90 #include <sys/domain.h> 91 #include <sys/kernel.h> 92 #include <sys/protosw.h> 93 #include <sys/socket.h> 94 #include <sys/socketvar.h> 95 #include <sys/signalvar.h> 96 #include <sys/resourcevar.h> 97 #include <sys/uidinfo.h> 98 #include <sys/event.h> 99 #include <sys/poll.h> 100 #include <sys/kauth.h> 101 #include <sys/mutex.h> 102 #include <sys/condvar.h> 103 #include <sys/kthread.h> 104 105 #ifdef COMPAT_50 106 #include <compat/sys/time.h> 107 #include <compat/sys/socket.h> 108 #endif 109 110 #include <uvm/uvm_extern.h> 111 #include <uvm/uvm_loan.h> 112 #include <uvm/uvm_page.h> 113 114 MALLOC_DEFINE(M_SONAME, "soname", "socket name"); 115 116 extern const struct fileops socketops; 117 118 extern int somaxconn; /* patchable (XXX sysctl) */ 119 int somaxconn = SOMAXCONN; 120 kmutex_t *softnet_lock; 121 122 #ifdef SOSEND_COUNTERS 123 #include <sys/device.h> 124 125 static struct evcnt sosend_loan_big = EVCNT_INITIALIZER(EVCNT_TYPE_MISC, 126 NULL, "sosend", "loan big"); 127 static struct evcnt sosend_copy_big = EVCNT_INITIALIZER(EVCNT_TYPE_MISC, 128 NULL, "sosend", "copy big"); 129 static struct evcnt sosend_copy_small = EVCNT_INITIALIZER(EVCNT_TYPE_MISC, 130 NULL, "sosend", "copy small"); 131 static struct evcnt sosend_kvalimit = EVCNT_INITIALIZER(EVCNT_TYPE_MISC, 132 NULL, "sosend", "kva limit"); 133 134 #define SOSEND_COUNTER_INCR(ev) (ev)->ev_count++ 135 136 EVCNT_ATTACH_STATIC(sosend_loan_big); 137 EVCNT_ATTACH_STATIC(sosend_copy_big); 138 EVCNT_ATTACH_STATIC(sosend_copy_small); 139 EVCNT_ATTACH_STATIC(sosend_kvalimit); 140 #else 141 142 #define SOSEND_COUNTER_INCR(ev) /* nothing */ 143 144 #endif /* SOSEND_COUNTERS */ 145 146 #if defined(SOSEND_NO_LOAN) || defined(MULTIPROCESSOR) 147 int sock_loan_thresh = -1; 148 #else 149 int sock_loan_thresh = 4096; 150 #endif 151 152 static kmutex_t so_pendfree_lock; 153 static struct mbuf *so_pendfree = NULL; 154 155 #ifndef SOMAXKVA 156 #define SOMAXKVA (16 * 1024 * 1024) 157 #endif 158 int somaxkva = SOMAXKVA; 159 static int socurkva; 160 static kcondvar_t socurkva_cv; 161 162 static kauth_listener_t socket_listener; 163 164 #define SOCK_LOAN_CHUNK 65536 165 166 static void sopendfree_thread(void *); 167 static kcondvar_t pendfree_thread_cv; 168 static lwp_t *sopendfree_lwp; 169 170 static void sysctl_kern_socket_setup(void); 171 static struct sysctllog *socket_sysctllog; 172 173 static vsize_t 174 sokvareserve(struct socket *so, vsize_t len) 175 { 176 int error; 177 178 mutex_enter(&so_pendfree_lock); 179 while (socurkva + len > somaxkva) { 180 SOSEND_COUNTER_INCR(&sosend_kvalimit); 181 error = cv_wait_sig(&socurkva_cv, &so_pendfree_lock); 182 if (error) { 183 len = 0; 184 break; 185 } 186 } 187 socurkva += len; 188 mutex_exit(&so_pendfree_lock); 189 return len; 190 } 191 192 static void 193 sokvaunreserve(vsize_t len) 194 { 195 196 mutex_enter(&so_pendfree_lock); 197 socurkva -= len; 198 cv_broadcast(&socurkva_cv); 199 mutex_exit(&so_pendfree_lock); 200 } 201 202 /* 203 * sokvaalloc: allocate kva for loan. 204 */ 205 206 vaddr_t 207 sokvaalloc(vaddr_t sva, vsize_t len, struct socket *so) 208 { 209 vaddr_t lva; 210 211 /* 212 * reserve kva. 213 */ 214 215 if (sokvareserve(so, len) == 0) 216 return 0; 217 218 /* 219 * allocate kva. 220 */ 221 222 lva = uvm_km_alloc(kernel_map, len, atop(sva) & uvmexp.colormask, 223 UVM_KMF_COLORMATCH | UVM_KMF_VAONLY | UVM_KMF_WAITVA); 224 if (lva == 0) { 225 sokvaunreserve(len); 226 return (0); 227 } 228 229 return lva; 230 } 231 232 /* 233 * sokvafree: free kva for loan. 234 */ 235 236 void 237 sokvafree(vaddr_t sva, vsize_t len) 238 { 239 240 /* 241 * free kva. 242 */ 243 244 uvm_km_free(kernel_map, sva, len, UVM_KMF_VAONLY); 245 246 /* 247 * unreserve kva. 248 */ 249 250 sokvaunreserve(len); 251 } 252 253 static void 254 sodoloanfree(struct vm_page **pgs, void *buf, size_t size) 255 { 256 vaddr_t sva, eva; 257 vsize_t len; 258 int npgs; 259 260 KASSERT(pgs != NULL); 261 262 eva = round_page((vaddr_t) buf + size); 263 sva = trunc_page((vaddr_t) buf); 264 len = eva - sva; 265 npgs = len >> PAGE_SHIFT; 266 267 pmap_kremove(sva, len); 268 pmap_update(pmap_kernel()); 269 uvm_unloan(pgs, npgs, UVM_LOAN_TOPAGE); 270 sokvafree(sva, len); 271 } 272 273 /* 274 * sopendfree_thread: free mbufs on "pendfree" list. 275 * unlock and relock so_pendfree_lock when freeing mbufs. 276 */ 277 278 static void 279 sopendfree_thread(void *v) 280 { 281 struct mbuf *m, *next; 282 size_t rv; 283 284 mutex_enter(&so_pendfree_lock); 285 286 for (;;) { 287 rv = 0; 288 while (so_pendfree != NULL) { 289 m = so_pendfree; 290 so_pendfree = NULL; 291 mutex_exit(&so_pendfree_lock); 292 293 for (; m != NULL; m = next) { 294 next = m->m_next; 295 KASSERT((~m->m_flags & (M_EXT|M_EXT_PAGES)) == 0); 296 KASSERT(m->m_ext.ext_refcnt == 0); 297 298 rv += m->m_ext.ext_size; 299 sodoloanfree(m->m_ext.ext_pgs, m->m_ext.ext_buf, 300 m->m_ext.ext_size); 301 pool_cache_put(mb_cache, m); 302 } 303 304 mutex_enter(&so_pendfree_lock); 305 } 306 if (rv) 307 cv_broadcast(&socurkva_cv); 308 cv_wait(&pendfree_thread_cv, &so_pendfree_lock); 309 } 310 panic("sopendfree_thread"); 311 /* NOTREACHED */ 312 } 313 314 void 315 soloanfree(struct mbuf *m, void *buf, size_t size, void *arg) 316 { 317 318 KASSERT(m != NULL); 319 320 /* 321 * postpone freeing mbuf. 322 * 323 * we can't do it in interrupt context 324 * because we need to put kva back to kernel_map. 325 */ 326 327 mutex_enter(&so_pendfree_lock); 328 m->m_next = so_pendfree; 329 so_pendfree = m; 330 cv_signal(&pendfree_thread_cv); 331 mutex_exit(&so_pendfree_lock); 332 } 333 334 static long 335 sosend_loan(struct socket *so, struct uio *uio, struct mbuf *m, long space) 336 { 337 struct iovec *iov = uio->uio_iov; 338 vaddr_t sva, eva; 339 vsize_t len; 340 vaddr_t lva; 341 int npgs, error; 342 vaddr_t va; 343 int i; 344 345 if (VMSPACE_IS_KERNEL_P(uio->uio_vmspace)) 346 return (0); 347 348 if (iov->iov_len < (size_t) space) 349 space = iov->iov_len; 350 if (space > SOCK_LOAN_CHUNK) 351 space = SOCK_LOAN_CHUNK; 352 353 eva = round_page((vaddr_t) iov->iov_base + space); 354 sva = trunc_page((vaddr_t) iov->iov_base); 355 len = eva - sva; 356 npgs = len >> PAGE_SHIFT; 357 358 KASSERT(npgs <= M_EXT_MAXPAGES); 359 360 lva = sokvaalloc(sva, len, so); 361 if (lva == 0) 362 return 0; 363 364 error = uvm_loan(&uio->uio_vmspace->vm_map, sva, len, 365 m->m_ext.ext_pgs, UVM_LOAN_TOPAGE); 366 if (error) { 367 sokvafree(lva, len); 368 return (0); 369 } 370 371 for (i = 0, va = lva; i < npgs; i++, va += PAGE_SIZE) 372 pmap_kenter_pa(va, VM_PAGE_TO_PHYS(m->m_ext.ext_pgs[i]), 373 VM_PROT_READ, 0); 374 pmap_update(pmap_kernel()); 375 376 lva += (vaddr_t) iov->iov_base & PAGE_MASK; 377 378 MEXTADD(m, (void *) lva, space, M_MBUF, soloanfree, so); 379 m->m_flags |= M_EXT_PAGES | M_EXT_ROMAP; 380 381 uio->uio_resid -= space; 382 /* uio_offset not updated, not set/used for write(2) */ 383 uio->uio_iov->iov_base = (char *)uio->uio_iov->iov_base + space; 384 uio->uio_iov->iov_len -= space; 385 if (uio->uio_iov->iov_len == 0) { 386 uio->uio_iov++; 387 uio->uio_iovcnt--; 388 } 389 390 return (space); 391 } 392 393 struct mbuf * 394 getsombuf(struct socket *so, int type) 395 { 396 struct mbuf *m; 397 398 m = m_get(M_WAIT, type); 399 MCLAIM(m, so->so_mowner); 400 return m; 401 } 402 403 static int 404 socket_listener_cb(kauth_cred_t cred, kauth_action_t action, void *cookie, 405 void *arg0, void *arg1, void *arg2, void *arg3) 406 { 407 int result; 408 enum kauth_network_req req; 409 410 