Annotation of src/sys/kern/uipc_socket.c, Revision 1.198
1.198 ! elad 1: /* $NetBSD: uipc_socket.c,v 1.197 2009/12/29 03:48:18 elad Exp $ */
1.64 thorpej 2:
3: /*-
1.188 ad 4: * Copyright (c) 2002, 2007, 2008, 2009 The NetBSD Foundation, Inc.
1.64 thorpej 5: * All rights reserved.
6: *
7: * This code is derived from software contributed to The NetBSD Foundation
1.188 ad 8: * by Jason R. Thorpe of Wasabi Systems, Inc, and by Andrew Doran.
1.64 thorpej 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: */
1.16 cgd 31:
1.1 cgd 32: /*
1.159 ad 33: * Copyright (c) 2004 The FreeBSD Foundation
34: * Copyright (c) 2004 Robert Watson
1.15 mycroft 35: * Copyright (c) 1982, 1986, 1988, 1990, 1993
36: * The Regents of the University of California. All rights reserved.
1.1 cgd 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.
1.85 agc 46: * 3. Neither the name of the University nor the names of its contributors
1.1 cgd 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: *
1.32 fvdl 62: * @(#)uipc_socket.c 8.6 (Berkeley) 5/2/95
1.1 cgd 63: */
1.59 lukem 64:
65: #include <sys/cdefs.h>
1.198 ! elad 66: __KERNEL_RCSID(0, "$NetBSD: uipc_socket.c,v 1.197 2009/12/29 03:48:18 elad Exp $");
1.64 thorpej 67:
1.179 christos 68: #include "opt_compat_netbsd.h"
1.64 thorpej 69: #include "opt_sock_counters.h"
70: #include "opt_sosend_loan.h"
1.81 martin 71: #include "opt_mbuftrace.h"
1.84 ragge 72: #include "opt_somaxkva.h"
1.167 ad 73: #include "opt_multiprocessor.h" /* XXX */
1.1 cgd 74:
1.9 mycroft 75: #include <sys/param.h>
76: #include <sys/systm.h>
77: #include <sys/proc.h>
78: #include <sys/file.h>
1.142 dyoung 79: #include <sys/filedesc.h>
1.173 plunky 80: #include <sys/kmem.h>
1.9 mycroft 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>
1.21 christos 87: #include <sys/signalvar.h>
1.9 mycroft 88: #include <sys/resourcevar.h>
1.174 pooka 89: #include <sys/uidinfo.h>
1.72 jdolecek 90: #include <sys/event.h>
1.89 christos 91: #include <sys/poll.h>
1.118 elad 92: #include <sys/kauth.h>
1.136 ad 93: #include <sys/mutex.h>
94: #include <sys/condvar.h>
1.37 thorpej 95:
1.179 christos 96: #ifdef COMPAT_50
97: #include <compat/sys/time.h>
1.184 christos 98: #include <compat/sys/socket.h>
1.179 christos 99: #endif
100:
1.64 thorpej 101: #include <uvm/uvm.h>
102:
1.77 thorpej 103: MALLOC_DEFINE(M_SOOPTS, "soopts", "socket options");
104: MALLOC_DEFINE(M_SONAME, "soname", "socket name");
1.37 thorpej 105:
1.142 dyoung 106: extern const struct fileops socketops;
107:
1.54 lukem 108: extern int somaxconn; /* patchable (XXX sysctl) */
109: int somaxconn = SOMAXCONN;
1.160 ad 110: kmutex_t *softnet_lock;
1.49 jonathan 111:
1.64 thorpej 112: #ifdef SOSEND_COUNTERS
113: #include <sys/device.h>
114:
1.113 thorpej 115: static struct evcnt sosend_loan_big = EVCNT_INITIALIZER(EVCNT_TYPE_MISC,
1.64 thorpej 116: NULL, "sosend", "loan big");
1.113 thorpej 117: static struct evcnt sosend_copy_big = EVCNT_INITIALIZER(EVCNT_TYPE_MISC,
1.64 thorpej 118: NULL, "sosend", "copy big");
1.113 thorpej 119: static struct evcnt sosend_copy_small = EVCNT_INITIALIZER(EVCNT_TYPE_MISC,
1.64 thorpej 120: NULL, "sosend", "copy small");
1.113 thorpej 121: static struct evcnt sosend_kvalimit = EVCNT_INITIALIZER(EVCNT_TYPE_MISC,
1.64 thorpej 122: NULL, "sosend", "kva limit");
123:
124: #define SOSEND_COUNTER_INCR(ev) (ev)->ev_count++
125:
1.101 matt 126: EVCNT_ATTACH_STATIC(sosend_loan_big);
127: EVCNT_ATTACH_STATIC(sosend_copy_big);
128: EVCNT_ATTACH_STATIC(sosend_copy_small);
129: EVCNT_ATTACH_STATIC(sosend_kvalimit);
1.64 thorpej 130: #else
131:
132: #define SOSEND_COUNTER_INCR(ev) /* nothing */
133:
134: #endif /* SOSEND_COUNTERS */
135:
1.119 yamt 136: static struct callback_entry sokva_reclaimerentry;
1.1 cgd 137:
1.167 ad 138: #if defined(SOSEND_NO_LOAN) || defined(MULTIPROCESSOR)
1.121 yamt 139: int sock_loan_thresh = -1;
1.71 thorpej 140: #else
1.121 yamt 141: int sock_loan_thresh = 4096;
1.65 thorpej 142: #endif
1.64 thorpej 143:
1.136 ad 144: static kmutex_t so_pendfree_lock;
1.113 thorpej 145: static struct mbuf *so_pendfree;
1.64 thorpej 146:
1.84 ragge 147: #ifndef SOMAXKVA
148: #define SOMAXKVA (16 * 1024 * 1024)
149: #endif
150: int somaxkva = SOMAXKVA;
1.113 thorpej 151: static int socurkva;
1.136 ad 152: static kcondvar_t socurkva_cv;
1.64 thorpej 153:
1.191 elad 154: static kauth_listener_t socket_listener;
155:
1.64 thorpej 156: #define SOCK_LOAN_CHUNK 65536
157:
1.117 yamt 158: static size_t sodopendfree(void);
159: static size_t sodopendfreel(void);
1.93 yamt 160:
1.178 pooka 161: static void sysctl_kern_somaxkva_setup(void);
162: static struct sysctllog *socket_sysctllog;
163:
1.113 thorpej 164: static vsize_t
1.129 yamt 165: sokvareserve(struct socket *so, vsize_t len)
1.80 yamt 166: {
1.98 christos 167: int error;
1.80 yamt 168:
1.136 ad 169: mutex_enter(&so_pendfree_lock);
1.80 yamt 170: while (socurkva + len > somaxkva) {
1.93 yamt 171: size_t freed;
172:
173: /*
174: * try to do pendfree.
175: */
176:
1.117 yamt 177: freed = sodopendfreel();
1.93 yamt 178:
179: /*
180: * if some kva was freed, try again.
181: */
182:
183: if (freed)
1.80 yamt 184: continue;
1.93 yamt 185:
1.80 yamt 186: SOSEND_COUNTER_INCR(&sosend_kvalimit);
1.136 ad 187: error = cv_wait_sig(&socurkva_cv, &so_pendfree_lock);
1.98 christos 188: if (error) {
189: len = 0;
190: break;
191: }
1.80 yamt 192: }
1.93 yamt 193: socurkva += len;
1.136 ad 194: mutex_exit(&so_pendfree_lock);
1.98 christos 195: return len;
1.95 yamt 196: }
197:
1.113 thorpej 198: static void
1.95 yamt 199: sokvaunreserve(vsize_t len)
200: {
201:
1.136 ad 202: mutex_enter(&so_pendfree_lock);
1.95 yamt 203: socurkva -= len;
1.136 ad 204: cv_broadcast(&socurkva_cv);
205: mutex_exit(&so_pendfree_lock);
1.95 yamt 206: }
207:
208: /*
209: * sokvaalloc: allocate kva for loan.
210: */
211:
212: vaddr_t
213: sokvaalloc(vsize_t len, struct socket *so)
214: {
215: vaddr_t lva;
216:
217: /*
218: * reserve kva.
219: */
220:
1.98 christos 221: if (sokvareserve(so, len) == 0)
222: return 0;
1.93 yamt 223:
224: /*
225: * allocate kva.
226: */
1.80 yamt 227:
1.109 yamt 228: lva = uvm_km_alloc(kernel_map, len, 0, UVM_KMF_VAONLY | UVM_KMF_WAITVA);
1.95 yamt 229: if (lva == 0) {
230: sokvaunreserve(len);
1.80 yamt 231: return (0);
1.95 yamt 232: }
1.80 yamt 233:
234: return lva;
235: }
236:
1.93 yamt 237: /*
238: * sokvafree: free kva for loan.
239: */
240:
1.80 yamt 241: void
242: sokvafree(vaddr_t sva, vsize_t len)
243: {
1.93 yamt 244:
245: /*
246: * free kva.
247: */
1.80 yamt 248:
1.109 yamt 249: uvm_km_free(kernel_map, sva, len, UVM_KMF_VAONLY);
1.93 yamt 250:
251: /*
252: * unreserve kva.
253: */
254:
1.95 yamt 255: sokvaunreserve(len);
1.80 yamt 256: }
257:
1.64 thorpej 258: static void
1.134 christos 259: sodoloanfree(struct vm_page **pgs, void *buf, size_t size)
1.64 thorpej 260: {
1.156 yamt 261: vaddr_t sva, eva;
1.64 thorpej 262: vsize_t len;
1.156 yamt 263: int npgs;
264:
265: KASSERT(pgs != NULL);
1.64 thorpej 266:
267: eva = round_page((vaddr_t) buf + size);
268: sva = trunc_page((vaddr_t) buf);
269: len = eva - sva;
270: npgs = len >> PAGE_SHIFT;
271:
272: pmap_kremove(sva, len);
273: pmap_update(pmap_kernel());
274: uvm_unloan(pgs, npgs, UVM_LOAN_TOPAGE);
1.80 yamt 275: sokvafree(sva, len);
1.64 thorpej 276: }
277:
278: static size_t
1.152 matt 279: sodopendfree(void)
1.64 thorpej 280: {
1.93 yamt 281: size_t rv;
1.64 thorpej 282:
1.160 ad 283: if (__predict_true(so_pendfree == NULL))
284: return 0;
285:
1.136 ad 286: mutex_enter(&so_pendfree_lock);
1.117 yamt 287: rv = sodopendfreel();
1.136 ad 288: mutex_exit(&so_pendfree_lock);
1.93 yamt 289:
290: return rv;
291: }
292:
293: /*
294: * sodopendfreel: free mbufs on "pendfree" list.
1.136 ad 295: * unlock and relock so_pendfree_lock when freeing mbufs.
1.93 yamt 296: *
1.136 ad 297: * => called with so_pendfree_lock held.