result = KAUTH_RESULT_DEFER; 411 req = (enum kauth_network_req)arg0; 412 413 if ((action != KAUTH_NETWORK_SOCKET) && 414 (action != KAUTH_NETWORK_BIND)) 415 return result; 416 417 switch (req) { 418 case KAUTH_REQ_NETWORK_BIND_PORT: 419 result = KAUTH_RESULT_ALLOW; 420 break; 421 422 case KAUTH_REQ_NETWORK_SOCKET_DROP: { 423 /* Normal users can only drop their own connections. */ 424 struct socket *so = (struct socket *)arg1; 425 426 if (so->so_cred && proc_uidmatch(cred, so->so_cred) == 0) 427 result = KAUTH_RESULT_ALLOW; 428 429 break; 430 } 431 432 case KAUTH_REQ_NETWORK_SOCKET_OPEN: 433 /* We allow "raw" routing/bluetooth sockets to anyone. */ 434 if ((u_long)arg1 == PF_ROUTE || (u_long)arg1 == PF_OROUTE 435 || (u_long)arg1 == PF_BLUETOOTH) { 436 result = KAUTH_RESULT_ALLOW; 437 } else { 438 /* Privileged, let secmodel handle this. */ 439 if ((u_long)arg2 == SOCK_RAW) 440 break; 441 } 442 443 result = KAUTH_RESULT_ALLOW; 444 445 break; 446 447 case KAUTH_REQ_NETWORK_SOCKET_CANSEE: 448 result = KAUTH_RESULT_ALLOW; 449 450 break; 451 452 default: 453 break; 454 } 455 456 return result; 457 } 458 459 void 460 soinit(void) 461 { 462 463 sysctl_kern_socket_setup(); 464 465 mutex_init(&so_pendfree_lock, MUTEX_DEFAULT, IPL_VM); 466 softnet_lock = mutex_obj_alloc(MUTEX_DEFAULT, IPL_NONE); 467 cv_init(&socurkva_cv, "sokva"); 468 cv_init(&pendfree_thread_cv, "sopendfr"); 469 soinit2(); 470 471 /* Set the initial adjusted socket buffer size. */ 472 if (sb_max_set(sb_max)) 473 panic("bad initial sb_max value: %lu", sb_max); 474 475 socket_listener = kauth_listen_scope(KAUTH_SCOPE_NETWORK, 476 socket_listener_cb, NULL); 477 } 478 479 void 480 soinit1(void) 481 { 482 int error = kthread_create(PRI_NONE, KTHREAD_MPSAFE, NULL, 483 sopendfree_thread, NULL, &sopendfree_lwp, "sopendfree"); 484 if (error) 485 panic("soinit1 %d", error); 486 } 487 488 /* 489 * socreate: create a new socket of the specified type and the protocol. 490 * 491 * => Caller may specify another socket for lock sharing (must not be held). 492 * => Returns the new socket without lock held. 493 */ 494 int 495 socreate(int dom, struct socket **aso, int type, int proto, struct lwp *l, 496 struct socket *lockso) 497 { 498 const struct protosw *prp; 499 struct socket *so; 500 uid_t uid; 501 int error; 502 kmutex_t *lock; 503 504 error = kauth_authorize_network(l->l_cred, KAUTH_NETWORK_SOCKET, 505 KAUTH_REQ_NETWORK_SOCKET_OPEN, KAUTH_ARG(dom), KAUTH_ARG(type), 506 KAUTH_ARG(proto)); 507 if (error != 0) 508 return error; 509 510 if (proto) 511 prp = pffindproto(dom, proto, type); 512 else 513 prp = pffindtype(dom, type); 514 if (prp == NULL) { 515 /* no support for domain */ 516 if (pffinddomain(dom) == 0) 517 return EAFNOSUPPORT; 518 /* no support for socket type */ 519 if (proto == 0 && type != 0) 520 return EPROTOTYPE; 521 return EPROTONOSUPPORT; 522 } 523 if (prp->pr_usrreqs == NULL) 524 return EPROTONOSUPPORT; 525 if (prp->pr_type != type) 526 return EPROTOTYPE; 527 528 so = soget(true); 529 so->so_type = type; 530 so->so_proto = prp; 531 so->so_send = sosend; 532 so->so_receive = soreceive; 533 #ifdef MBUFTRACE 534 so->so_rcv.sb_mowner = &prp->pr_domain->dom_mowner; 535 so->so_snd.sb_mowner = &prp->pr_domain->dom_mowner; 536 so->so_mowner = &prp->pr_domain->dom_mowner; 537 #endif 538 uid = kauth_cred_geteuid(l->l_cred); 539 so->so_uidinfo = uid_find(uid); 540 so->so_cpid = l->l_proc->p_pid; 541 542 /* 543 * Lock assigned and taken during PCB attach, unless we share 544 * the lock with another socket, e.g. socketpair(2) case. 545 */ 546 if (lockso) { 547 lock = lockso->so_lock; 548 so->so_lock = lock; 549 mutex_obj_hold(lock); 550 mutex_enter(lock); 551 } 552 553 /* Attach the PCB (returns with the socket lock held). */ 554 error = (*prp->pr_usrreqs->pr_attach)(so, proto); 555 KASSERT(solocked(so)); 556 557 if (error) { 558 KASSERT(so->so_pcb == NULL); 559 so->so_state |= SS_NOFDREF; 560 sofree(so); 561 return error; 562 } 563 so->so_cred = kauth_cred_dup(l->l_cred); 564 sounlock(so); 565 566 *aso = so; 567 return 0; 568 } 569 570 /* 571 * fsocreate: create a socket and a file descriptor associated with it. 572 * 573 * => On success, write file descriptor to fdout and return zero. 574 * => On failure, return non-zero; *fdout will be undefined. 575 */ 576 int 577 fsocreate(int domain, struct socket **sop, int type, int proto, int *fdout) 578 { 579 lwp_t *l = curlwp; 580 int error, fd, flags; 581 struct socket *so; 582 struct file *fp; 583 584 if ((error = fd_allocfile(&fp, &fd)) != 0) { 585 return error; 586 } 587 flags = type & SOCK_FLAGS_MASK; 588 fd_set_exclose(l, fd, (flags & SOCK_CLOEXEC) != 0); 589 fp->f_flag = FREAD|FWRITE|((flags & SOCK_NONBLOCK) ? FNONBLOCK : 0)| 590 ((flags & SOCK_NOSIGPIPE) ? FNOSIGPIPE : 0); 591 fp->f_type = DTYPE_SOCKET; 592 fp->f_ops = &socketops; 593 594 type &= ~SOCK_FLAGS_MASK; 595 error = socreate(domain, &so, type, proto, l, NULL); 596 if (error) { 597 fd_abort(curproc, fp, fd); 598 return error; 599 } 600 if (flags & SOCK_NONBLOCK) { 601 so->so_state |= SS_NBIO; 602 } 603 fp->f_data = so; 604 fd_affix(curproc, fp, fd); 605 606 if (sop != NULL) { 607 *sop = so; 608 } 609 *fdout = fd; 610 return error; 611 } 612 613 int 614 sofamily(const struct socket *so) 615 { 616 const struct protosw *pr; 617 const struct domain *dom; 618 619 if ((pr = so->so_proto) == NULL) 620 return AF_UNSPEC; 621 if ((dom = pr->pr_domain) == NULL) 622 return AF_UNSPEC; 623 return dom->dom_family; 624 } 625 626 int 627 sobind(struct socket *so, struct mbuf *nam, struct lwp *l) 628 { 629 int error; 630 631 solock(so); 632 error = (*so->so_proto->pr_usrreqs->pr_generic)(so, 633 PRU_BIND, NULL, nam, NULL, l); 634 sounlock(so); 635 return error; 636 } 637 638 int 639 solisten(struct socket *so, int backlog, struct lwp *l) 640 { 641 int error; 642 643 solock(so); 644 if ((so->so_state & (SS_ISCONNECTED | SS_ISCONNECTING | 645 SS_ISDISCONNECTING)) != 0) { 646 sounlock(so); 647 return EINVAL; 648 } 649 error = (*so->so_proto->pr_usrreqs->pr_generic)(so, 650 PRU_LISTEN, NULL, NULL, NULL, l); 651 if (error != 0) { 652 sounlock(so); 653 return error; 654 } 655 if (TAILQ_EMPTY(&so->so_q)) 656 so->so_options |= SO_ACCEPTCONN; 657 if (backlog < 0) 658 backlog = 0; 659 so->so_qlimit = min(backlog, somaxconn); 660 sounlock(so); 661 return 0; 662 } 663 664 void 665 sofree(struct socket *so) 666 { 667 u_int refs; 668 669 KASSERT(solocked(so)); 670 671 if (so->so_pcb || (so->so_state & SS_NOFDREF) == 0) { 672 sounlock(so); 673 return; 674 } 675 if (so->so_head) { 676 /* 677 * We must not decommission a socket that's on the accept(2) 678 * queue. If we do, then accept(2) may hang after select(2) 679 * indicated that the listening socket was ready. 680 */ 681 if (!soqremque(so, 0)) { 682 sounlock(so); 683 return; 684 } 685 } 686 if (so->so_rcv.sb_hiwat) 687 (void)chgsbsize(so->so_uidinfo, &so->so_rcv.sb_hiwat, 0, 688 RLIM_INFINITY); 689 if (so->so_snd.sb_hiwat) 690 (void)chgsbsize(so->so_uidinfo, &so->so_snd.sb_hiwat, 0, 691 RLIM_INFINITY); 692 sbrelease(&so->so_snd, so); 693 KASSERT(!cv_has_waiters(&so->so_cv)); 694 KASSERT(!cv_has_waiters(&so->so_rcv.sb_cv)); 695 KASSERT(!cv_has_waiters(&so->so_snd.sb_cv)); 696 sorflush(so); 697 refs = so->so_aborting; /* XXX */ 698 /* Remove acccept filter if one is present. */ 699 if (so->so_accf != NULL) 700 (void)accept_filt_clear(so); 701 sounlock(so); 702 if (refs == 0) /* XXX */ 703 soput(so); 704 } 705 706 /* 707 * soclose: close a socket on last file table reference removal. 708 * Initiate disconnect if connected. Free socket when disconnect complete. 709 */ 710 int 711 soclose(struct socket *so) 712 { 713 struct socket *so2; 714 int error = 0; 715 716 solock(so); 717 if (so->so_options & SO_ACCEPTCONN) { 718 for (;;) { 719 if ((so2 = TAILQ_FIRST(&so->so_q0)) != 0) { 720 KASSERT(solocked2(so, so2)); 721 (void) soqremque(so2, 0); 722 /* soabort drops the lock. */ 723 (void) soabort(so2); 724 solock(so); 725 continue; 726 } 727 if ((so2 = TAILQ_FIRST(&so->so_q)) != 0) { 728 KASSERT(solocked2(so, so2)); 729 (void) soqremque(so2, 1); 730 /* soabort drops the lock. */ 731 (void) soabort(so2); 732 solock(so); 733 continue; 734 } 735 break; 736 } 737 } 738 if (so->so_pcb == NULL) 739 goto discard; 740 if (so->so_state & SS_ISCONNECTED) { 741 if ((so->so_state & SS_ISDISCONNECTING) == 0) { 742 error = sodisconnect(so); 743 if (error) 744 goto drop; 745 } 746 if (so->so_options & SO_LINGER) { 747 if ((so->so_state & (SS_ISDISCONNECTING|SS_NBIO)) == 748 (SS_ISDISCONNECTING|SS_NBIO)) 749 goto drop; 750 while (so->so_state & SS_ISCONNECTED) { 751 error = sowait(so, true, so->so_linger * hz); 752 if (error) 753 break; 754 } 755 } 756 } 757 drop: 758 if (so->so_pcb) { 759 KASSERT(solocked(so)); 760 (*so->so_proto->pr_usrreqs->pr_detach)(so); 761 } 762 discard: 763 KASSERT((so->so_state & SS_NOFDREF) == 0); 764 kauth_cred_free(so->so_cred); 765 so->so_state |= SS_NOFDREF; 766 sofree(so); 767 return error; 768 } 769 770 /* 771 * Must be called with the socket locked.. Will return with it unlocked. 772 */ 773 int 774 soabort(struct socket *so) 775 { 776 u_int refs; 777 int error; 778 779 KASSERT(solocked(so)); 780 KASSERT(so->so_head == NULL); 781 782 so->so_aborting++; /* XXX */ 783 error = (*so->so_proto->pr_usrreqs->pr_generic)(so, 784 PRU_ABORT, NULL, NULL, NULL, NULL); 785 refs = --so->so_aborting; /* XXX */ 786 if (error || (refs == 0)) { 787 sofree(so); 788 } else { 789 sounlock(so); 790 } 791 return error; 792 } 793 794 int 795 soaccept(struct socket *so, struct mbuf *nam) 796 { 797 int error; 798 799 KASSERT(solocked(so)); 800 KASSERT((so->so_state & SS_NOFDREF) != 0); 801 802 so->so_state &= ~SS_NOFDREF; 803 if ((so->so_state & SS_ISDISCONNECTED) == 0 || 804 (so->so_proto->pr_flags & PR_ABRTACPTDIS) == 0) 805 error = (*so->so_proto->pr_usrreqs->pr_accept)(so, nam); 806 else 807 error = ECONNABORTED; 808 809 return error; 810 } 811 812 int 813 soconnect(struct socket *so, struct mbuf *nam, struct lwp *l) 814 { 815 int error; 816 817 KASSERT(solocked(so)); 818 819 if (so->so_options & SO_ACCEPTCONN) 820 return EOPNOTSUPP; 821 /* 822 * If protocol is connection-based, can only connect once. 823 * Otherwise, if connected, try to disconnect first. 824 * This allows user to disconnect by connecting to, e.g., 825 * a null address. 826 */ 827 if (so->so_state & (SS_ISCONNECTED|SS_ISCONNECTING) && 828 ((so->so_proto->pr_flags & PR_CONNREQUIRED) || 829 (error = sodisconnect(so)))) 830 error = EISCONN; 831 else 832 error = (*so->so_proto->pr_usrreqs->pr_generic)(so, 833 PRU_CONNECT, NULL, nam, NULL, l); 834 835 return error; 836 } 837 838 int 839 soconnect2(struct socket *so1, struct socket *so2) 840 { 841 KASSERT(solocked2(so1, so2)); 842 843 return (*so1->so_proto->pr_usrreqs->pr_generic)(so1, 844 PRU_CONNECT2, NULL, (struct mbuf *)so2, NULL, NULL); 845 } 846 847 int 848 sodisconnect(struct socket *so) 849 { 850 int error; 851 852 KASSERT(solocked(so)); 853 854 if ((so->so_state & SS_ISCONNECTED) == 0) { 855 error = ENOTCONN; 856 } else if (so->so_state & SS_ISDISCONNECTING) { 857 error = EALREADY; 858 } else { 859 error = (*so->so_proto->pr_usrreqs->pr_generic)(so, 860 PRU_DISCONNECT, NULL, NULL, NULL, NULL); 861 } 862 return (error); 863 } 864 865 #define SBLOCKWAIT(f) (((f) & MSG_DONTWAIT) ? M_NOWAIT : M_WAITOK) 866 /* 867 * Send on a socket. 868 * If send must go all at once and message is larger than 869 * send buffering, then hard error. 870 * Lock against other senders. 871 * If must go all at once and not enough room now, then 872 * inform user that this would block and do nothing. 873 * Otherwise, if nonblocking, send as much as possible. 874 * The data to be sent is described by "uio" if nonzero, 875 * otherwise by the mbuf chain "top" (which must be null 876 * if uio is not). Data provided in mbuf chain must be small 877 * enough to send all at once. 878 * 879 * Returns nonzero on error, timeout or signal; callers 880 * must check for short counts if EINTR/ERESTART are returned. 881 * Data and control buffers are freed on return. 882 */ 883 int 884 sosend(struct socket *so, struct mbuf *addr, struct uio *uio, struct mbuf *top, 885 struct mbuf *control, int flags, struct lwp *l) 886 { 887 struct mbuf **mp, *m; 888 long space, len, resid, clen, mlen; 889 int error, s, dontroute, atomic; 890 short wakeup_state = 0; 891 892 clen = 0; 893 894 /* 895 * solock() provides atomicity of access. splsoftnet() prevents 896 * protocol processing soft interrupts from interrupting us and 897 * blocking (expensive). 898 */ 899 s = splsoftnet(); 900 solock(so); 901 atomic = sosendallatonce(so) || top; 902 if (uio) 903 resid = uio->uio_resid; 904 else 905 resid = top->m_pkthdr.len; 906 /* 907 * In theory resid should be unsigned. 908 * However, space must be signed, as it might be less than 0 909 * if we over-committed, and we must use a signed comparison 910 * of space and resid. On the other hand, a negative resid 911 * causes us to loop sending 0-length segments to the protocol. 912 */ 913 if (resid < 0) { 914 error = EINVAL; 915 goto out; 916 } 917 dontroute = 918 (flags & MSG_DONTROUTE) && (so->so_options & SO_DONTROUTE) == 0 && 919 (so->so_proto->pr_flags & PR_ATOMIC); 920 l->l_ru.ru_msgsnd++; 921 if (control) 922 clen = control->m_len; 923 restart: 924 if ((error = sblock(&so->so_snd, SBLOCKWAIT(flags))) != 0) 925 goto out; 926 do { 927 if (so->so_state & SS_CANTSENDMORE) { 928 error = EPIPE; 929 goto release; 930 } 931 if (so->so_error) { 932 error = so->so_error; 933 so->so_error = 0; 934 goto release; 935 } 936 if ((so->so_state & SS_ISCONNECTED) == 0) { 937 if (so->so_proto->pr_flags & PR_CONNREQUIRED) { 938 if (resid || clen == 0) { 939 error = ENOTCONN; 940 goto release; 941 } 942 } else if (addr == 0) { 943 error = EDESTADDRREQ; 944 goto release; 945 } 946 } 947 space = sbspace(&so->so_snd); 948 if (flags & MSG_OOB) 949 space += 1024; 950 if ((atomic && resid > so->so_snd.sb_hiwat) || 951 clen > so->so_snd.sb_hiwat) { 952 error = EMSGSIZE; 953 goto release; 954 } 955 if (space < resid + clen && 956 (atomic || space < so->so_snd.sb_lowat || space < clen)) { 957 if ((so->so_state & SS_NBIO) || (flags & MSG_NBIO)) { 958 error = EWOULDBLOCK; 959 goto release; 960 } 961 sbunlock(&so->so_snd); 962 if (wakeup_state & SS_RESTARTSYS) { 963 error = ERESTART; 964 goto out; 965 } 966 error = sbwait(&so->so_snd); 967 if (error) 968 goto out; 969 wakeup_state = so->so_state; 970 goto restart; 971 } 972 wakeup_state = 0; 973 mp = ⊤ 974 space -= clen; 975 do { 976 if (uio == NULL) { 977 /* 978 * Data is prepackaged in "top". 