1.93 yamt 298: */
299:
300: static size_t
1.152 matt 301: sodopendfreel(void)
1.93 yamt 302: {
1.137 ad 303: struct mbuf *m, *next;
1.93 yamt 304: size_t rv = 0;
305:
1.136 ad 306: KASSERT(mutex_owned(&so_pendfree_lock));
1.64 thorpej 307:
1.137 ad 308: while (so_pendfree != NULL) {
1.64 thorpej 309: m = so_pendfree;
1.93 yamt 310: so_pendfree = NULL;
1.136 ad 311: mutex_exit(&so_pendfree_lock);
1.93 yamt 312:
313: for (; m != NULL; m = next) {
314: next = m->m_next;
1.156 yamt 315: KASSERT((~m->m_flags & (M_EXT|M_EXT_PAGES)) == 0);
316: KASSERT(m->m_ext.ext_refcnt == 0);
1.93 yamt 317:
318: rv += m->m_ext.ext_size;
1.156 yamt 319: sodoloanfree(m->m_ext.ext_pgs, m->m_ext.ext_buf,
1.93 yamt 320: m->m_ext.ext_size);
1.145 ad 321: pool_cache_put(mb_cache, m);
1.93 yamt 322: }
1.64 thorpej 323:
1.136 ad 324: mutex_enter(&so_pendfree_lock);
1.64 thorpej 325: }
326:
327: return (rv);
328: }
329:
1.80 yamt 330: void
1.134 christos 331: soloanfree(struct mbuf *m, void *buf, size_t size, void *arg)
1.64 thorpej 332: {
333:
1.156 yamt 334: KASSERT(m != NULL);
1.64 thorpej 335:
1.93 yamt 336: /*
337: * postpone freeing mbuf.
338: *
339: * we can't do it in interrupt context
340: * because we need to put kva back to kernel_map.
341: */
342:
1.136 ad 343: mutex_enter(&so_pendfree_lock);
1.92 yamt 344: m->m_next = so_pendfree;
345: so_pendfree = m;
1.136 ad 346: cv_broadcast(&socurkva_cv);
347: mutex_exit(&so_pendfree_lock);
1.64 thorpej 348: }
349:
350: static long
351: sosend_loan(struct socket *so, struct uio *uio, struct mbuf *m, long space)
352: {
353: struct iovec *iov = uio->uio_iov;
354: vaddr_t sva, eva;
355: vsize_t len;
1.156 yamt 356: vaddr_t lva;
357: int npgs, error;
358: vaddr_t va;
359: int i;
1.64 thorpej 360:
1.116 yamt 361: if (VMSPACE_IS_KERNEL_P(uio->uio_vmspace))
1.64 thorpej 362: return (0);
363:
364: if (iov->iov_len < (size_t) space)
365: space = iov->iov_len;
366: if (space > SOCK_LOAN_CHUNK)
367: space = SOCK_LOAN_CHUNK;
368:
369: eva = round_page((vaddr_t) iov->iov_base + space);
370: sva = trunc_page((vaddr_t) iov->iov_base);
371: len = eva - sva;
372: npgs = len >> PAGE_SHIFT;
373:
1.79 thorpej 374: KASSERT(npgs <= M_EXT_MAXPAGES);
375:
1.80 yamt 376: lva = sokvaalloc(len, so);
1.64 thorpej 377: if (lva == 0)
1.80 yamt 378: return 0;
1.64 thorpej 379:
1.116 yamt 380: error = uvm_loan(&uio->uio_vmspace->vm_map, sva, len,
1.79 thorpej 381: m->m_ext.ext_pgs, UVM_LOAN_TOPAGE);
1.64 thorpej 382: if (error) {
1.80 yamt 383: sokvafree(lva, len);
1.64 thorpej 384: return (0);
385: }
386:
387: for (i = 0, va = lva; i < npgs; i++, va += PAGE_SIZE)
1.79 thorpej 388: pmap_kenter_pa(va, VM_PAGE_TO_PHYS(m->m_ext.ext_pgs[i]),
1.194 cegger 389: VM_PROT_READ, 0);
1.64 thorpej 390: pmap_update(pmap_kernel());
391:
392: lva += (vaddr_t) iov->iov_base & PAGE_MASK;
393:
1.134 christos 394: MEXTADD(m, (void *) lva, space, M_MBUF, soloanfree, so);
1.79 thorpej 395: m->m_flags |= M_EXT_PAGES | M_EXT_ROMAP;
1.64 thorpej 396:
397: uio->uio_resid -= space;
398: /* uio_offset not updated, not set/used for write(2) */
1.134 christos 399: uio->uio_iov->iov_base = (char *)uio->uio_iov->iov_base + space;
1.64 thorpej 400: uio->uio_iov->iov_len -= space;
401: if (uio->uio_iov->iov_len == 0) {
402: uio->uio_iov++;
403: uio->uio_iovcnt--;
404: }
405:
406: return (space);
407: }
408:
1.119 yamt 409: static int
1.129 yamt 410: sokva_reclaim_callback(struct callback_entry *ce, void *obj, void *arg)
1.119 yamt 411: {
412:
413: KASSERT(ce == &sokva_reclaimerentry);
414: KASSERT(obj == NULL);
415:
416: sodopendfree();
417: if (!vm_map_starved_p(kernel_map)) {
418: return CALLBACK_CHAIN_ABORT;
419: }
420: return CALLBACK_CHAIN_CONTINUE;
421: }
422:
1.142 dyoung 423: struct mbuf *
1.147 dyoung 424: getsombuf(struct socket *so, int type)
1.142 dyoung 425: {
426: struct mbuf *m;
427:
1.147 dyoung 428: m = m_get(M_WAIT, type);
1.142 dyoung 429: MCLAIM(m, so->so_mowner);
430: return m;
431: }
432:
1.191 elad 433: static int
434: socket_listener_cb(kauth_cred_t cred, kauth_action_t action, void *cookie,
435: void *arg0, void *arg1, void *arg2, void *arg3)
436: {
437: int result;
438: enum kauth_network_req req;
439:
440: result = KAUTH_RESULT_DEFER;
441: req = (enum kauth_network_req)arg0;
442:
1.193 elad 443: if ((action != KAUTH_NETWORK_SOCKET) &&
444: (action != KAUTH_NETWORK_BIND))
1.191 elad 445: return result;
446:
447: switch (req) {
1.193 elad 448: case KAUTH_REQ_NETWORK_BIND_PORT:
449: result = KAUTH_RESULT_ALLOW;
450: break;
451:
1.191 elad 452: case KAUTH_REQ_NETWORK_SOCKET_DROP: {
453: /* Normal users can only drop their own connections. */
454: struct socket *so = (struct socket *)arg1;
455: uid_t sockuid = so->so_uidinfo->ui_uid;
456:
457: if (sockuid == kauth_cred_getuid(cred) ||
458: sockuid == kauth_cred_geteuid(cred))
459: result = KAUTH_RESULT_ALLOW;
460:
461: break;
462: }
463:
464: case KAUTH_REQ_NETWORK_SOCKET_OPEN:
465: /* We allow "raw" routing/bluetooth sockets to anyone. */
466: if ((u_long)arg1 == PF_ROUTE || (u_long)arg1 == PF_BLUETOOTH)
467: result = KAUTH_RESULT_ALLOW;
468: else {
469: /* Privileged, let secmodel handle this. */
470: if ((u_long)arg2 == SOCK_RAW)
471: break;
472: }
473:
474: result = KAUTH_RESULT_ALLOW;
475:
476: break;
477:
1.192 elad 478: case KAUTH_REQ_NETWORK_SOCKET_CANSEE:
479: result = KAUTH_RESULT_ALLOW;
480:
481: break;
482:
1.191 elad 483: default:
484: break;
485: }
486:
487: return result;
488: }
489:
1.119 yamt 490: void
491: soinit(void)
492: {
493:
1.178 pooka 494: sysctl_kern_somaxkva_setup();
495:
1.148 ad 496: mutex_init(&so_pendfree_lock, MUTEX_DEFAULT, IPL_VM);
1.160 ad 497: softnet_lock = mutex_obj_alloc(MUTEX_DEFAULT, IPL_NONE);
1.136 ad 498: cv_init(&socurkva_cv, "sokva");
1.166 ad 499: soinit2();
1.136 ad 500:
1.119 yamt 501: /* Set the initial adjusted socket buffer size. */
502: if (sb_max_set(sb_max))
503: panic("bad initial sb_max value: %lu", sb_max);
504:
505: callback_register(&vm_map_to_kernel(kernel_map)->vmk_reclaim_callback,
506: &sokva_reclaimerentry, NULL, sokva_reclaim_callback);
1.191 elad 507:
508: socket_listener = kauth_listen_scope(KAUTH_SCOPE_NETWORK,
509: socket_listener_cb, NULL);
1.119 yamt 510: }
511:
1.1 cgd 512: /*
513: * Socket operation routines.
514: * These routines are called by the routines in
515: * sys_socket.c or from a system process, and
516: * implement the semantics of socket operations by
517: * switching out to the protocol specific routines.
518: */
519: /*ARGSUSED*/
1.3 andrew 520: int
1.160 ad 521: socreate(int dom, struct socket **aso, int type, int proto, struct lwp *l,
522: struct socket *lockso)
1.1 cgd 523: {
1.99 matt 524: const struct protosw *prp;
1.54 lukem 525: struct socket *so;
1.115 yamt 526: uid_t uid;
1.160 ad 527: int error;
528: kmutex_t *lock;
1.1 cgd 529:
1.132 elad 530: error = kauth_authorize_network(l->l_cred, KAUTH_NETWORK_SOCKET,
531: KAUTH_REQ_NETWORK_SOCKET_OPEN, KAUTH_ARG(dom), KAUTH_ARG(type),
532: KAUTH_ARG(proto));
1.140 dyoung 533: if (error != 0)
534: return error;
1.127 elad 535:
1.1 cgd 536: if (proto)
537: prp = pffindproto(dom, proto, type);
538: else
539: prp = pffindtype(dom, type);
1.140 dyoung 540: if (prp == NULL) {
1.120 ginsbach 541: /* no support for domain */
542: if (pffinddomain(dom) == 0)
1.140 dyoung 543: return EAFNOSUPPORT;
1.120 ginsbach 544: /* no support for socket type */
545: if (proto == 0 && type != 0)
1.140 dyoung 546: return EPROTOTYPE;
547: return EPROTONOSUPPORT;
1.120 ginsbach 548: }
1.140 dyoung 549: if (prp->pr_usrreq == NULL)
550: return EPROTONOSUPPORT;
1.1 cgd 551: if (prp->pr_type != type)
1.140 dyoung 552: return EPROTOTYPE;
1.160 ad 553:
554: so = soget(true);
1.1 cgd 555: so->so_type = type;
556: so->so_proto = prp;
1.33 matt 557: so->so_send = sosend;
558: so->so_receive = soreceive;
1.78 matt 559: #ifdef MBUFTRACE
560: so->so_rcv.sb_mowner = &prp->pr_domain->dom_mowner;
561: so->so_snd.sb_mowner = &prp->pr_domain->dom_mowner;
562: so->so_mowner = &prp->pr_domain->dom_mowner;
563: #endif
1.138 rmind 564: uid = kauth_cred_geteuid(l->l_cred);
1.115 yamt 565: so->so_uidinfo = uid_find(uid);
1.168 yamt 566: so->so_egid = kauth_cred_getegid(l->l_cred);
567: so->so_cpid = l->l_proc->p_pid;
1.160 ad 568: if (lockso != NULL) {
569: /* Caller wants us to share a lock. */
570: lock = lockso->so_lock;
571: so->so_lock = lock;
572: mutex_obj_hold(lock);
573: mutex_enter(lock);
574: } else {
575: /* Lock assigned and taken during PRU_ATTACH. */
576: }
1.140 dyoung 577: error = (*prp->pr_usrreq)(so, PRU_ATTACH, NULL,
578: (struct mbuf *)(long)proto, NULL, l);
1.160 ad 579: KASSERT(solocked(so));
1.140 dyoung 580: if (error != 0) {
1.1 cgd 581: so->so_state |= SS_NOFDREF;
582: sofree(so);
1.140 dyoung 583: return error;
1.1 cgd 584: }
1.198 ! elad 585: so->so_cred = kauth_cred_dup(l->l_cred);
1.160 ad 586: sounlock(so);
1.1 cgd 587: *aso = so;
1.140 dyoung 588: return 0;
1.1 cgd 589: }
590:
1.142 dyoung 591: /* On success, write file descriptor to fdout and return zero. On
592: * failure, return non-zero; *fdout will be undefined.
593: */
594: int
595: fsocreate(int domain, struct socket **sop, int type, int protocol,
596: struct lwp *l, int *fdout)
597: {
598: struct socket *so;
599: struct file *fp;
600: int fd, error;
601:
1.155 ad 602: if ((error = fd_allocfile(&fp, &fd)) != 0)
1.142 dyoung 603: return (error);
604: fp->f_flag = FREAD|FWRITE;
605: fp->f_type = DTYPE_SOCKET;
606: fp->f_ops = &socketops;
1.160 ad 607: error = socreate(domain, &so, type, protocol, l, NULL);
1.142 dyoung 608: if (error != 0) {
1.155 ad 609: fd_abort(curproc, fp, fd);
1.142 dyoung 610: } else {
611: if (sop != NULL)
612: *sop = so;
613: fp->f_data = so;
1.155 ad 614: fd_affix(curproc, fp, fd);
1.142 dyoung 615: *fdout = fd;
616: }
617: return error;
618: }
619:
1.3 andrew 620: int
1.190 dyoung 621: sofamily(const struct socket *so)
622: {
623: const struct protosw *pr;
624: const struct domain *dom;
625:
626: if ((pr = so->so_proto) == NULL)
627: return AF_UNSPEC;
628: if ((dom = pr->pr_domain) == NULL)
629: return AF_UNSPEC;
630: return dom->dom_family;
631: }
632:
633: int
1.114 christos 634: sobind(struct socket *so, struct mbuf *nam, struct lwp *l)
1.1 cgd 635: {
1.160 ad 636: int error;
1.1 cgd 637:
1.160 ad 638: solock(so);
1.140 dyoung 639: error = (*so->so_proto->pr_usrreq)(so, PRU_BIND, NULL, nam, NULL, l);
1.160 ad 640: sounlock(so);
1.140 dyoung 641: return error;
1.1 cgd 642: }
643:
1.3 andrew 644: int
1.150 elad 645: solisten(struct socket *so, int backlog, struct lwp *l)
1.1 cgd 646: {
1.160 ad 647: int error;
1.1 cgd 648:
1.160 ad 649: solock(so);
1.158 ad 650: if ((so->so_state & (SS_ISCONNECTED | SS_ISCONNECTING |
1.163 ad 651: SS_ISDISCONNECTING)) != 0) {
652: sounlock(so);
1.158 ad 653: return (EOPNOTSUPP);
1.163 ad 654: }
1.140 dyoung 655: error = (*so->so_proto->pr_usrreq)(so, PRU_LISTEN, NULL,
1.150 elad 656: NULL, NULL, l);
1.140 dyoung 657: if (error != 0) {
1.160 ad 658: sounlock(so);
1.140 dyoung 659: return error;
1.1 cgd 660: }
1.63 matt 661: if (TAILQ_EMPTY(&so->so_q))
1.1 cgd 662: so->so_options |= SO_ACCEPTCONN;
663: if (backlog < 0)
664: backlog = 0;
1.49 jonathan 665: so->so_qlimit = min(backlog, somaxconn);
1.160 ad 666: sounlock(so);
1.140 dyoung 667: return 0;
1.1 cgd 668: }
669:
1.21 christos 670: void
1.54 lukem 671: sofree(struct socket *so)
1.1 cgd 672: {
1.161 ad 673: u_int refs;
1.1 cgd 674:
1.160 ad 675: KASSERT(solocked(so));
676:
677: if (so->so_pcb || (so->so_state & SS_NOFDREF) == 0) {
678: sounlock(so);
1.1 cgd 679: return;
1.160 ad 680: }
1.43 mycroft 681: if (so->so_head) {
682: /*
683: * We must not decommission a socket that's on the accept(2)
684: * queue. If we do, then accept(2) may hang after select(2)
685: * indicated that the listening socket was ready.
686: */
1.160 ad 687: if (!soqremque(so, 0)) {
688: sounlock(so);
1.43 mycroft 689: return;
1.160 ad 690: }
1.43 mycroft 691: }
1.98 christos 692: if (so->so_rcv.sb_hiwat)
1.110 christos 693: (void)chgsbsize(so->so_uidinfo, &so->so_rcv.sb_hiwat, 0,
1.98 christos 694: RLIM_INFINITY);
695: if (so->so_snd.sb_hiwat)
1.110 christos 696: (void)chgsbsize(so->so_uidinfo, &so->so_snd.sb_hiwat, 0,
1.98 christos 697: RLIM_INFINITY);
698: sbrelease(&so->so_snd, so);
1.160 ad 699: KASSERT(!cv_has_waiters(&so->so_cv));
700: KASSERT(!cv_has_waiters(&so->so_rcv.sb_cv));
701: KASSERT(!cv_has_waiters(&so->so_snd.sb_cv));
1.1 cgd 702: sorflush(so);
1.161 ad 703: refs = so->so_aborting; /* XXX */
1.177 ad 704: /* Remove acccept filter if one is present. */
1.170 tls 705: if (so->so_accf != NULL)
1.177 ad 706: (void)accept_filt_clear(so);
1.160 ad 707: sounlock(so);
1.161 ad 708: if (refs == 0) /* XXX */
709: soput(so);
1.1 cgd 710: }
711:
712: /*
713: * Close a socket on last file table reference removal.
714: * Initiate disconnect if connected.
715: * Free socket when disconnect complete.
716: */
1.3 andrew 717: int
1.54 lukem 718: soclose(struct socket *so)
1.1 cgd 719: {
1.54 lukem 720: struct socket *so2;
1.160 ad 721: int error;
722: int error2;
1.1 cgd 723:
1.54 lukem 724: error = 0;
1.160 ad 725: solock(so);
1.1 cgd 726: if (so->so_options & SO_ACCEPTCONN) {
1.172 ad 727: for (;;) {
728: if ((so2 = TAILQ_FIRST(&so->so_q0)) != 0) {
1.160 ad 729: KASSERT(solocked2(so, so2));
730: (void) soqremque(so2, 0);
731: /* soabort drops the lock. */
732: (void) soabort(so2);
733: solock(so);
1.172 ad 734: continue;
1.160 ad 735: }
1.172 ad 736: if ((so2 = TAILQ_FIRST(&so->so_q)) != 0) {
1.160 ad 737: KASSERT(solocked2(so, so2));
738: (void) soqremque(so2, 1);
739: /* soabort drops the lock. */
740: (void) soabort(so2);
741: solock(so);
1.172 ad 742: continue;
1.160 ad 743: }
1.172 ad 744: break;
745: }
1.1 cgd 746: }
747: if (so->so_pcb == 0)
748: goto discard;
749: if (so->so_state & SS_ISCONNECTED) {
750: if ((so->so_state & SS_ISDISCONNECTING) == 0) {
751: error = sodisconnect(so);
752: if (error)
753: goto drop;
754: }
755: if (so->so_options & SO_LINGER) {
1.151 ad 756: if ((so->so_state & SS_ISDISCONNECTING) && so->so_nbio)
1.1 cgd 757: goto drop;
1.21 christos 758: while (so->so_state & SS_ISCONNECTED) {
1.185 yamt 759: error = sowait(so, true, so->so_linger * hz);
1.21 christos 760: if (error)
1.1 cgd 761: break;
1.21 christos 762: }
1.1 cgd 763: }
764: }
1.54 lukem 765: drop:
1.1 cgd 766: if (so->so_pcb) {
1.160 ad 767: error2 = (*so->so_proto->pr_usrreq)(so, PRU_DETACH,
1.140 dyoung 768: NULL, NULL, NULL, NULL);
1.1 cgd 769: if (error == 0)
770: error = error2;
771: }
1.54 lukem 772: discard:
1.1 cgd 773: if (so->so_state & SS_NOFDREF)
774: panic("soclose: NOFDREF");
1.198 ! elad 775: kauth_cred_free(so->so_cred);
1.1 cgd 776: so->so_state |= SS_NOFDREF;
777: sofree(so);
778: return (error);
779: }
780:
781: /*
1.160 ad 782: * Must be called with the socket locked.. Will return with it unlocked.
1.1 cgd 783: */
1.3 andrew 784: int
1.54 lukem 785: soabort(struct socket *so)
1.1 cgd 786: {
1.161 ad 787: u_int refs;
1.139 yamt 788: int error;
1.160 ad 789:
790: KASSERT(solocked(so));
791: KASSERT(so->so_head == NULL);
1.1 cgd 792:
1.161 ad 793: so->so_aborting++; /* XXX */
1.140 dyoung 794: error = (*so->so_proto->pr_usrreq)(so, PRU_ABORT, NULL,
795: NULL, NULL, NULL);
1.161 ad 796: refs = --so->so_aborting; /* XXX */
1.164 drochner 797: if (error || (refs == 0)) {
1.139 yamt 798: sofree(so);
1.160 ad 799: } else {
800: sounlock(so);
1.139 yamt 801: }
802: return error;
1.1 cgd 803: }
804:
1.3 andrew 805: int
1.54 lukem 806: soaccept(struct socket *so, struct mbuf *nam)
1.1 cgd 807: {
1.160 ad 808: int error;
809:
810: KASSERT(solocked(so));
1.1 cgd 811:
1.54 lukem 812: error = 0;
1.1 cgd 813: if ((so->so_state & SS_NOFDREF) == 0)
814: panic("soaccept: !NOFDREF");
815: so->so_state &= ~SS_NOFDREF;
1.55 thorpej 816: if ((so->so_state & SS_ISDISCONNECTED) == 0 ||
817: (so->so_proto->pr_flags & PR_ABRTACPTDIS) == 0)
1.41 mycroft 818: error = (*so->so_proto->pr_usrreq)(so, PRU_ACCEPT,
1.140 dyoung 819: NULL, nam, NULL, NULL);
1.41 mycroft 820: else
1.53 itojun 821: error = ECONNABORTED;
1.52 itojun 822:
1.1 cgd 823: return (error);
824: }
825:
1.3 andrew 826: int
1.114 christos 827: soconnect(struct socket *so, struct mbuf *nam, struct lwp *l)
1.1 cgd 828: {
1.160 ad 829: int error;
830:
831: KASSERT(solocked(so));
1.1 cgd 832:
833: if (so->so_options & SO_ACCEPTCONN)
834: return (EOPNOTSUPP);
835: /*
836: * If protocol is connection-based, can only connect once.
837: * Otherwise, if connected, try to disconnect first.
838: * This allows user to disconnect by connecting to, e.g.,
839: * a null address.