979 */ 980 resid = 0; 981 if (flags & MSG_EOR) 982 top->m_flags |= M_EOR; 983 } else do { 984 sounlock(so); 985 splx(s); 986 if (top == NULL) { 987 m = m_gethdr(M_WAIT, MT_DATA); 988 mlen = MHLEN; 989 m->m_pkthdr.len = 0; 990 m->m_pkthdr.rcvif = NULL; 991 } else { 992 m = m_get(M_WAIT, MT_DATA); 993 mlen = MLEN; 994 } 995 MCLAIM(m, so->so_snd.sb_mowner); 996 if (sock_loan_thresh >= 0 && 997 uio->uio_iov->iov_len >= sock_loan_thresh && 998 space >= sock_loan_thresh && 999 (len = sosend_loan(so, uio, m, 1000 space)) != 0) { 1001 SOSEND_COUNTER_INCR(&sosend_loan_big); 1002 space -= len; 1003 goto have_data; 1004 } 1005 if (resid >= MINCLSIZE && space >= MCLBYTES) { 1006 SOSEND_COUNTER_INCR(&sosend_copy_big); 1007 m_clget(m, M_DONTWAIT); 1008 if ((m->m_flags & M_EXT) == 0) 1009 goto nopages; 1010 mlen = MCLBYTES; 1011 if (atomic && top == 0) { 1012 len = lmin(MCLBYTES - max_hdr, 1013 resid); 1014 m->m_data += max_hdr; 1015 } else 1016 len = lmin(MCLBYTES, resid); 1017 space -= len; 1018 } else { 1019 nopages: 1020 SOSEND_COUNTER_INCR(&sosend_copy_small); 1021 len = lmin(lmin(mlen, resid), space); 1022 space -= len; 1023 /* 1024 * For datagram protocols, leave room 1025 * for protocol headers in first mbuf. 1026 */ 1027 if (atomic && top == 0 && len < mlen) 1028 MH_ALIGN(m, len); 1029 } 1030 error = uiomove(mtod(m, void *), (int)len, uio); 1031 have_data: 1032 resid = uio->uio_resid; 1033 m->m_len = len; 1034 *mp = m; 1035 top->m_pkthdr.len += len; 1036 s = splsoftnet(); 1037 solock(so); 1038 if (error != 0) 1039 goto release; 1040 mp = &m->m_next; 1041 if (resid <= 0) { 1042 if (flags & MSG_EOR) 1043 top->m_flags |= M_EOR; 1044 break; 1045 } 1046 } while (space > 0 && atomic); 1047 1048 if (so->so_state & SS_CANTSENDMORE) { 1049 error = EPIPE; 1050 goto release; 1051 } 1052 if (dontroute) 1053 so->so_options |= SO_DONTROUTE; 1054 if (resid > 0) 1055 so->so_state |= SS_MORETOCOME; 1056 error = (*so->so_proto->pr_usrreqs->pr_generic)(so, 1057 (flags & MSG_OOB) ? PRU_SENDOOB : PRU_SEND, 1058 top, addr, control, curlwp); 1059 if (dontroute) 1060 so->so_options &= ~SO_DONTROUTE; 1061 if (resid > 0) 1062 so->so_state &= ~SS_MORETOCOME; 1063 clen = 0; 1064 control = NULL; 1065 top = NULL; 1066 mp = ⊤ 1067 if (error != 0) 1068 goto release; 1069 } while (resid && space > 0); 1070 } while (resid); 1071 1072 release: 1073 sbunlock(&so->so_snd); 1074 out: 1075 sounlock(so); 1076 splx(s); 1077 if (top) 1078 m_freem(top); 1079 if (control) 1080 m_freem(control); 1081 return (error); 1082 } 1083 1084 /* 1085 * Following replacement or removal of the first mbuf on the first 1086 * mbuf chain of a socket buffer, push necessary state changes back 1087 * into the socket buffer so that other consumers see the values 1088 * consistently. 'nextrecord' is the callers locally stored value of 1089 * the original value of sb->sb_mb->m_nextpkt which must be restored 1090 * when the lead mbuf changes. NOTE: 'nextrecord' may be NULL. 1091 */ 1092 static void 1093 sbsync(struct sockbuf *sb, struct mbuf *nextrecord) 1094 { 1095 1096 KASSERT(solocked(sb->sb_so)); 1097 1098 /* 1099 * First, update for the new value of nextrecord. If necessary, 1100 * make it the first record. 1101 */ 1102 if (sb->sb_mb != NULL) 1103 sb->sb_mb->m_nextpkt = nextrecord; 1104 else 1105 sb->sb_mb = nextrecord; 1106 1107 /* 1108 * Now update any dependent socket buffer fields to reflect 1109 * the new state. This is an inline of SB_EMPTY_FIXUP, with 1110 * the addition of a second clause that takes care of the 1111 * case where sb_mb has been updated, but remains the last 1112 * record. 1113 */ 1114 if (sb->sb_mb == NULL) { 1115 sb->sb_mbtail = NULL; 1116 sb->sb_lastrecord = NULL; 1117 } else if (sb->sb_mb->m_nextpkt == NULL) 1118 sb->sb_lastrecord = sb->sb_mb; 1119 } 1120 1121 /* 1122 * Implement receive operations on a socket. 1123 * We depend on the way that records are added to the sockbuf 1124 * by sbappend*. In particular, each record (mbufs linked through m_next) 1125 * must begin with an address if the protocol so specifies, 1126 * followed by an optional mbuf or mbufs containing ancillary data, 1127 * and then zero or more mbufs of data. 1128 * In order to avoid blocking network interrupts for the entire time here, 1129 * we splx() while doing the actual copy to user space. 1130 * Although the sockbuf is locked, new data may still be appended, 1131 * and thus we must maintain consistency of the sockbuf during that time. 1132 * 1133 * The caller may receive the data as a single mbuf chain by supplying 1134 * an mbuf **mp0 for use in returning the chain. The uio is then used 1135 * only for the count in uio_resid. 1136 */ 1137 int 1138 soreceive(struct socket *so, struct mbuf **paddr, struct uio *uio, 1139 struct mbuf **mp0, struct mbuf **controlp, int *flagsp) 1140 { 1141 struct lwp *l = curlwp; 1142 struct mbuf *m, **mp, *mt; 1143 size_t len, offset, moff, orig_resid; 1144 int atomic, flags, error, s, type; 1145 const struct protosw *pr; 1146 struct mbuf *nextrecord; 1147 int mbuf_removed = 0; 1148 const struct domain *dom; 1149 short wakeup_state = 0; 1150 1151 pr = so->so_proto; 1152 atomic = pr->pr_flags & PR_ATOMIC; 1153 dom = pr->pr_domain; 1154 mp = mp0; 1155 type = 0; 1156 orig_resid = uio->uio_resid; 1157 1158 if (paddr != NULL) 1159 *paddr = NULL; 1160 if (controlp != NULL) 1161 *controlp = NULL; 1162 if (flagsp != NULL) 1163 flags = *flagsp &~ MSG_EOR; 1164 else 1165 flags = 0; 1166 1167 if (flags & MSG_OOB) { 1168 m = m_get(M_WAIT, MT_DATA); 1169 solock(so); 1170 error = (*pr->pr_usrreqs->pr_generic)(so, PRU_RCVOOB, m, 1171 (struct mbuf *)(long)(flags & MSG_PEEK), NULL, l); 1172 sounlock(so); 1173 if (error) 1174 goto bad; 1175 do { 1176 error = uiomove(mtod(m, void *), 1177 MIN(uio->uio_resid, m->m_len), uio); 1178 m = m_free(m); 1179 } while (uio->uio_resid > 0 && error == 0 && m); 1180 bad: 1181 if (m != NULL) 1182 m_freem(m); 1183 return error; 1184 } 1185 if (mp != NULL) 1186 *mp = NULL; 1187 1188 /* 1189 * solock() provides atomicity of access. splsoftnet() prevents 1190 * protocol processing soft interrupts from interrupting us and 1191 * blocking (expensive). 1192 */ 1193 s = splsoftnet(); 1194 solock(so); 1195 restart: 1196 if ((error = sblock(&so->so_rcv, SBLOCKWAIT(flags))) != 0) { 1197 sounlock(so); 1198 splx(s); 1199 return error; 1200 } 1201 1202 m = so->so_rcv.sb_mb; 1203 /* 1204 * If we have less data than requested, block awaiting more 1205 * (subject to any timeout) if: 1206 * 1. the current count is less than the low water mark, 1207 * 2. MSG_WAITALL is set, and it is possible to do the entire 1208 * receive operation at once if we block (resid <= hiwat), or 1209 * 3. MSG_DONTWAIT is not set. 1210 * If MSG_WAITALL is set but resid is larger than the receive buffer, 1211 * we have to do the receive in sections, and thus risk returning 1212 * a short count if a timeout or signal occurs after we start. 1213 */ 1214 if (m == NULL || 1215 ((flags & MSG_DONTWAIT) == 0 && 1216 so->so_rcv.sb_cc < uio->uio_resid && 1217 (so->so_rcv.sb_cc < so->so_rcv.sb_lowat || 1218 ((flags & MSG_WAITALL) && 1219 uio->uio_resid <= so->so_rcv.sb_hiwat)) && 1220 m->m_nextpkt == NULL && !atomic)) { 1221 #ifdef DIAGNOSTIC 1222 if (m == NULL && so->so_rcv.sb_cc) 1223 panic("receive 1"); 1224 #endif 1225 if (so->so_error) { 1226 if (m != NULL) 1227 goto dontblock; 1228 error = so->so_error; 1229 if ((flags & MSG_PEEK) == 0) 1230 so->so_error = 0; 1231 goto release; 1232 } 1233 if (so->so_state & SS_CANTRCVMORE) { 1234 if (m != NULL) 1235 goto dontblock; 1236 else 1237 goto release; 1238 } 1239 for (; m != NULL; m = m->m_next) 1240 if (m->m_type == MT_OOBDATA || (m->m_flags & M_EOR)) { 1241 m = so->so_rcv.sb_mb; 1242 goto dontblock; 1243 } 1244 if ((so->so_state & (SS_ISCONNECTED|SS_ISCONNECTING)) == 0 && 1245 (so->so_proto->pr_flags & PR_CONNREQUIRED)) { 1246 error = ENOTCONN; 1247 goto release; 1248 } 1249 if (uio->uio_resid == 0) 1250 goto release; 1251 if ((so->so_state & SS_NBIO) || 1252 (flags & (MSG_DONTWAIT|MSG_NBIO))) { 1253 error = EWOULDBLOCK; 1254 goto release; 1255 } 1256 SBLASTRECORDCHK(&so->so_rcv, "soreceive sbwait 1"); 1257 SBLASTMBUFCHK(&so->so_rcv, "soreceive sbwait 1"); 1258 sbunlock(&so->so_rcv); 1259 if (wakeup_state & SS_RESTARTSYS) 1260 error = ERESTART; 1261 else 1262 error = sbwait(&so->so_rcv); 1263 if (error != 0) { 1264 sounlock(so); 1265 splx(s); 1266 return error; 1267 } 1268 wakeup_state = so->so_state; 1269 goto restart; 1270 } 1271 dontblock: 1272 /* 1273 * On entry here, m points to the first record of the socket buffer. 