840: */
841: if (so->so_state & (SS_ISCONNECTED|SS_ISCONNECTING) &&
842: ((so->so_proto->pr_flags & PR_CONNREQUIRED) ||
843: (error = sodisconnect(so))))
844: error = EISCONN;
845: else
846: error = (*so->so_proto->pr_usrreq)(so, PRU_CONNECT,
1.140 dyoung 847: NULL, nam, NULL, l);
1.1 cgd 848: return (error);
849: }
850:
1.3 andrew 851: int
1.54 lukem 852: soconnect2(struct socket *so1, struct socket *so2)
1.1 cgd 853: {
1.160 ad 854: int error;
855:
856: KASSERT(solocked2(so1, so2));
1.1 cgd 857:
1.22 mycroft 858: error = (*so1->so_proto->pr_usrreq)(so1, PRU_CONNECT2,
1.140 dyoung 859: NULL, (struct mbuf *)so2, NULL, NULL);
1.1 cgd 860: return (error);
861: }
862:
1.3 andrew 863: int
1.54 lukem 864: sodisconnect(struct socket *so)
1.1 cgd 865: {
1.160 ad 866: int error;
867:
868: KASSERT(solocked(so));
1.1 cgd 869:
870: if ((so->so_state & SS_ISCONNECTED) == 0) {
871: error = ENOTCONN;
1.160 ad 872: } else if (so->so_state & SS_ISDISCONNECTING) {
1.1 cgd 873: error = EALREADY;
1.160 ad 874: } else {
875: error = (*so->so_proto->pr_usrreq)(so, PRU_DISCONNECT,
876: NULL, NULL, NULL, NULL);
1.1 cgd 877: }
1.117 yamt 878: sodopendfree();
1.1 cgd 879: return (error);
880: }
881:
1.15 mycroft 882: #define SBLOCKWAIT(f) (((f) & MSG_DONTWAIT) ? M_NOWAIT : M_WAITOK)
1.1 cgd 883: /*
884: * Send on a socket.
885: * If send must go all at once and message is larger than
886: * send buffering, then hard error.
887: * Lock against other senders.
888: * If must go all at once and not enough room now, then
889: * inform user that this would block and do nothing.
890: * Otherwise, if nonblocking, send as much as possible.
891: * The data to be sent is described by "uio" if nonzero,
892: * otherwise by the mbuf chain "top" (which must be null
893: * if uio is not). Data provided in mbuf chain must be small
894: * enough to send all at once.
895: *
896: * Returns nonzero on error, timeout or signal; callers
897: * must check for short counts if EINTR/ERESTART are returned.
898: * Data and control buffers are freed on return.
899: */
1.3 andrew 900: int
1.54 lukem 901: sosend(struct socket *so, struct mbuf *addr, struct uio *uio, struct mbuf *top,
1.114 christos 902: struct mbuf *control, int flags, struct lwp *l)
1.1 cgd 903: {
1.54 lukem 904: struct mbuf **mp, *m;
1.114 christos 905: struct proc *p;
1.58 jdolecek 906: long space, len, resid, clen, mlen;
907: int error, s, dontroute, atomic;
1.196 dsl 908: short wakeup_state = 0;
1.54 lukem 909:
1.114 christos 910: p = l->l_proc;
1.117 yamt 911: sodopendfree();
1.160 ad 912: clen = 0;
1.64 thorpej 913:
1.160 ad 914: /*
915: * solock() provides atomicity of access. splsoftnet() prevents
916: * protocol processing soft interrupts from interrupting us and
917: * blocking (expensive).
918: */
919: s = splsoftnet();
920: solock(so);
1.54 lukem 921: atomic = sosendallatonce(so) || top;
1.1 cgd 922: if (uio)
923: resid = uio->uio_resid;
924: else
925: resid = top->m_pkthdr.len;
1.7 cgd 926: /*
927: * In theory resid should be unsigned.
928: * However, space must be signed, as it might be less than 0
929: * if we over-committed, and we must use a signed comparison
930: * of space and resid. On the other hand, a negative resid
931: * causes us to loop sending 0-length segments to the protocol.
932: */
1.29 mycroft 933: if (resid < 0) {
934: error = EINVAL;
935: goto out;
936: }
1.1 cgd 937: dontroute =
938: (flags & MSG_DONTROUTE) && (so->so_options & SO_DONTROUTE) == 0 &&
939: (so->so_proto->pr_flags & PR_ATOMIC);
1.165 christos 940: l->l_ru.ru_msgsnd++;
1.1 cgd 941: if (control)
942: clen = control->m_len;
1.54 lukem 943: restart:
1.21 christos 944: if ((error = sblock(&so->so_snd, SBLOCKWAIT(flags))) != 0)
1.1 cgd 945: goto out;
946: do {
1.160 ad 947: if (so->so_state & SS_CANTSENDMORE) {
948: error = EPIPE;
949: goto release;
950: }
1.48 thorpej 951: if (so->so_error) {
952: error = so->so_error;
953: so->so_error = 0;
954: goto release;
955: }
1.1 cgd 956: if ((so->so_state & SS_ISCONNECTED) == 0) {
957: if (so->so_proto->pr_flags & PR_CONNREQUIRED) {
958: if ((so->so_state & SS_ISCONFIRMING) == 0 &&
1.160 ad 959: !(resid == 0 && clen != 0)) {
960: error = ENOTCONN;
961: goto release;
962: }
963: } else if (addr == 0) {
964: error = EDESTADDRREQ;
965: goto release;
966: }
1.1 cgd 967: }
968: space = sbspace(&so->so_snd);
969: if (flags & MSG_OOB)
970: space += 1024;
1.21 christos 971: if ((atomic && resid > so->so_snd.sb_hiwat) ||
1.160 ad 972: clen > so->so_snd.sb_hiwat) {
973: error = EMSGSIZE;
974: goto release;
975: }
1.96 mycroft 976: if (space < resid + clen &&
1.1 cgd 977: (atomic || space < so->so_snd.sb_lowat || space < clen)) {
1.160 ad 978: if (so->so_nbio) {
979: error = EWOULDBLOCK;
980: goto release;
981: }
1.1 cgd 982: sbunlock(&so->so_snd);
1.196 dsl 983: if (wakeup_state & SS_RESTARTSYS) {
984: error = ERESTART;
985: goto out;
986: }
1.1 cgd 987: error = sbwait(&so->so_snd);
988: if (error)
989: goto out;
1.196 dsl 990: wakeup_state = so->so_state;
1.1 cgd 991: goto restart;
992: }
1.196 dsl 993: wakeup_state = 0;
1.1 cgd 994: mp = ⊤
995: space -= clen;
996: do {
1.45 tv 997: if (uio == NULL) {
998: /*
999: * Data is prepackaged in "top".
1000: */
1001: resid = 0;
1002: if (flags & MSG_EOR)
1003: top->m_flags |= M_EOR;
1004: } else do {
1.160 ad 1005: sounlock(so);
1006: splx(s);
1.144 dyoung 1007: if (top == NULL) {
1.78 matt 1008: m = m_gethdr(M_WAIT, MT_DATA);
1.45 tv 1009: mlen = MHLEN;
1010: m->m_pkthdr.len = 0;
1.140 dyoung 1011: m->m_pkthdr.rcvif = NULL;
1.45 tv 1012: } else {
1.78 matt 1013: m = m_get(M_WAIT, MT_DATA);
1.45 tv 1014: mlen = MLEN;
1015: }
1.78 matt 1016: MCLAIM(m, so->so_snd.sb_mowner);
1.121 yamt 1017: if (sock_loan_thresh >= 0 &&
1018: uio->uio_iov->iov_len >= sock_loan_thresh &&
1019: space >= sock_loan_thresh &&
1.64 thorpej 1020: (len = sosend_loan(so, uio, m,
1021: space)) != 0) {
1022: SOSEND_COUNTER_INCR(&sosend_loan_big);
1023: space -= len;
1024: goto have_data;
1025: }
1.45 tv 1026: if (resid >= MINCLSIZE && space >= MCLBYTES) {
1.64 thorpej 1027: SOSEND_COUNTER_INCR(&sosend_copy_big);
1.78 matt 1028: m_clget(m, M_WAIT);
1.45 tv 1029: if ((m->m_flags & M_EXT) == 0)
1030: goto nopages;
1031: mlen = MCLBYTES;
1032: if (atomic && top == 0) {
1.58 jdolecek 1033: len = lmin(MCLBYTES - max_hdr,
1.54 lukem 1034: resid);
1.45 tv 1035: m->m_data += max_hdr;
1036: } else
1.58 jdolecek 1037: len = lmin(MCLBYTES, resid);
1.45 tv 1038: space -= len;
1039: } else {
1.64 thorpej 1040: nopages:
1041: SOSEND_COUNTER_INCR(&sosend_copy_small);
1.58 jdolecek 1042: len = lmin(lmin(mlen, resid), space);
1.45 tv 1043: space -= len;
1044: /*
1045: * For datagram protocols, leave room
1046: * for protocol headers in first mbuf.
1047: */
1048: if (atomic && top == 0 && len < mlen)
1049: MH_ALIGN(m, len);
1050: }
1.144 dyoung 1051: error = uiomove(mtod(m, void *), (int)len, uio);
1.64 thorpej 1052: have_data:
1.45 tv 1053: resid = uio->uio_resid;
1054: m->m_len = len;
1055: *mp = m;
1056: top->m_pkthdr.len += len;
1.160 ad 1057: s = splsoftnet();
1058: solock(so);
1.144 dyoung 1059: if (error != 0)
1.45 tv 1060: goto release;
1061: mp = &m->m_next;
1062: if (resid <= 0) {
1063: if (flags & MSG_EOR)
1064: top->m_flags |= M_EOR;
1065: break;
1066: }
1067: } while (space > 0 && atomic);
1.108 perry 1068:
1.160 ad 1069: if (so->so_state & SS_CANTSENDMORE) {
1070: error = EPIPE;
1071: goto release;
1072: }
1.45 tv 1073: if (dontroute)
1074: so->so_options |= SO_DONTROUTE;
1075: if (resid > 0)
1076: so->so_state |= SS_MORETOCOME;
1.46 sommerfe 1077: error = (*so->so_proto->pr_usrreq)(so,
1078: (flags & MSG_OOB) ? PRU_SENDOOB : PRU_SEND,
1.160 ad 1079: top, addr, control, curlwp);
1.45 tv 1080: if (dontroute)
1081: so->so_options &= ~SO_DONTROUTE;
1082: if (resid > 0)
1083: so->so_state &= ~SS_MORETOCOME;
1084: clen = 0;
1.144 dyoung 1085: control = NULL;
1086: top = NULL;
1.45 tv 1087: mp = ⊤
1.144 dyoung 1088: if (error != 0)
1.1 cgd 1089: goto release;
1090: } while (resid && space > 0);
1091: } while (resid);
1092:
1.54 lukem 1093: release:
1.1 cgd 1094: sbunlock(&so->so_snd);
1.54 lukem 1095: out:
1.160 ad 1096: sounlock(so);
1097: splx(s);
1.1 cgd 1098: if (top)
1099: m_freem(top);
1100: if (control)
1101: m_freem(control);
1102: return (error);
1103: }
1104:
1105: /*
1.159 ad 1106: * Following replacement or removal of the first mbuf on the first
1107: * mbuf chain of a socket buffer, push necessary state changes back
1108: * into the socket buffer so that other consumers see the values
1109: * consistently. 'nextrecord' is the callers locally stored value of
1110: * the original value of sb->sb_mb->m_nextpkt which must be restored
1111: * when the lead mbuf changes. NOTE: 'nextrecord' may be NULL.
1112: */
1113: static void
1114: sbsync(struct sockbuf *sb, struct mbuf *nextrecord)
1115: {
1116:
1.160 ad 1117: KASSERT(solocked(sb->sb_so));
1118:
1.159 ad 1119: /*
1120: * First, update for the new value of nextrecord. If necessary,
1121: * make it the first record.