1274 * From this point onward, we maintain 'nextrecord' as a cache of the 1275 * pointer to the next record in the socket buffer. We must keep the 1276 * various socket buffer pointers and local stack versions of the 1277 * pointers in sync, pushing out modifications before dropping the 1278 * socket lock, and re-reading them when picking it up. 1279 * 1280 * Otherwise, we will race with the network stack appending new data 1281 * or records onto the socket buffer by using inconsistent/stale 1282 * versions of the field, possibly resulting in socket buffer 1283 * corruption. 1284 * 1285 * By holding the high-level sblock(), we prevent simultaneous 1286 * readers from pulling off the front of the socket buffer. 1287 */ 1288 if (l != NULL) 1289 l->l_ru.ru_msgrcv++; 1290 KASSERT(m == so->so_rcv.sb_mb); 1291 SBLASTRECORDCHK(&so->so_rcv, "soreceive 1"); 1292 SBLASTMBUFCHK(&so->so_rcv, "soreceive 1"); 1293 nextrecord = m->m_nextpkt; 1294 if (pr->pr_flags & PR_ADDR) { 1295 #ifdef DIAGNOSTIC 1296 if (m->m_type != MT_SONAME) 1297 panic("receive 1a"); 1298 #endif 1299 orig_resid = 0; 1300 if (flags & MSG_PEEK) { 1301 if (paddr) 1302 *paddr = m_copy(m, 0, m->m_len); 1303 m = m->m_next; 1304 } else { 1305 sbfree(&so->so_rcv, m); 1306 mbuf_removed = 1; 1307 if (paddr != NULL) { 1308 *paddr = m; 1309 so->so_rcv.sb_mb = m->m_next; 1310 m->m_next = NULL; 1311 m = so->so_rcv.sb_mb; 1312 } else { 1313 MFREE(m, so->so_rcv.sb_mb); 1314 m = so->so_rcv.sb_mb; 1315 } 1316 sbsync(&so->so_rcv, nextrecord); 1317 } 1318 } 1319 1320 /* 1321 * Process one or more MT_CONTROL mbufs present before any data mbufs 1322 * in the first mbuf chain on the socket buffer. If MSG_PEEK, we 1323 * just copy the data; if !MSG_PEEK, we call into the protocol to 1324 * perform externalization (or freeing if controlp == NULL). 1325 */ 1326 if (__predict_false(m != NULL && m->m_type == MT_CONTROL)) { 1327 struct mbuf *cm = NULL, *cmn; 1328 struct mbuf **cme = &cm; 1329 1330 do { 1331 if (flags & MSG_PEEK) { 1332 if (controlp != NULL) { 1333 *controlp = m_copy(m, 0, m->m_len); 1334 controlp = &(*controlp)->m_next; 1335 } 1336 m = m->m_next; 1337 } else { 1338 sbfree(&so->so_rcv, m); 1339 so->so_rcv.sb_mb = m->m_next; 1340 m->m_next = NULL; 1341 *cme = m; 1342 cme = &(*cme)->m_next; 1343 m = so->so_rcv.sb_mb; 1344 } 1345 } while (m != NULL && m->m_type == MT_CONTROL); 1346 if ((flags & MSG_PEEK) == 0) 1347 sbsync(&so->so_rcv, nextrecord); 1348 for (; cm != NULL; cm = cmn) { 1349 cmn = cm->m_next; 1350 cm->m_next = NULL; 1351 type = mtod(cm, struct cmsghdr *)->cmsg_type; 1352 if (controlp != NULL) { 1353 if (dom->dom_externalize != NULL && 1354 type == SCM_RIGHTS) { 1355 sounlock(so); 1356 splx(s); 1357 error = (*dom->dom_externalize)(cm, l, 1358 (flags & MSG_CMSG_CLOEXEC) ? 1359 O_CLOEXEC : 0); 1360 s = splsoftnet(); 1361 solock(so); 1362 } 1363 *controlp = cm; 1364 while (*controlp != NULL) 1365 controlp = &(*controlp)->m_next; 1366 } else { 1367 /* 1368 * Dispose of any SCM_RIGHTS message that went 1369 * through the read path rather than recv. 1370 */ 1371 if (dom->dom_dispose != NULL && 1372 type == SCM_RIGHTS) { 1373 sounlock(so); 1374 (*dom->dom_dispose)(cm); 1375 solock(so); 1376 } 1377 m_freem(cm); 1378 } 1379 } 1380 if (m != NULL) 1381 nextrecord = so->so_rcv.sb_mb->m_nextpkt; 1382 else 1383 nextrecord = so->so_rcv.sb_mb; 1384 orig_resid = 0; 1385 } 1386 1387 /* If m is non-NULL, we have some data to read. */ 1388 if (__predict_true(m != NULL)) { 1389 type = m->m_type; 1390 if (type == MT_OOBDATA) 1391 flags |= MSG_OOB; 1392 } 1393 SBLASTRECORDCHK(&so->so_rcv, "soreceive 2"); 1394 SBLASTMBUFCHK(&so->so_rcv, "soreceive 2"); 1395 1396 moff = 0; 1397 offset = 0; 1398 while (m != NULL && uio->uio_resid > 0 && error == 0) { 1399 if (m->m_type == MT_OOBDATA) { 1400 if (type != MT_OOBDATA) 1401 break; 1402 } else if (type == MT_OOBDATA) 1403 break; 1404 #ifdef DIAGNOSTIC 1405 else if (m->m_type != MT_DATA && m->m_type != MT_HEADER) 1406 panic("receive 3"); 1407 #endif 1408 so->so_state &= ~SS_RCVATMARK; 1409 wakeup_state = 0; 1410 len = uio->uio_resid; 1411 if (so->so_oobmark && len > so->so_oobmark - offset) 1412 len = so->so_oobmark - offset; 1413 if (len > m->m_len - moff) 1414 len = m->m_len - moff; 1415 /* 1416 * If mp is set, just pass back the mbufs. 1417 * Otherwise copy them out via the uio, then free. 1418 * Sockbuf must be consistent here (points to current mbuf, 1419 * it points to next record) when we drop priority; 1420 * we must note any additions to the sockbuf when we 1421 * block interrupts again. 1422 */ 1423 if (mp == NULL) { 1424 SBLASTRECORDCHK(&so->so_rcv, "soreceive uiomove"); 1425 SBLASTMBUFCHK(&so->so_rcv, "soreceive uiomove"); 1426 sounlock(so); 1427 splx(s); 1428 error = uiomove(mtod(m, char *) + moff, len, uio); 1429 s = splsoftnet(); 1430 solock(so); 1431 if (error != 0) { 1432 /* 1433 * If any part of the record has been removed 1434 * (such as the MT_SONAME mbuf, which will 1435 * happen when PR_ADDR, and thus also 1436 * PR_ATOMIC, is set), then drop the entire 1437 * record to maintain the atomicity of the 1438 * receive operation. 1439 * 1440 * This avoids a later panic("receive 1a") 1441 * when compiled with DIAGNOSTIC. 1442 */ 1443 if (m && mbuf_removed && atomic) 1444 (void) sbdroprecord(&so->so_rcv); 1445 1446 goto release; 1447 } 1448 } else 1449 uio->uio_resid -= len; 1450 if (len == m->m_len - moff) { 1451 if (m->m_flags & M_EOR) 1452 flags |= MSG_EOR; 1453 if (flags & MSG_PEEK) { 1454 m = m->m_next; 1455 moff = 0; 1456 } else { 1457 nextrecord = m->m_nextpkt; 1458 sbfree(&so->so_rcv, m); 1459 if (mp) { 1460 *mp = m; 1461 mp = &m->m_next; 1462 so->so_rcv.sb_mb = m = m->m_next; 1463 *mp = NULL; 1464 } else { 1465 MFREE(m, so->so_rcv.sb_mb); 1466 m = so->so_rcv.sb_mb; 1467 } 1468 /* 1469 * If m != NULL, we also know that 1470 * so->so_rcv.sb_mb != NULL. 1471 */ 1472 KASSERT(so->so_rcv.sb_mb == m); 1473 if (m) { 1474 m->m_nextpkt = nextrecord; 1475 if (nextrecord == NULL) 1476 so->so_rcv.sb_lastrecord = m; 1477 } else { 1478 so->so_rcv.sb_mb = nextrecord; 1479 SB_EMPTY_FIXUP(&so->so_rcv); 1480 } 1481 SBLASTRECORDCHK(&so->so_rcv, "soreceive 3"); 1482 SBLASTMBUFCHK(&so->so_rcv, "soreceive 3"); 1483 } 1484 } else if (flags & MSG_PEEK) 1485 moff += len; 1486 else { 1487 if (mp != NULL) { 1488 mt = m_copym(m, 0, len, M_NOWAIT); 1489 if (__predict_false(mt == NULL)) { 1490 sounlock(so); 1491 mt = m_copym(m, 0, len, M_WAIT); 1492 solock(so); 1493 } 1494 *mp = mt; 1495 } 1496 m->m_data += len; 1497 m->m_len -= len; 1498 so->so_rcv.sb_cc -= len; 1499 } 1500 if (so->so_oobmark) { 1501 if ((flags & MSG_PEEK) == 0) { 1502 so->so_oobmark -= len; 1503 if (so->so_oobmark == 0) { 1504 so->so_state |= SS_RCVATMARK; 1505 break; 1506 } 1507 } else { 1508 offset += len; 1509 if (offset == so->so_oobmark) 1510 break; 1511 } 1512 } 1513 if (flags & MSG_EOR) 1514 break; 1515 /* 1516 * If the MSG_WAITALL flag is set (for non-atomic socket), 1517 * we must not quit until "uio->uio_resid == 0" or an error 1518 * termination. If a signal/timeout occurs, return 1519 * with a short count but without error. 