1122: */
1123: if (sb->sb_mb != NULL)
1124: sb->sb_mb->m_nextpkt = nextrecord;
1125: else
1126: sb->sb_mb = nextrecord;
1127:
1128: /*
1129: * Now update any dependent socket buffer fields to reflect
1130: * the new state. This is an inline of SB_EMPTY_FIXUP, with
1131: * the addition of a second clause that takes care of the
1132: * case where sb_mb has been updated, but remains the last
1133: * record.
1134: */
1135: if (sb->sb_mb == NULL) {
1136: sb->sb_mbtail = NULL;
1137: sb->sb_lastrecord = NULL;
1138: } else if (sb->sb_mb->m_nextpkt == NULL)
1139: sb->sb_lastrecord = sb->sb_mb;
1140: }
1141:
1142: /*
1.1 cgd 1143: * Implement receive operations on a socket.
1144: * We depend on the way that records are added to the sockbuf
1145: * by sbappend*. In particular, each record (mbufs linked through m_next)
1146: * must begin with an address if the protocol so specifies,
1147: * followed by an optional mbuf or mbufs containing ancillary data,
1148: * and then zero or more mbufs of data.
1149: * In order to avoid blocking network interrupts for the entire time here,
1150: * we splx() while doing the actual copy to user space.
1151: * Although the sockbuf is locked, new data may still be appended,
1152: * and thus we must maintain consistency of the sockbuf during that time.
1153: *
1154: * The caller may receive the data as a single mbuf chain by supplying
1155: * an mbuf **mp0 for use in returning the chain. The uio is then used
1156: * only for the count in uio_resid.
1157: */
1.3 andrew 1158: int
1.54 lukem 1159: soreceive(struct socket *so, struct mbuf **paddr, struct uio *uio,
1160: struct mbuf **mp0, struct mbuf **controlp, int *flagsp)
1.1 cgd 1161: {
1.116 yamt 1162: struct lwp *l = curlwp;
1.160 ad 1163: struct mbuf *m, **mp, *mt;
1.146 dyoung 1164: int atomic, flags, len, error, s, offset, moff, type, orig_resid;
1.99 matt 1165: const struct protosw *pr;
1.54 lukem 1166: struct mbuf *nextrecord;
1.67 he 1167: int mbuf_removed = 0;
1.146 dyoung 1168: const struct domain *dom;
1.196 dsl 1169: short wakeup_state = 0;
1.64 thorpej 1170:
1.54 lukem 1171: pr = so->so_proto;
1.146 dyoung 1172: atomic = pr->pr_flags & PR_ATOMIC;
1173: dom = pr->pr_domain;
1.1 cgd 1174: mp = mp0;
1.54 lukem 1175: type = 0;
1176: orig_resid = uio->uio_resid;
1.102 jonathan 1177:
1.144 dyoung 1178: if (paddr != NULL)
1179: *paddr = NULL;
1180: if (controlp != NULL)
1181: *controlp = NULL;
1182: if (flagsp != NULL)
1.1 cgd 1183: flags = *flagsp &~ MSG_EOR;
1184: else
1185: flags = 0;
1.66 enami 1186:
1187: if ((flags & MSG_DONTWAIT) == 0)
1.117 yamt 1188: sodopendfree();
1.66 enami 1189:
1.1 cgd 1190: if (flags & MSG_OOB) {
1191: m = m_get(M_WAIT, MT_DATA);
1.160 ad 1192: solock(so);
1.17 cgd 1193: error = (*pr->pr_usrreq)(so, PRU_RCVOOB, m,
1.140 dyoung 1194: (struct mbuf *)(long)(flags & MSG_PEEK), NULL, l);
1.160 ad 1195: sounlock(so);
1.1 cgd 1196: if (error)
1197: goto bad;
1198: do {
1.134 christos 1199: error = uiomove(mtod(m, void *),
1.1 cgd 1200: (int) min(uio->uio_resid, m->m_len), uio);
1201: m = m_free(m);
1.144 dyoung 1202: } while (uio->uio_resid > 0 && error == 0 && m);
1.54 lukem 1203: bad:
1.144 dyoung 1204: if (m != NULL)
1.1 cgd 1205: m_freem(m);
1.144 dyoung 1206: return error;
1.1 cgd 1207: }
1.144 dyoung 1208: if (mp != NULL)
1.140 dyoung 1209: *mp = NULL;
1.160 ad 1210:
1211: /*
1212: * solock() provides atomicity of access. splsoftnet() prevents
1213: * protocol processing soft interrupts from interrupting us and
1214: * blocking (expensive).
1215: */
1216: s = splsoftnet();
1217: solock(so);
1.1 cgd 1218: if (so->so_state & SS_ISCONFIRMING && uio->uio_resid)
1.140 dyoung 1219: (*pr->pr_usrreq)(so, PRU_RCVD, NULL, NULL, NULL, l);
1.1 cgd 1220:
1.54 lukem 1221: restart:
1.160 ad 1222: if ((error = sblock(&so->so_rcv, SBLOCKWAIT(flags))) != 0) {
1223: sounlock(so);
1224: splx(s);
1.144 dyoung 1225: return error;
1.160 ad 1226: }
1.1 cgd 1227:
1228: m = so->so_rcv.sb_mb;
1229: /*
1230: * If we have less data than requested, block awaiting more
1231: * (subject to any timeout) if:
1.15 mycroft 1232: * 1. the current count is less than the low water mark,
1.1 cgd 1233: * 2. MSG_WAITALL is set, and it is possible to do the entire
1.15 mycroft 1234: * receive operation at once if we block (resid <= hiwat), or
1235: * 3. MSG_DONTWAIT is not set.
1.1 cgd 1236: * If MSG_WAITALL is set but resid is larger than the receive buffer,
1237: * we have to do the receive in sections, and thus risk returning
1238: * a short count if a timeout or signal occurs after we start.
1239: */
1.144 dyoung 1240: if (m == NULL ||
1241: ((flags & MSG_DONTWAIT) == 0 &&
1242: so->so_rcv.sb_cc < uio->uio_resid &&
1243: (so->so_rcv.sb_cc < so->so_rcv.sb_lowat ||
1244: ((flags & MSG_WAITALL) &&
1245: uio->uio_resid <= so->so_rcv.sb_hiwat)) &&
1.146 dyoung 1246: m->m_nextpkt == NULL && !atomic)) {
1.1 cgd 1247: #ifdef DIAGNOSTIC
1.144 dyoung 1248: if (m == NULL && so->so_rcv.sb_cc)
1.1 cgd 1249: panic("receive 1");
1250: #endif
1251: if (so->so_error) {
1.144 dyoung 1252: if (m != NULL)
1.15 mycroft 1253: goto dontblock;
1.1 cgd 1254: error = so->so_error;
1255: if ((flags & MSG_PEEK) == 0)
1256: so->so_error = 0;
1257: goto release;
1258: }
1259: if (so->so_state & SS_CANTRCVMORE) {
1.144 dyoung 1260: if (m != NULL)
1.15 mycroft 1261: goto dontblock;
1.1 cgd 1262: else
1263: goto release;
1264: }
1.144 dyoung 1265: for (; m != NULL; m = m->m_next)
1.1 cgd 1266: if (m->m_type == MT_OOBDATA || (m->m_flags & M_EOR)) {
1267: m = so->so_rcv.sb_mb;
1268: goto dontblock;
1269: }
1270: if ((so->so_state & (SS_ISCONNECTED|SS_ISCONNECTING)) == 0 &&
1271: (so->so_proto->pr_flags & PR_CONNREQUIRED)) {
1272: error = ENOTCONN;
1273: goto release;
1274: }
1275: if (uio->uio_resid == 0)
1276: goto release;
1.151 ad 1277: if (so->so_nbio || (flags & MSG_DONTWAIT)) {
1.1 cgd 1278: error = EWOULDBLOCK;
1279: goto release;
1280: }
1.69 thorpej 1281: SBLASTRECORDCHK(&so->so_rcv, "soreceive sbwait 1");
1282: SBLASTMBUFCHK(&so->so_rcv, "soreceive sbwait 1");
1.1 cgd 1283: sbunlock(&so->so_rcv);
1.196 dsl 1284: if (wakeup_state & SS_RESTARTSYS)
1285: error = ERESTART;
1286: else
1287: error = sbwait(&so->so_rcv);
1.160 ad 1288: if (error != 0) {
1289: sounlock(so);
1290: splx(s);
1.144 dyoung 1291: return error;
1.160 ad 1292: }
1.196 dsl 1293: wakeup_state = so->so_state;
1.1 cgd 1294: goto restart;
1295: }
1.54 lukem 1296: dontblock:
1.69 thorpej 1297: /*
1298: * On entry here, m points to the first record of the socket buffer.
1.159 ad 1299: * From this point onward, we maintain 'nextrecord' as a cache of the
1300: * pointer to the next record in the socket buffer. We must keep the
1301: * various socket buffer pointers and local stack versions of the
1302: * pointers in sync, pushing out modifications before dropping the
1.160 ad 1303: * socket lock, and re-reading them when picking it up.
1.159 ad 1304: *
1305: * Otherwise, we will race with the network stack appending new data
1306: * or records onto the socket buffer by using inconsistent/stale
1307: * versions of the field, possibly resulting in socket buffer
1308: * corruption.
1309: *
1310: * By holding the high-level sblock(), we prevent simultaneous
1311: * readers from pulling off the front of the socket buffer.
1.69 thorpej 1312: */
1.144 dyoung 1313: if (l != NULL)
1.157 ad 1314: l->l_ru.ru_msgrcv++;
1.69 thorpej 1315: KASSERT(m == so->so_rcv.sb_mb);
1316: SBLASTRECORDCHK(&so->so_rcv, "soreceive 1");
1317: SBLASTMBUFCHK(&so->so_rcv, "soreceive 1");
1.1 cgd 1318: nextrecord = m->m_nextpkt;
1319: if (pr->pr_flags & PR_ADDR) {
1320: #ifdef DIAGNOSTIC
1321: if (m->m_type != MT_SONAME)
1322: panic("receive 1a");
1323: #endif
1.3 andrew 1324: orig_resid = 0;
1.1 cgd 1325: if (flags & MSG_PEEK) {
1326: if (paddr)
1327: *paddr = m_copy(m, 0, m->m_len);
1328: m = m->m_next;
1329: } else {
1330: sbfree(&so->so_rcv, m);
1.67 he 1331: mbuf_removed = 1;
1.144 dyoung 1332: if (paddr != NULL) {
1.1 cgd 1333: *paddr = m;
1334: so->so_rcv.sb_mb = m->m_next;
1.144 dyoung 1335: m->m_next = NULL;
1.1 cgd 1336: m = so->so_rcv.sb_mb;
1337: } else {
1338: MFREE(m, so->so_rcv.sb_mb);
1339: m = so->so_rcv.sb_mb;
1340: }
1.159 ad 1341: sbsync(&so->so_rcv, nextrecord);
1.1 cgd 1342: }
1343: }
1.159 ad 1344:
1345: /*
1346: * Process one or more MT_CONTROL mbufs present before any data mbufs
1347: * in the first mbuf chain on the socket buffer. If MSG_PEEK, we
1348: * just copy the data; if !MSG_PEEK, we call into the protocol to
1349: * perform externalization (or freeing if controlp == NULL).
1350: */
1351: if (__predict_false(m != NULL && m->m_type == MT_CONTROL)) {
1352: struct mbuf *cm = NULL, *cmn;
1353: struct mbuf **cme = &cm;
1354:
1355: do {
1356: if (flags & MSG_PEEK) {
1357: if (controlp != NULL) {
1358: *controlp = m_copy(m, 0, m->m_len);
1359: controlp = &(*controlp)->m_next;
1360: }
1361: m = m->m_next;
1362: } else {
1363: sbfree(&so->so_rcv, m);
1.1 cgd 1364: so->so_rcv.sb_mb = m->m_next;
1.144 dyoung 1365: m->m_next = NULL;
1.159 ad 1366: *cme = m;
1367: cme = &(*cme)->m_next;
1.1 cgd 1368: m = so->so_rcv.sb_mb;
1.159 ad 1369: }
1370: } while (m != NULL && m->m_type == MT_CONTROL);
1371: if ((flags & MSG_PEEK) == 0)
1372: sbsync(&so->so_rcv, nextrecord);
1373: for (; cm != NULL; cm = cmn) {
1374: cmn = cm->m_next;
1375: cm->m_next = NULL;
1376: type = mtod(cm, struct cmsghdr *)->cmsg_type;
1377: if (controlp != NULL) {
1378: if (dom->dom_externalize != NULL &&
1379: type == SCM_RIGHTS) {
1.160 ad 1380: sounlock(so);
1.159 ad 1381: splx(s);
1382: error = (*dom->dom_externalize)(cm, l);
1383: s = splsoftnet();
1.160 ad 1384: solock(so);
1.159 ad 1385: }
1386: *controlp = cm;
1387: while (*controlp != NULL)
1388: controlp = &(*controlp)->m_next;
1.1 cgd 1389: } else {
1.106 itojun 1390: /*
1391: * Dispose of any SCM_RIGHTS message that went
1392: * through the read path rather than recv.
1393: */
1.159 ad 1394: if (dom->dom_dispose != NULL &&
1395: type == SCM_RIGHTS) {
1.160 ad 1396: sounlock(so);
1.159 ad 1397: (*dom->dom_dispose)(cm);
1.160 ad 1398: solock(so);
1.159 ad 1399: }
1400: m_freem(cm);
1.1 cgd 1401: }
1402: }
1.159 ad 1403: if (m != NULL)
1404: nextrecord = so->so_rcv.sb_mb->m_nextpkt;
1405: else
1406: nextrecord = so->so_rcv.sb_mb;
1407: orig_resid = 0;
1.1 cgd 1408: }
1.69 thorpej 1409:
1.159 ad 1410: /* If m is non-NULL, we have some data to read. */
1411: if (__predict_true(m != NULL)) {
1.1 cgd 1412: type = m->m_type;
1413: if (type == MT_OOBDATA)
1414: flags |= MSG_OOB;
1415: }
1.69 thorpej 1416: SBLASTRECORDCHK(&so->so_rcv, "soreceive 2");
1417: SBLASTMBUFCHK(&so->so_rcv, "soreceive 2");
1418:
1.1 cgd 1419: moff = 0;
1420: offset = 0;
1.144 dyoung 1421: while (m != NULL && uio->uio_resid > 0 && error == 0) {
1.1 cgd 1422: if (m->m_type == MT_OOBDATA) {
1423: if (type != MT_OOBDATA)
1424: break;
1425: } else if (type == MT_OOBDATA)
1426: break;
1427: #ifdef DIAGNOSTIC
1428: else if (m->m_type != MT_DATA && m->m_type != MT_HEADER)
1429: panic("receive 3");
1430: #endif
1431: so->so_state &= ~SS_RCVATMARK;
1.196 dsl 1432: wakeup_state = 0;
1.1 cgd 1433: len = uio->uio_resid;
1434: if (so->so_oobmark && len > so->so_oobmark - offset)
1435: len = so->so_oobmark - offset;
1436: if (len > m->m_len - moff)
1437: len = m->m_len - moff;
1438: /*
1439: * If mp is set, just pass back the mbufs.
1440: * Otherwise copy them out via the uio, then free.
1441: * Sockbuf must be consistent here (points to current mbuf,
1442: * it points to next record) when we drop priority;
1443: * we must note any additions to the sockbuf when we
1444: * block interrupts again.
1445: */
1.144 dyoung 1446: if (mp == NULL) {
1.69 thorpej 1447: SBLASTRECORDCHK(&so->so_rcv, "soreceive uiomove");
1448: SBLASTMBUFCHK(&so->so_rcv, "soreceive uiomove");
1.160 ad 1449: sounlock(so);
1.1 cgd 1450: splx(s);
1.134 christos 1451: error = uiomove(mtod(m, char *) + moff, (int)len, uio);
1.20 mycroft 1452: s = splsoftnet();
1.160 ad 1453: solock(so);
1.144 dyoung 1454: if (error != 0) {
1.67 he 1455: /*
1456: * If any part of the record has been removed
1457: * (such as the MT_SONAME mbuf, which will
1458: * happen when PR_ADDR, and thus also
1459: * PR_ATOMIC, is set), then drop the entire
1460: * record to maintain the atomicity of the
1461: * receive operation.
1462: *
1463: * This avoids a later panic("receive 1a")
1464: * when compiled with DIAGNOSTIC.
1465: */
1.146 dyoung 1466: if (m && mbuf_removed && atomic)
1.67 he 1467: (void) sbdroprecord(&so->so_rcv);
1468:
1.57 jdolecek 1469: goto release;
1.67 he 1470: }
1.1 cgd 1471: } else
1472: uio->uio_resid -= len;
1473: if (len == m->m_len - moff) {
1474: if (m->m_flags & M_EOR)
1475: flags |= MSG_EOR;
1476: if (flags & MSG_PEEK) {
1477: m = m->m_next;
1478: moff = 0;
1479: } else {
1480: nextrecord = m->m_nextpkt;
1481: sbfree(&so->so_rcv, m);
1482: if (mp) {
1483: *mp = m;
1484: mp = &m->m_next;
1485: so->so_rcv.sb_mb = m = m->m_next;
1.140 dyoung 1486: *mp = NULL;
1.1 cgd 1487: } else {
1488: MFREE(m, so->so_rcv.sb_mb);
1489: m = so->so_rcv.sb_mb;
1490: }
1.69 thorpej 1491: /*
1492: * If m != NULL, we also know that
1493: * so->so_rcv.sb_mb != NULL.
1494: */
1495: KASSERT(so->so_rcv.sb_mb == m);
1496: if (m) {
1.1 cgd 1497: m->m_nextpkt = nextrecord;
1.69 thorpej 1498: if (nextrecord == NULL)
1499: so->so_rcv.sb_lastrecord = m;
1500: } else {
1501: so->so_rcv.sb_mb = nextrecord;
1.70 thorpej 1502: SB_EMPTY_FIXUP(&so->so_rcv);
1.69 thorpej 1503: }
1504: SBLASTRECORDCHK(&so->so_rcv, "soreceive 3");
1505: SBLASTMBUFCHK(&so->so_rcv, "soreceive 3");
1.1 cgd 1506: }
1.144 dyoung 1507: } else if (flags & MSG_PEEK)
1508: moff += len;
1509: else {
1.160 ad 1510: if (mp != NULL) {
1511: mt = m_copym(m, 0, len, M_NOWAIT);
1512: if (__predict_false(mt == NULL)) {
1513: sounlock(so);
1514: mt = m_copym(m, 0, len, M_WAIT);
1515: solock(so);
1516: }
1517: *mp = mt;
1518: }
1.144 dyoung 1519: m->m_data += len;
1520: m->m_len -= len;
1521: so->so_rcv.sb_cc -= len;
1.1 cgd 1522: }
1523: if (so->so_oobmark) {
1524: if ((flags & MSG_PEEK) == 0) {
1525: so->so_oobmark -= len;
1526: if (so->so_oobmark == 0) {
1527: so->so_state |= SS_RCVATMARK;
1528: break;
1529: }
1.7 cgd 1530: } else {
1.1 cgd 1531: offset += len;
1.7 cgd 1532: if (offset == so->so_oobmark)
1533: break;
1534: }
1.1 cgd 1535: }
1536: if (flags & MSG_EOR)
1537: break;
1538: /*
1539: * If the MSG_WAITALL flag is set (for non-atomic socket),
1540: * we must not quit until "uio->uio_resid == 0" or an error
1541: * termination. If a signal/timeout occurs, return
1542: * with a short count but without error.
1543: * Keep sockbuf locked against other readers.
1544: */
1.144 dyoung 1545: while (flags & MSG_WAITALL && m == NULL && uio->uio_resid > 0 &&
1.3 andrew 1546: !sosendallatonce(so) && !nextrecord) {
1.1 cgd 1547: if (so->so_error || so->so_state & SS_CANTRCVMORE)
1548: break;
1.68 matt 1549: /*
1550: * If we are peeking and the socket receive buffer is
1551: * full, stop since we can't get more data to peek at.
1552: */
1553: if ((flags & MSG_PEEK) && sbspace(&so->so_rcv) <= 0)
1554: break;
1555: /*
1556: * If we've drained the socket buffer, tell the
1557: * protocol in case it needs to do something to
1558: * get it filled again.
1559: */
1560: if ((pr->pr_flags & PR_WANTRCVD) && so->so_pcb)
1561: (*pr->pr_usrreq)(so, PRU_RCVD,
1.140 dyoung 1562: NULL, (struct mbuf *)(long)flags, NULL, l);
1.69 thorpej 1563: SBLASTRECORDCHK(&so->so_rcv, "soreceive sbwait 2");
1564: SBLASTMBUFCHK(&so->so_rcv, "soreceive sbwait 2");
1.196 dsl 1565: if (wakeup_state & SS_RESTARTSYS)
1566: error = ERESTART;
1567: else
1568: error = sbwait(&so->so_rcv);
1.144 dyoung 1569: if (error != 0) {
1.1 cgd 1570: sbunlock(&so->so_rcv);
1.160 ad 1571: sounlock(so);
1.1 cgd 1572: splx(s);
1.144 dyoung 1573: return 0;
1.1 cgd 1574: }
1.21 christos 1575: if ((m = so->so_rcv.sb_mb) != NULL)
1.1 cgd 1576: nextrecord = m->m_nextpkt;
1.196 dsl 1577: wakeup_state = so->so_state;
1.1 cgd 1578: }
1579: }
1.3 andrew 1580:
1.146 dyoung 1581: if (m && atomic) {
1.3 andrew 1582: flags |= MSG_TRUNC;
1583: if ((flags & MSG_PEEK) == 0)
1584: (void) sbdroprecord(&so->so_rcv);
1585: }
1.1 cgd 1586: if ((flags & MSG_PEEK) == 0) {
1.144 dyoung 1587: if (m == NULL) {
1.69 thorpej 1588: /*
1.70 thorpej 1589: * First part is an inline SB_EMPTY_FIXUP(). Second
1.69 thorpej 1590: * part makes sure sb_lastrecord is up-to-date if
1591: * there is still data in the socket buffer.