1520 * Keep sockbuf locked against other readers. 1521 */ 1522 while (flags & MSG_WAITALL && m == NULL && uio->uio_resid > 0 && 1523 !sosendallatonce(so) && !nextrecord) { 1524 if (so->so_error || so->so_state & SS_CANTRCVMORE) 1525 break; 1526 /* 1527 * If we are peeking and the socket receive buffer is 1528 * full, stop since we can't get more data to peek at. 1529 */ 1530 if ((flags & MSG_PEEK) && sbspace(&so->so_rcv) <= 0) 1531 break; 1532 /* 1533 * If we've drained the socket buffer, tell the 1534 * protocol in case it needs to do something to 1535 * get it filled again. 1536 */ 1537 if ((pr->pr_flags & PR_WANTRCVD) && so->so_pcb) 1538 (*pr->pr_usrreqs->pr_generic)(so, PRU_RCVD, 1539 NULL, (struct mbuf *)(long)flags, NULL, l); 1540 SBLASTRECORDCHK(&so->so_rcv, "soreceive sbwait 2"); 1541 SBLASTMBUFCHK(&so->so_rcv, "soreceive sbwait 2"); 1542 if (wakeup_state & SS_RESTARTSYS) 1543 error = ERESTART; 1544 else 1545 error = sbwait(&so->so_rcv); 1546 if (error != 0) { 1547 sbunlock(&so->so_rcv); 1548 sounlock(so); 1549 splx(s); 1550 return 0; 1551 } 1552 if ((m = so->so_rcv.sb_mb) != NULL) 1553 nextrecord = m->m_nextpkt; 1554 wakeup_state = so->so_state; 1555 } 1556 } 1557 1558 if (m && atomic) { 1559 flags |= MSG_TRUNC; 1560 if ((flags & MSG_PEEK) == 0) 1561 (void) sbdroprecord(&so->so_rcv); 1562 } 1563 if ((flags & MSG_PEEK) == 0) { 1564 if (m == NULL) { 1565 /* 1566 * First part is an inline SB_EMPTY_FIXUP(). Second 1567 * part makes sure sb_lastrecord is up-to-date if 1568 * there is still data in the socket buffer. 1569 */ 1570 so->so_rcv.sb_mb = nextrecord; 1571 if (so->so_rcv.sb_mb == NULL) { 1572 so->so_rcv.sb_mbtail = NULL; 1573 so->so_rcv.sb_lastrecord = NULL; 1574 } else if (nextrecord->m_nextpkt == NULL) 1575 so->so_rcv.sb_lastrecord = nextrecord; 1576 } 1577 SBLASTRECORDCHK(&so->so_rcv, "soreceive 4"); 1578 SBLASTMBUFCHK(&so->so_rcv, "soreceive 4"); 1579 if (pr->pr_flags & PR_WANTRCVD && so->so_pcb) 1580 (*pr->pr_usrreqs->pr_generic)(so, PRU_RCVD, NULL, 1581 (struct mbuf *)(long)flags, NULL, l); 1582 } 1583 if (orig_resid == uio->uio_resid && orig_resid && 1584 (flags & MSG_EOR) == 0 && (so->so_state & SS_CANTRCVMORE) == 0) { 1585 sbunlock(&so->so_rcv); 1586 goto restart; 1587 } 1588 1589 if (flagsp != NULL) 1590 *flagsp |= flags; 1591 release: 1592 sbunlock(&so->so_rcv); 1593 sounlock(so); 1594 splx(s); 1595 return error; 1596 } 1597 1598 int 1599 soshutdown(struct socket *so, int how) 1600 { 1601 const struct protosw *pr; 1602 int error; 1603 1604 KASSERT(solocked(so)); 1605 1606 pr = so->so_proto; 1607 if (!(how == SHUT_RD || how == SHUT_WR || how == SHUT_RDWR)) 1608 return (EINVAL); 1609 1610 if (how == SHUT_RD || how == SHUT_RDWR) { 1611 sorflush(so); 1612 error = 0; 1613 } 1614 if (how == SHUT_WR || how == SHUT_RDWR) 1615 error = (*pr->pr_usrreqs->pr_generic)(so, 1616 PRU_SHUTDOWN, NULL, NULL, NULL, NULL); 1617 1618 return error; 1619 } 1620 1621 void 1622 sorestart(struct socket *so) 1623 { 1624 /* 1625 * An application has called close() on an fd on which another 1626 * of its threads has called a socket system call. 1627 * Mark this and wake everyone up, and code that would block again 1628 * instead returns ERESTART. 1629 * On system call re-entry the fd is validated and EBADF returned. 1630 * Any other fd will block again on the 2nd syscall. 1631 */ 1632 solock(so); 1633 so->so_state |= SS_RESTARTSYS; 1634 cv_broadcast(&so->so_cv); 1635 cv_broadcast(&so->so_snd.sb_cv); 1636 cv_broadcast(&so->so_rcv.sb_cv); 1637 sounlock(so); 1638 } 1639 1640 void 1641 sorflush(struct socket *so) 1642 { 1643 struct sockbuf *sb, asb; 1644 const struct protosw *pr; 1645 1646 KASSERT(solocked(so)); 1647 1648 sb = &so->so_rcv; 1649 pr = so->so_proto; 1650 socantrcvmore(so); 1651 sb->sb_flags |= SB_NOINTR; 1652 (void )sblock(sb, M_WAITOK); 1653 sbunlock(sb); 1654 asb = *sb; 1655 /* 1656 * Clear most of the sockbuf structure, but leave some of the 1657 * fields valid. 1658 */ 1659 memset(&sb->sb_startzero, 0, 1660 sizeof(*sb) - offsetof(struct sockbuf, sb_startzero)); 1661 if (pr->pr_flags & PR_RIGHTS && pr->pr_domain->dom_dispose) { 1662 sounlock(so); 1663 (*pr->pr_domain->dom_dispose)(asb.sb_mb); 1664 solock(so); 1665 } 1666 sbrelease(&asb, so); 1667 } 1668 1669 /* 1670 * internal set SOL_SOCKET options 1671 */ 1672 static int 1673 sosetopt1(struct socket *so, const struct sockopt *sopt) 1674 { 1675 int error = EINVAL, opt; 1676 int optval = 0; /* XXX: gcc */ 1677 struct linger l; 1678 struct timeval tv; 1679 1680 switch ((opt = sopt->sopt_name)) { 1681 1682 case SO_ACCEPTFILTER: 1683 error = accept_filt_setopt(so, sopt); 1684 KASSERT(solocked(so)); 1685 break; 1686 1687 case SO_LINGER: 1688 error = sockopt_get(sopt, &l, sizeof(l)); 1689 solock(so); 1690 if (error) 1691 break; 1692 if (l.l_linger < 0 || l.l_linger > USHRT_MAX || 1693 l.l_linger > (INT_MAX / hz)) { 1694 error = EDOM; 1695 break; 1696 } 1697 so->so_linger = l.l_linger; 1698 if (l.l_onoff) 1699 so->so_options |= SO_LINGER; 1700 else 1701 so->so_options &= ~SO_LINGER; 1702 break; 1703 1704 case SO_DEBUG: 1705 case SO_KEEPALIVE: 1706 case SO_DONTROUTE: 1707 case SO_USELOOPBACK: 1708 case SO_BROADCAST: 1709 case SO_REUSEADDR: 1710 case SO_REUSEPORT: 1711 case SO_OOBINLINE: 1712 case SO_TIMESTAMP: 1713 case SO_NOSIGPIPE: 1714 #ifdef SO_OTIMESTAMP 1715 case SO_OTIMESTAMP: 1716 #endif 1717 error = sockopt_getint(sopt, &optval); 1718 solock(so); 1719 if (error) 1720 break; 1721 if (optval) 1722 so->so_options |= opt; 1723 else 1724 so->so_options &= ~opt; 1725 break; 1726 1727 case SO_SNDBUF: 1728 case SO_RCVBUF: 1729 case SO_SNDLOWAT: 1730 case SO_RCVLOWAT: 1731 error = sockopt_getint(sopt, &optval); 1732 solock(so); 1733 if (error) 1734 break; 1735 1736 /* 1737 * Values < 1 make no sense for any of these 1738 * options, so disallow them. 1739 */ 1740 if (optval < 1) { 1741 error = EINVAL; 1742 break; 1743 } 1744 1745 switch (opt) { 1746 case SO_SNDBUF: 1747 if (sbreserve(&so->so_snd, (u_long)optval, so) == 0) { 1748 error = ENOBUFS; 1749 break; 1750 } 1751 so->so_snd.sb_flags &= ~SB_AUTOSIZE; 1752 break; 1753 1754 case SO_RCVBUF: 1755 if (sbreserve(&so->so_rcv, (u_long)optval, so) == 0) { 1756 error = ENOBUFS; 1757 break; 1758 } 1759 so->so_rcv.sb_flags &= ~SB_AUTOSIZE; 1760 break; 1761 1762 /* 1763 * Make sure the low-water is never greater than 1764 * the high-water. 