1592: */
1.1 cgd 1593: so->so_rcv.sb_mb = nextrecord;
1.69 thorpej 1594: if (so->so_rcv.sb_mb == NULL) {
1595: so->so_rcv.sb_mbtail = NULL;
1596: so->so_rcv.sb_lastrecord = NULL;
1597: } else if (nextrecord->m_nextpkt == NULL)
1598: so->so_rcv.sb_lastrecord = nextrecord;
1599: }
1600: SBLASTRECORDCHK(&so->so_rcv, "soreceive 4");
1601: SBLASTMBUFCHK(&so->so_rcv, "soreceive 4");
1.1 cgd 1602: if (pr->pr_flags & PR_WANTRCVD && so->so_pcb)
1.140 dyoung 1603: (*pr->pr_usrreq)(so, PRU_RCVD, NULL,
1604: (struct mbuf *)(long)flags, NULL, l);
1.1 cgd 1605: }
1.3 andrew 1606: if (orig_resid == uio->uio_resid && orig_resid &&
1607: (flags & MSG_EOR) == 0 && (so->so_state & SS_CANTRCVMORE) == 0) {
1608: sbunlock(&so->so_rcv);
1609: goto restart;
1610: }
1.108 perry 1611:
1.144 dyoung 1612: if (flagsp != NULL)
1.1 cgd 1613: *flagsp |= flags;
1.54 lukem 1614: release:
1.1 cgd 1615: sbunlock(&so->so_rcv);
1.160 ad 1616: sounlock(so);
1.1 cgd 1617: splx(s);
1.144 dyoung 1618: return error;
1.1 cgd 1619: }
1620:
1.14 mycroft 1621: int
1.54 lukem 1622: soshutdown(struct socket *so, int how)
1.1 cgd 1623: {
1.99 matt 1624: const struct protosw *pr;
1.160 ad 1625: int error;
1626:
1627: KASSERT(solocked(so));
1.34 kleink 1628:
1.54 lukem 1629: pr = so->so_proto;
1.34 kleink 1630: if (!(how == SHUT_RD || how == SHUT_WR || how == SHUT_RDWR))
1631: return (EINVAL);
1.1 cgd 1632:
1.160 ad 1633: if (how == SHUT_RD || how == SHUT_RDWR) {
1.1 cgd 1634: sorflush(so);
1.160 ad 1635: error = 0;
1636: }
1.34 kleink 1637: if (how == SHUT_WR || how == SHUT_RDWR)
1.160 ad 1638: error = (*pr->pr_usrreq)(so, PRU_SHUTDOWN, NULL,
1.140 dyoung 1639: NULL, NULL, NULL);
1.160 ad 1640:
1641: return error;
1.1 cgd 1642: }
1643:
1.195 dsl 1644: void
1.196 dsl 1645: sorestart(struct socket *so)
1.188 ad 1646: {
1.196 dsl 1647: /*
1648: * An application has called close() on an fd on which another
1649: * of its threads has called a socket system call.
1650: * Mark this and wake everyone up, and code that would block again
1651: * instead returns ERESTART.
1652: * On system call re-entry the fd is validated and EBADF returned.
1653: * Any other fd will block again on the 2nd syscall.
1654: */
1.188 ad 1655: solock(so);
1.196 dsl 1656: so->so_state |= SS_RESTARTSYS;
1.188 ad 1657: cv_broadcast(&so->so_cv);
1.196 dsl 1658: cv_broadcast(&so->so_snd.sb_cv);
1659: cv_broadcast(&so->so_rcv.sb_cv);
1.188 ad 1660: sounlock(so);
1661: }
1662:
1.14 mycroft 1663: void
1.54 lukem 1664: sorflush(struct socket *so)
1.1 cgd 1665: {
1.54 lukem 1666: struct sockbuf *sb, asb;
1.99 matt 1667: const struct protosw *pr;
1.160 ad 1668:
1669: KASSERT(solocked(so));
1.1 cgd 1670:
1.54 lukem 1671: sb = &so->so_rcv;
1672: pr = so->so_proto;
1.160 ad 1673: socantrcvmore(so);
1.1 cgd 1674: sb->sb_flags |= SB_NOINTR;
1.160 ad 1675: (void )sblock(sb, M_WAITOK);
1.1 cgd 1676: sbunlock(sb);
1677: asb = *sb;
1.86 wrstuden 1678: /*
1679: * Clear most of the sockbuf structure, but leave some of the
1680: * fields valid.
1681: */
1682: memset(&sb->sb_startzero, 0,
1683: sizeof(*sb) - offsetof(struct sockbuf, sb_startzero));
1.160 ad 1684: if (pr->pr_flags & PR_RIGHTS && pr->pr_domain->dom_dispose) {
1685: sounlock(so);
1.1 cgd 1686: (*pr->pr_domain->dom_dispose)(asb.sb_mb);
1.160 ad 1687: solock(so);
1688: }
1.98 christos 1689: sbrelease(&asb, so);
1.1 cgd 1690: }
1691:
1.171 plunky 1692: /*
1693: * internal set SOL_SOCKET options
1694: */
1.142 dyoung 1695: static int
1.171 plunky 1696: sosetopt1(struct socket *so, const struct sockopt *sopt)
1.1 cgd 1697: {
1.182 christos 1698: int error = EINVAL, optval, opt;
1.171 plunky 1699: struct linger l;
1700: struct timeval tv;
1.142 dyoung 1701:
1.179 christos 1702: switch ((opt = sopt->sopt_name)) {
1.142 dyoung 1703:
1.170 tls 1704: case SO_ACCEPTFILTER:
1.177 ad 1705: error = accept_filt_setopt(so, sopt);
1706: KASSERT(solocked(so));
1.170 tls 1707: break;
1708:
1.171 plunky 1709: case SO_LINGER:
1710: error = sockopt_get(sopt, &l, sizeof(l));
1.177 ad 1711: solock(so);
1.171 plunky 1712: if (error)
1.177 ad 1713: break;
1.171 plunky 1714: if (l.l_linger < 0 || l.l_linger > USHRT_MAX ||
1.177 ad 1715: l.l_linger > (INT_MAX / hz)) {
1716: error = EDOM;
1717: break;
1718: }
1.171 plunky 1719: so->so_linger = l.l_linger;
1720: if (l.l_onoff)
1721: so->so_options |= SO_LINGER;
1722: else
1723: so->so_options &= ~SO_LINGER;
1.177 ad 1724: break;
1.1 cgd 1725:
1.142 dyoung 1726: case SO_DEBUG:
1727: case SO_KEEPALIVE:
1728: case SO_DONTROUTE:
1729: case SO_USELOOPBACK:
1730: case SO_BROADCAST:
1731: case SO_REUSEADDR:
1732: case SO_REUSEPORT:
1733: case SO_OOBINLINE:
1734: case SO_TIMESTAMP:
1.184 christos 1735: #ifdef SO_OTIMESTAMP
1736: case SO_OTIMESTAMP:
1737: #endif
1.171 plunky 1738: error = sockopt_getint(sopt, &optval);
1.177 ad 1739: solock(so);
1.171 plunky 1740: if (error)
1.177 ad 1741: break;
1.171 plunky 1742: if (optval)
1.179 christos 1743: so->so_options |= opt;
1.142 dyoung 1744: else
1.179 christos 1745: so->so_options &= ~opt;
1.142 dyoung 1746: break;
1747:
1748: case SO_SNDBUF:
1749: case SO_RCVBUF:
1750: case SO_SNDLOWAT:
1751: case SO_RCVLOWAT:
1.171 plunky 1752: error = sockopt_getint(sopt, &optval);
1.177 ad 1753: solock(so);
1.171 plunky 1754: if (error)
1.177 ad 1755: break;
1.1 cgd 1756:
1.142 dyoung 1757: /*
1758: * Values < 1 make no sense for any of these
1759: * options, so disallow them.
1760: */
1.177 ad 1761: if (optval < 1) {
1762: error = EINVAL;
1763: break;
1764: }
1.1 cgd 1765:
1.179 christos 1766: switch (opt) {
1.171 plunky 1767: case SO_SNDBUF:
1.177 ad 1768: if (sbreserve(&so->so_snd, (u_long)optval, so) == 0) {
1769: error = ENOBUFS;
1770: break;
1771: }
1.171 plunky 1772: so->so_snd.sb_flags &= ~SB_AUTOSIZE;
1773: break;
1.1 cgd 1774:
1775: case SO_RCVBUF:
1.177 ad 1776: if (sbreserve(&so->so_rcv, (u_long)optval, so) == 0) {
1777: error = ENOBUFS;
1778: break;
1779: }
1.171 plunky 1780: so->so_rcv.sb_flags &= ~SB_AUTOSIZE;
1.142 dyoung 1781: break;
1782:
1783: /*
1784: * Make sure the low-water is never greater than
1785: * the high-water.
1786: */
1.1 cgd 1787: case SO_SNDLOWAT:
1.171 plunky 1788: if (optval > so->so_snd.sb_hiwat)
1789: optval = so->so_snd.sb_hiwat;
1790:
1791: so->so_snd.sb_lowat = optval;
1.142 dyoung 1792: break;
1.171 plunky 1793:
1.1 cgd 1794: case SO_RCVLOWAT:
1.171 plunky 1795: if (optval > so->so_rcv.sb_hiwat)
1796: optval = so->so_rcv.sb_hiwat;
1797:
1798: so->so_rcv.sb_lowat = optval;
1.142 dyoung 1799: break;
1800: }
1801: break;
1.28 thorpej 1802:
1.179 christos 1803: #ifdef COMPAT_50
1804: case SO_OSNDTIMEO:
1805: case SO_ORCVTIMEO: {
1806: struct timeval50 otv;
1807: error = sockopt_get(sopt, &otv, sizeof(otv));
1.186 pooka 1808: if (error) {
1809: solock(so);
1.183 christos 1810: break;
1.186 pooka 1811: }
1.179 christos 1812: timeval50_to_timeval(&otv, &tv);
1813: opt = opt == SO_OSNDTIMEO ? SO_SNDTIMEO : SO_RCVTIMEO;
1.182 christos 1814: error = 0;
1.179 christos 1815: /*FALLTHROUGH*/
1816: }
1817: #endif /* COMPAT_50 */
1818:
1.142 dyoung 1819: case SO_SNDTIMEO:
1820: case SO_RCVTIMEO:
1.182 christos 1821: if (error)
1.179 christos 1822: error = sockopt_get(sopt, &tv, sizeof(tv));
1.177 ad 1823: solock(so);
1.171 plunky 1824: if (error)
1.177 ad 1825: break;
1.171 plunky 1826:
1.177 ad 1827: if (tv.tv_sec > (INT_MAX - tv.tv_usec / tick) / hz) {
1828: error = EDOM;
1829: break;
1830: }
1.28 thorpej 1831:
1.171 plunky 1832: optval = tv.tv_sec * hz + tv.tv_usec / tick;
1833: if (optval == 0 && tv.tv_usec != 0)
1834: optval = 1;
1.28 thorpej 1835:
1.179 christos 1836: switch (opt) {
1.142 dyoung 1837: case SO_SNDTIMEO:
1.171 plunky 1838: so->so_snd.sb_timeo = optval;
1.1 cgd 1839: break;
1840: case SO_RCVTIMEO:
1.171 plunky 1841: so->so_rcv.sb_timeo = optval;
1.142 dyoung 1842: break;
1843: }
1844: break;
1.1 cgd 1845:
1.142 dyoung 1846: default:
1.177 ad 1847: solock(so);
1848: error = ENOPROTOOPT;
1849: break;
1.142 dyoung 1850: }
1.177 ad 1851: KASSERT(solocked(so));
1852: return error;
1.142 dyoung 1853: }
1.1 cgd 1854:
1.142 dyoung 1855: int
1.171 plunky 1856: sosetopt(struct socket *so, struct sockopt *sopt)
1.142 dyoung 1857: {
1858: int error, prerr;
1.1 cgd 1859:
1.177 ad 1860: if (sopt->sopt_level == SOL_SOCKET) {
1.171 plunky 1861: error = sosetopt1(so, sopt);
1.177 ad 1862: KASSERT(solocked(so));
1863: } else {
1.142 dyoung 1864: error = ENOPROTOOPT;