1765 */ 1766 case SO_SNDLOWAT: 1767 if (optval > so->so_snd.sb_hiwat) 1768 optval = so->so_snd.sb_hiwat; 1769 1770 so->so_snd.sb_lowat = optval; 1771 break; 1772 1773 case SO_RCVLOWAT: 1774 if (optval > so->so_rcv.sb_hiwat) 1775 optval = so->so_rcv.sb_hiwat; 1776 1777 so->so_rcv.sb_lowat = optval; 1778 break; 1779 } 1780 break; 1781 1782 #ifdef COMPAT_50 1783 case SO_OSNDTIMEO: 1784 case SO_ORCVTIMEO: { 1785 struct timeval50 otv; 1786 error = sockopt_get(sopt, &otv, sizeof(otv)); 1787 if (error) { 1788 solock(so); 1789 break; 1790 } 1791 timeval50_to_timeval(&otv, &tv); 1792 opt = opt == SO_OSNDTIMEO ? SO_SNDTIMEO : SO_RCVTIMEO; 1793 error = 0; 1794 /*FALLTHROUGH*/ 1795 } 1796 #endif /* COMPAT_50 */ 1797 1798 case SO_SNDTIMEO: 1799 case SO_RCVTIMEO: 1800 if (error) 1801 error = sockopt_get(sopt, &tv, sizeof(tv)); 1802 solock(so); 1803 if (error) 1804 break; 1805 1806 if (tv.tv_sec > (INT_MAX - tv.tv_usec / tick) / hz) { 1807 error = EDOM; 1808 break; 1809 } 1810 1811 optval = tv.tv_sec * hz + tv.tv_usec / tick; 1812 if (optval == 0 && tv.tv_usec != 0) 1813 optval = 1; 1814 1815 switch (opt) { 1816 case SO_SNDTIMEO: 1817 so->so_snd.sb_timeo = optval; 1818 break; 1819 case SO_RCVTIMEO: 1820 so->so_rcv.sb_timeo = optval; 1821 break; 1822 } 1823 break; 1824 1825 default: 1826 solock(so); 1827 error = ENOPROTOOPT; 1828 break; 1829 } 1830 KASSERT(solocked(so)); 1831 return error; 1832 } 1833 1834 int 1835 sosetopt(struct socket *so, struct sockopt *sopt) 1836 { 1837 int error, prerr; 1838 1839 if (sopt->sopt_level == SOL_SOCKET) { 1840 error = sosetopt1(so, sopt); 1841 KASSERT(solocked(so)); 1842 } else { 1843 error = ENOPROTOOPT; 1844 solock(so); 1845 } 1846 1847 if ((error == 0 || error == ENOPROTOOPT) && 1848 so->so_proto != NULL && so->so_proto->pr_ctloutput != NULL) { 1849 /* give the protocol stack a shot */ 1850 prerr = (*so->so_proto->pr_ctloutput)(PRCO_SETOPT, so, sopt); 1851 if (prerr == 0) 1852 error = 0; 1853 else if (prerr != ENOPROTOOPT) 1854 error = prerr; 1855 } 1856 sounlock(so); 1857 return error; 1858 } 1859 1860 /* 1861 * so_setsockopt() is a wrapper providing a sockopt structure for sosetopt() 1862 */ 1863 int 1864 so_setsockopt(struct lwp *l, struct socket *so, int level, int name, 1865 const void *val, size_t valsize) 1866 { 1867 struct sockopt sopt; 1868 int error; 1869 1870 KASSERT(valsize == 0 || val != NULL); 1871 1872 sockopt_init(&sopt, level, name, valsize); 1873 sockopt_set(&sopt, val, valsize); 1874 1875 error = sosetopt(so, &sopt); 1876 1877 sockopt_destroy(&sopt); 1878 1879 return error; 1880 } 1881 1882 /* 1883 * internal get SOL_SOCKET options 1884 */ 1885 static int 1886 sogetopt1(struct socket *so, struct sockopt *sopt) 1887 { 1888 int error, optval, opt; 1889 struct linger l; 1890 struct timeval tv; 1891 1892 switch ((opt = sopt->sopt_name)) { 1893 1894 case SO_ACCEPTFILTER: 1895 error = accept_filt_getopt(so, sopt); 1896 break; 1897 1898 case SO_LINGER: 1899 l.l_onoff = (so->so_options & SO_LINGER) ? 1 : 0; 1900 l.l_linger = so->so_linger; 1901 1902 error = sockopt_set(sopt, &l, sizeof(l)); 1903 break; 1904 1905 case SO_USELOOPBACK: 1906 case SO_DONTROUTE: 1907 case SO_DEBUG: 1908 case SO_KEEPALIVE: 1909 case SO_REUSEADDR: 1910 case SO_REUSEPORT: 1911 case SO_BROADCAST: 1912 case SO_OOBINLINE: 1913 case SO_TIMESTAMP: 1914 case SO_NOSIGPIPE: 1915 #ifdef SO_OTIMESTAMP 1916 case SO_OTIMESTAMP: 1917 #endif 1918 case SO_ACCEPTCONN: 1919 error = sockopt_setint(sopt, (so->so_options & opt) ? 1 : 0); 1920 break; 1921 1922 case SO_TYPE: 1923 error = sockopt_setint(sopt, so->so_type); 1924 break; 1925 1926 case SO_ERROR: 1927 error = sockopt_setint(sopt, so->so_error); 1928 so->so_error = 0; 1929 break; 1930 1931 case SO_SNDBUF: 1932 error = sockopt_setint(sopt, so->so_snd.sb_hiwat); 1933 break; 1934 1935 case SO_RCVBUF: 1936 error = sockopt_setint(sopt, so->so_rcv.sb_hiwat); 1937 break; 1938 1939 case SO_SNDLOWAT: 1940 error = sockopt_setint(sopt, so->so_snd.sb_lowat); 1941 break; 1942 1943 case SO_RCVLOWAT: 1944 error = sockopt_setint(sopt, so->so_rcv.sb_lowat); 1945 break; 1946 1947 #ifdef COMPAT_50 1948 case SO_OSNDTIMEO: 1949 case SO_ORCVTIMEO: { 1950 struct timeval50 otv; 1951 1952 optval = (opt == SO_OSNDTIMEO ? 1953 so->so_snd.sb_timeo : so->so_rcv.sb_timeo); 1954 1955 otv.tv_sec = optval / hz; 1956 otv.tv_usec = (optval % hz) * tick; 1957 1958 error = sockopt_set(sopt, &otv, sizeof(otv)); 1959 break; 1960 } 1961 #endif /* COMPAT_50 */ 1962 1963 case SO_SNDTIMEO: 1964 case SO_RCVTIMEO: 1965 optval = (opt == SO_SNDTIMEO ? 1966 so->so_snd.sb_timeo : so->so_rcv.sb_timeo); 1967 1968 tv.tv_sec = optval / hz; 1969 tv.tv_usec = (optval % hz) * tick; 1970 1971 error = sockopt_set(sopt, &tv, sizeof(tv)); 1972 break; 1973 1974 case SO_OVERFLOWED: 1975 error = sockopt_setint(sopt, so->so_rcv.sb_overflowed); 1976 break; 1977 1978 default: 1979 error = ENOPROTOOPT; 1980 break; 1981 } 1982 1983 return (error); 1984 } 1985 1986 int 1987 sogetopt(struct socket *so, struct sockopt *sopt) 1988 { 1989 int error; 1990 1991 solock(so); 1992 if (sopt->sopt_level != SOL_SOCKET) { 1993 if (so->so_proto && so->so_proto->pr_ctloutput) { 1994 error = ((*so->so_proto->pr_ctloutput) 1995 (PRCO_GETOPT, so, sopt)); 1996 } else 1997 error = (ENOPROTOOPT); 1998 } else { 1999 error = sogetopt1(so, sopt); 2000 } 2001 sounlock(so); 2002 return (error); 2003 } 2004 2005 /* 2006 * alloc sockopt data buffer buffer 2007 * - will be released at destroy 2008 */ 2009 static int 2010 sockopt_alloc(struct sockopt *sopt, size_t len, km_flag_t kmflag) 2011 { 2012 2013 KASSERT(sopt->sopt_size == 0); 2014 2015 if (len > sizeof(sopt->sopt_buf)) { 2016 sopt->sopt_data = kmem_zalloc(len, kmflag); 2017 if (sopt->sopt_data == NULL) 2018 return ENOMEM; 2019 } else 2020 sopt->sopt_data = sopt->sopt_buf; 2021 2022 sopt->sopt_size = len; 2023 return 0; 2024 } 2025 2026 /* 2027 * initialise sockopt storage 2028 * - MAY sleep during allocation 2029 */ 2030 void 2031 sockopt_init(struct sockopt *sopt, int level, int name, size_t size) 2032 { 2033 2034 memset(sopt, 0, sizeof(*sopt)); 2035 2036 sopt->sopt_level = level; 2037 sopt->sopt_name = name; 2038 (void)sockopt_alloc(sopt, size, KM_SLEEP); 2039 } 2040 2041 /* 2042 * destroy sockopt storage 2043 * - will release any held memory references 2044 */ 2045 void 2046 sockopt_destroy(struct sockopt *sopt) 2047 { 2048 2049 if (sopt->sopt_data != sopt->sopt_buf) 2050 kmem_free(sopt->sopt_data, sopt->sopt_size); 2051 2052 memset(sopt, 0, sizeof(*sopt)); 2053 } 2054 2055 /* 2056 * set sockopt value 2057 * - value is copied into sockopt 2058 * - memory is allocated when necessary, will not sleep 2059 */ 2060 int 2061 sockopt_set(struct sockopt *sopt, const void *buf, size_t len) 2062 { 2063 int error; 2064 2065 if (sopt->sopt_size == 0) { 2066 error = sockopt_alloc(sopt, len, KM_NOSLEEP); 2067 if (error) 2068 return error; 2069 } 2070 2071 KASSERT(sopt->sopt_size == len); 2072 memcpy(sopt->sopt_data, buf, len); 2073 return 0; 2074 } 2075 2076 /* 2077 * common case of set sockopt integer value 2078 */ 2079 int 2080 sockopt_setint(struct sockopt *sopt, int val) 2081 { 2082 2083 return sockopt_set(sopt, &val, sizeof(int)); 2084 } 2085 2086 /* 2087 * get sockopt value 2088 * - correct size must be given 2089 */ 2090 int 2091 sockopt_get(const struct sockopt *sopt, void *buf, size_t len) 2092 { 2093 2094 if (sopt->sopt_size != len) 2095 return EINVAL; 2096 2097 memcpy(buf, sopt->sopt_data, len); 2098 return 0; 2099 } 2100 2101 /* 2102 * common case of get sockopt integer value 2103 */ 2104 int 2105 sockopt_getint(const struct sockopt *sopt, int *valp) 2106 { 2107 2108 return sockopt_get(sopt, valp, sizeof(int)); 2109 } 2110 2111 /* 2112 * set sockopt value from mbuf 2113 * - ONLY for legacy code 2114 * - mbuf is released by sockopt 2115 * - will not sleep 2116 */ 2117 int 2118 sockopt_setmbuf(struct sockopt *sopt, struct mbuf *m) 2119 { 2120 size_t len; 2121 int error; 2122 2123 len = m_length(m); 2124 2125 if (sopt->sopt_size == 0) { 2126 error = sockopt_alloc(sopt, len, KM_NOSLEEP); 2127 if (error) 2128 return error; 2129 } 2130 2131 KASSERT(sopt->sopt_size == len); 2132 m_copydata(m, 0, len, sopt->sopt_data); 2133 m_freem(m); 2134 2135 return 0; 2136 } 2137 2138 /* 2139 * get sockopt value into mbuf 2140 * - ONLY for legacy code 2141 * - mbuf to be released by the caller 2142 * - will not sleep 2143 */ 2144 struct mbuf * 2145 sockopt_getmbuf(const struct sockopt *sopt) 2146 { 2147 struct mbuf *m; 2148 2149 if (sopt->sopt_size > MCLBYTES) 2150 return NULL; 2151 2152 m = m_get(M_DONTWAIT, MT_SOOPTS); 2153 if (m == NULL) 2154 return NULL; 2155 2156 if (sopt->sopt_size > MLEN) { 2157 MCLGET(m, M_DONTWAIT); 2158 if ((m->m_flags & M_EXT) == 0) { 2159 m_free(m); 2160 return NULL; 2161 } 2162 } 2163 2164 memcpy(mtod(m, void *), sopt->sopt_data, sopt->sopt_size); 2165 m->m_len = sopt->sopt_size; 2166 2167 return m; 2168 } 2169 2170 void 2171 sohasoutofband(struct socket *so) 2172 { 2173 2174 fownsignal(so->so_pgid, SIGURG, POLL_PRI, POLLPRI|POLLRDBAND, so); 2175 selnotify(&so->so_rcv.sb_sel, POLLPRI | POLLRDBAND, NOTE_SUBMIT); 2176 } 2177 2178 static void 2179 filt_sordetach(struct knote *kn) 2180 { 2181 struct socket *so; 2182 2183 so = ((file_t *)kn->kn_obj)->f_data; 2184 solock(so); 2185 SLIST_REMOVE(&so->so_rcv.sb_sel.sel_klist, kn, knote, kn_selnext); 2186 if (SLIST_EMPTY(&so->so_rcv.sb_sel.sel_klist)) 2187 so->so_rcv.sb_flags &= ~SB_KNOTE; 2188 sounlock(so); 2189 } 2190 2191 /*ARGSUSED*/ 2192 static int 2193 filt_soread(struct knote *kn, long hint) 2194 { 2195 struct socket *so; 2196 int rv; 2197 2198 so = ((file_t *)kn->kn_obj)->f_data; 2199 if (hint != NOTE_SUBMIT) 2200 solock(so); 2201 kn->kn_data = so->so_rcv.sb_cc; 2202 if (so->so_state & SS_CANTRCVMORE) { 2203 kn->kn_flags |= EV_EOF; 2204 kn->kn_fflags = so->so_error; 2205 rv = 1; 2206 } else if (so->so_error) /* temporary udp error */ 2207 rv = 1; 2208 else if (kn->kn_sfflags & NOTE_LOWAT) 2209 rv = (kn->kn_data >= kn->kn_sdata); 2210 else 2211 rv = (kn->kn_data >= so->so_rcv.sb_lowat); 2212 if (hint != NOTE_SUBMIT) 2213 sounlock(so); 2214 return rv; 2215 } 2216 2217 static void 2218 filt_sowdetach(struct knote *kn) 2219 { 2220 struct socket *so; 2221 2222 so = ((file_t *)kn->kn_obj)->f_data; 2223 solock(so); 2224 SLIST_REMOVE(&so->so_snd.sb_sel.sel_klist, kn, knote, kn_selnext); 2225 if (SLIST_EMPTY(&so->so_snd.sb_sel.sel_klist)) 2226 so->so_snd.sb_flags &= ~SB_KNOTE; 2227 sounlock(so); 2228 } 2229 2230 /*ARGSUSED*/ 2231 static int 2232 filt_sowrite(struct knote *kn, long hint) 2233 { 2234 struct socket *so; 2235 int rv; 2236 2237 so = ((file_t *)kn->kn_obj)->f_data; 2238 if (hint != NOTE_SUBMIT) 2239 solock(so); 2240 kn->kn_data = sbspace(&so->so_snd); 2241 if (so->so_state & SS_CANTSENDMORE) { 2242 kn->kn_flags |= EV_EOF; 2243 kn->kn_fflags = so->so_error; 2244 rv = 1; 2245 } else if (so->so_error) /* temporary udp error */ 2246 rv = 1; 2247 else if (((so->so_state & SS_ISCONNECTED) == 0) && 2248 (so->so_proto->pr_flags & PR_CONNREQUIRED)) 2249 rv = 0; 2250 else if (kn->kn_sfflags & NOTE_LOWAT) 2251 rv = (kn->kn_data >= kn->kn_sdata); 2252 else 2253 rv = (kn->kn_data >= so->so_snd.sb_lowat); 2254 if (hint != NOTE_SUBMIT) 2255 sounlock(so); 2256 return rv; 2257 } 2258 2259 /*ARGSUSED*/ 2260 static int 2261 filt_solisten(struct knote *kn, long hint) 2262 { 2263 struct socket *so; 2264 int rv; 2265 2266 so = ((file_t *)kn->kn_obj)->f_data; 2267 2268 /* 2269 * Set kn_data to number of incoming connections, not 2270 * counting partial (incomplete) connections. 2271 */ 2272 if (hint != NOTE_SUBMIT) 2273 solock(so); 2274 kn->kn_data = so->so_qlen; 2275 rv = (kn->kn_data > 0); 2276 if (hint != NOTE_SUBMIT) 2277 sounlock(so); 2278 return rv; 2279 } 2280 2281 static const struct filterops solisten_filtops = 2282 { 1, NULL, filt_sordetach, filt_solisten }; 2283 static const struct filterops soread_filtops = 2284 { 1, NULL, filt_sordetach, filt_soread }; 2285 static const struct filterops sowrite_filtops = 2286 { 1, NULL, filt_sowdetach, filt_sowrite }; 2287 2288 int 2289 soo_kqfilter(struct file *fp, struct knote *kn) 2290 { 2291 struct socket *so; 2292 struct sockbuf *sb; 2293 2294 so = ((file_t *)kn->kn_obj)->f_data; 2295 solock(so); 2296 switch (kn->kn_filter) { 2297 case EVFILT_READ: 2298 if (so->so_options & SO_ACCEPTCONN) 2299 kn->kn_fop = &solisten_filtops; 2300 else 2301 kn->kn_fop = &soread_filtops; 2302 sb = &so->so_rcv; 2303 break; 2304 case EVFILT_WRITE: 2305 kn->kn_fop = &sowrite_filtops; 2306 sb = &so->so_snd; 2307 break; 2308 default: 2309 sounlock(so); 2310 return (EINVAL); 2311 } 2312 SLIST_INSERT_HEAD(&sb->sb_sel.sel_klist, kn, kn_selnext); 2313 sb->sb_flags |= SB_KNOTE; 2314 sounlock(so); 2315 return (0); 2316 } 2317 2318 static int 2319 sodopoll(struct socket *so, int events) 2320 { 2321 int revents; 2322 2323 revents = 0; 2324 2325 if (events & (POLLIN | POLLRDNORM)) 2326 if (soreadable(so)) 2327 revents |= events & (POLLIN | POLLRDNORM); 2328 2329 if (events & (POLLOUT | POLLWRNORM)) 2330 if (sowritable(so)) 2331 revents |= events & (POLLOUT | POLLWRNORM); 2332 2333 if (events & (POLLPRI | POLLRDBAND)) 2334 if (so->so_oobmark || (so->so_state & SS_RCVATMARK)) 2335 revents |= events & (POLLPRI | POLLRDBAND); 2336 2337 return revents; 2338 } 2339 2340 int 2341 sopoll(struct socket *so, int events) 2342 { 2343 int revents = 0; 2344 2345 #ifndef DIAGNOSTIC 2346 /* 2347 * Do a quick, unlocked check in expectation that the socket 2348 * will be ready for I/O. Don't do this check if DIAGNOSTIC, 2349 * as the solocked() assertions will fail. 2350 */ 2351 if ((revents = sodopoll(so, events)) != 0) 2352 return revents; 2353 #endif 2354 2355 solock(so); 2356 if ((revents = sodopoll(so, events)) == 0) { 2357 if (events & (POLLIN | POLLPRI | POLLRDNORM | POLLRDBAND)) { 2358 selrecord(curlwp, &so->so_rcv.sb_sel); 2359 so->so_rcv.sb_flags |= SB_NOTIFY; 2360 } 2361 2362 if (events & (POLLOUT | POLLWRNORM)) { 2363 selrecord(curlwp, &so->so_snd.sb_sel); 2364 so->so_snd.sb_flags |= SB_NOTIFY; 2365 } 2366 } 2367 sounlock(so); 2368 2369 return revents; 2370 } 2371 2372 2373 #include <sys/sysctl.h> 2374 2375 static int sysctl_kern_somaxkva(SYSCTLFN_PROTO); 2376 static int sysctl_kern_sbmax(SYSCTLFN_PROTO); 2377 2378 /* 2379 * sysctl helper routine for kern.somaxkva. ensures that the given 2380 * value is not too small. 2381 * (XXX should we maybe make sure it's not too large as well?) 2382 */ 2383 static int 2384 sysctl_kern_somaxkva(SYSCTLFN_ARGS) 2385 { 2386 int error, new_somaxkva; 2387 struct sysctlnode node; 2388 2389 new_somaxkva = somaxkva; 2390 node = *rnode; 2391 node.sysctl_data = &new_somaxkva; 2392 error = sysctl_lookup(SYSCTLFN_CALL(&node)); 2393 if (error || newp == NULL) 2394 return (error); 2395 2396 if (new_somaxkva < (16 * 1024 * 1024)) /* sanity */ 2397 return (EINVAL); 2398 2399 mutex_enter(&so_pendfree_lock); 2400 somaxkva = new_somaxkva; 2401 cv_broadcast(&socurkva_cv); 2402 mutex_exit(&so_pendfree_lock); 2403 2404 return (error); 2405 } 2406 2407 /* 2408 * sysctl helper routine for kern.sbmax. Basically just ensures that 2409 * any new value is not too small. 2410 */ 2411 static int 2412 sysctl_kern_sbmax(SYSCTLFN_ARGS) 2413 { 2414 int error, new_sbmax; 2415 struct sysctlnode node; 2416 2417 new_sbmax = sb_max; 2418 node = *rnode; 2419 node.sysctl_data = &new_sbmax; 2420 error = sysctl_lookup(SYSCTLFN_CALL(&node)); 2421 if (error || newp == NULL) 2422 return (error); 2423 2424 KERNEL_LOCK(1, NULL); 2425 error = sb_max_set(new_sbmax); 2426 KERNEL_UNLOCK_ONE(NULL); 2427 2428 return (error); 2429 } 2430 2431 static void 2432 sysctl_kern_socket_setup(void) 2433 { 2434 2435 KASSERT(socket_sysctllog == NULL); 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
Indexes created Sat Sep 20 22:09:52 GMT 2025