1.177 ad 1865: solock(so);
1866: }
1.1 cgd 1867:
1.142 dyoung 1868: if ((error == 0 || error == ENOPROTOOPT) &&
1869: so->so_proto != NULL && so->so_proto->pr_ctloutput != NULL) {
1870: /* give the protocol stack a shot */
1.171 plunky 1871: prerr = (*so->so_proto->pr_ctloutput)(PRCO_SETOPT, so, sopt);
1.142 dyoung 1872: if (prerr == 0)
1873: error = 0;
1874: else if (prerr != ENOPROTOOPT)
1875: error = prerr;
1.171 plunky 1876: }
1.160 ad 1877: sounlock(so);
1.142 dyoung 1878: return error;
1.1 cgd 1879: }
1880:
1.171 plunky 1881: /*
1882: * so_setsockopt() is a wrapper providing a sockopt structure for sosetopt()
1883: */
1884: int
1885: so_setsockopt(struct lwp *l, struct socket *so, int level, int name,
1886: const void *val, size_t valsize)
1887: {
1888: struct sockopt sopt;
1889: int error;
1890:
1891: KASSERT(valsize == 0 || val != NULL);
1892:
1893: sockopt_init(&sopt, level, name, valsize);
1894: sockopt_set(&sopt, val, valsize);
1895:
1896: error = sosetopt(so, &sopt);
1897:
1898: sockopt_destroy(&sopt);
1899:
1900: return error;
1901: }
1902:
1903: /*
1904: * internal get SOL_SOCKET options
1905: */
1906: static int
1907: sogetopt1(struct socket *so, struct sockopt *sopt)
1908: {
1.179 christos 1909: int error, optval, opt;
1.171 plunky 1910: struct linger l;
1911: struct timeval tv;
1912:
1.179 christos 1913: switch ((opt = sopt->sopt_name)) {
1.171 plunky 1914:
1915: case SO_ACCEPTFILTER:
1.177 ad 1916: error = accept_filt_getopt(so, sopt);
1.171 plunky 1917: break;
1918:
1919: case SO_LINGER:
1920: l.l_onoff = (so->so_options & SO_LINGER) ? 1 : 0;
1921: l.l_linger = so->so_linger;
1922:
1923: error = sockopt_set(sopt, &l, sizeof(l));
1924: break;
1925:
1926: case SO_USELOOPBACK:
1927: case SO_DONTROUTE:
1928: case SO_DEBUG:
1929: case SO_KEEPALIVE:
1930: case SO_REUSEADDR:
1931: case SO_REUSEPORT:
1932: case SO_BROADCAST:
1933: case SO_OOBINLINE:
1934: case SO_TIMESTAMP:
1.184 christos 1935: #ifdef SO_OTIMESTAMP
1936: case SO_OTIMESTAMP:
1937: #endif
1.179 christos 1938: error = sockopt_setint(sopt, (so->so_options & opt) ? 1 : 0);
1.171 plunky 1939: break;
1940:
1941: case SO_TYPE:
1942: error = sockopt_setint(sopt, so->so_type);
1943: break;
1944:
1945: case SO_ERROR:
1946: error = sockopt_setint(sopt, so->so_error);
1947: so->so_error = 0;
1948: break;
1949:
1950: case SO_SNDBUF:
1951: error = sockopt_setint(sopt, so->so_snd.sb_hiwat);
1952: break;
1953:
1954: case SO_RCVBUF:
1955: error = sockopt_setint(sopt, so->so_rcv.sb_hiwat);
1956: break;
1957:
1958: case SO_SNDLOWAT:
1959: error = sockopt_setint(sopt, so->so_snd.sb_lowat);
1960: break;
1961:
1962: case SO_RCVLOWAT:
1963: error = sockopt_setint(sopt, so->so_rcv.sb_lowat);
1964: break;
1965:
1.179 christos 1966: #ifdef COMPAT_50
1967: case SO_OSNDTIMEO:
1968: case SO_ORCVTIMEO: {
1969: struct timeval50 otv;
1970:
1971: optval = (opt == SO_OSNDTIMEO ?
1972: so->so_snd.sb_timeo : so->so_rcv.sb_timeo);
1973:
1974: otv.tv_sec = optval / hz;
1975: otv.tv_usec = (optval % hz) * tick;
1976:
1977: error = sockopt_set(sopt, &otv, sizeof(otv));
1978: break;
1979: }
1980: #endif /* COMPAT_50 */
1981:
1.171 plunky 1982: case SO_SNDTIMEO:
1983: case SO_RCVTIMEO:
1.179 christos 1984: optval = (opt == SO_SNDTIMEO ?
1.171 plunky 1985: so->so_snd.sb_timeo : so->so_rcv.sb_timeo);
1986:
1987: tv.tv_sec = optval / hz;
1988: tv.tv_usec = (optval % hz) * tick;
1989:
1990: error = sockopt_set(sopt, &tv, sizeof(tv));
1991: break;
1992:
1993: case SO_OVERFLOWED:
1994: error = sockopt_setint(sopt, so->so_rcv.sb_overflowed);
1995: break;
1996:
1997: default:
1998: error = ENOPROTOOPT;
1999: break;
2000: }
2001:
2002: return (error);
2003: }
2004:
1.14 mycroft 2005: int
1.171 plunky 2006: sogetopt(struct socket *so, struct sockopt *sopt)
1.1 cgd 2007: {
1.160 ad 2008: int error;
1.1 cgd 2009:
1.160 ad 2010: solock(so);
1.171 plunky 2011: if (sopt->sopt_level != SOL_SOCKET) {
1.1 cgd 2012: if (so->so_proto && so->so_proto->pr_ctloutput) {
1.160 ad 2013: error = ((*so->so_proto->pr_ctloutput)
1.171 plunky 2014: (PRCO_GETOPT, so, sopt));
1.1 cgd 2015: } else
1.160 ad 2016: error = (ENOPROTOOPT);
1.1 cgd 2017: } else {
1.171 plunky 2018: error = sogetopt1(so, sopt);
2019: }
2020: sounlock(so);
2021: return (error);
2022: }
2023:
2024: /*
2025: * alloc sockopt data buffer buffer
2026: * - will be released at destroy
2027: */
1.176 plunky 2028: static int
2029: sockopt_alloc(struct sockopt *sopt, size_t len, km_flag_t kmflag)
1.171 plunky 2030: {
2031:
2032: KASSERT(sopt->sopt_size == 0);
2033:
1.176 plunky 2034: if (len > sizeof(sopt->sopt_buf)) {
2035: sopt->sopt_data = kmem_zalloc(len, kmflag);
2036: if (sopt->sopt_data == NULL)
2037: return ENOMEM;
2038: } else
1.171 plunky 2039: sopt->sopt_data = sopt->sopt_buf;
2040:
2041: sopt->sopt_size = len;
1.176 plunky 2042: return 0;
1.171 plunky 2043: }
2044:
2045: /*
2046: * initialise sockopt storage
1.176 plunky 2047: * - MAY sleep during allocation
1.171 plunky 2048: */
2049: void
2050: sockopt_init(struct sockopt *sopt, int level, int name, size_t size)
2051: {
1.1 cgd 2052:
1.171 plunky 2053: memset(sopt, 0, sizeof(*sopt));
1.1 cgd 2054:
1.171 plunky 2055: sopt->sopt_level = level;
2056: sopt->sopt_name = name;
1.176 plunky 2057: (void)sockopt_alloc(sopt, size, KM_SLEEP);
1.171 plunky 2058: }
2059:
2060: /*
2061: * destroy sockopt storage
2062: * - will release any held memory references
2063: */
2064: void
2065: sockopt_destroy(struct sockopt *sopt)
2066: {
2067:
2068: if (sopt->sopt_data != sopt->sopt_buf)
1.173 plunky 2069: kmem_free(sopt->sopt_data, sopt->sopt_size);
1.171 plunky 2070:
2071: memset(sopt, 0, sizeof(*sopt));
2072: }
2073:
2074: /*
2075: * set sockopt value
2076: * - value is copied into sockopt
1.176 plunky 2077: * - memory is allocated when necessary, will not sleep
1.171 plunky 2078: */
2079: int
2080: sockopt_set(struct sockopt *sopt, const void *buf, size_t len)
2081: {
1.176 plunky 2082: int error;
1.171 plunky 2083:
1.176 plunky 2084: if (sopt->sopt_size == 0) {
2085: error = sockopt_alloc(sopt, len, KM_NOSLEEP);
2086: if (error)
2087: return error;
2088: }
1.171 plunky 2089:
2090: KASSERT(sopt->sopt_size == len);
2091: memcpy(sopt->sopt_data, buf, len);
2092: return 0;
2093: }
2094:
2095: /*
2096: * common case of set sockopt integer value
2097: */
2098: int
2099: sockopt_setint(struct sockopt *sopt, int val)
2100: {
2101:
2102: return sockopt_set(sopt, &val, sizeof(int));
2103: }
2104:
2105: /*
2106: * get sockopt value
2107: * - correct size must be given
2108: */
2109: int
2110: sockopt_get(const struct sockopt *sopt, void *buf, size_t len)
2111: {
1.170 tls 2112:
1.171 plunky 2113: if (sopt->sopt_size != len)
2114: return EINVAL;
1.1 cgd 2115:
1.171 plunky 2116: memcpy(buf, sopt->sopt_data, len);
2117: return 0;
2118: }
1.1 cgd 2119:
1.171 plunky 2120: /*
2121: * common case of get sockopt integer value
2122: */
2123: int
2124: sockopt_getint(const struct sockopt *sopt, int *valp)
2125: {
1.1 cgd 2126:
1.171 plunky 2127: return sockopt_get(sopt, valp, sizeof(int));
2128: }
1.1 cgd 2129:
1.171 plunky 2130: /*
2131: * set sockopt value from mbuf
2132: * - ONLY for legacy code
2133: * - mbuf is released by sockopt
1.176 plunky 2134: * - will not sleep
1.171 plunky 2135: */
2136: int
2137: sockopt_setmbuf(struct sockopt *sopt, struct mbuf *m)
2138: {
2139: size_t len;
1.176 plunky 2140: int error;
1.1 cgd 2141:
1.171 plunky 2142: len = m_length(m);
1.1 cgd 2143:
1.176 plunky 2144: if (sopt->sopt_size == 0) {
2145: error = sockopt_alloc(sopt, len, KM_NOSLEEP);
2146: if (error)
2147: return error;
2148: }
1.1 cgd 2149:
1.171 plunky 2150: KASSERT(sopt->sopt_size == len);
2151: m_copydata(m, 0, len, sopt->sopt_data);
2152: m_freem(m);
1.1 cgd 2153:
1.171 plunky 2154: return 0;
2155: }
1.1 cgd 2156:
1.171 plunky 2157: /*
2158: * get sockopt value into mbuf
2159: * - ONLY for legacy code
2160: * - mbuf to be released by the caller
1.176 plunky 2161: * - will not sleep
1.171 plunky 2162: */
2163: struct mbuf *
2164: sockopt_getmbuf(const struct sockopt *sopt)
2165: {
2166: struct mbuf *m;
1.107 darrenr 2167:
1.176 plunky 2168: if (sopt->sopt_size > MCLBYTES)
2169: return NULL;
2170:
2171: m = m_get(M_DONTWAIT, MT_SOOPTS);
1.171 plunky 2172: if (m == NULL)
2173: return NULL;
2174:
1.176 plunky 2175: if (sopt->sopt_size > MLEN) {
2176: MCLGET(m, M_DONTWAIT);
2177: if ((m->m_flags & M_EXT) == 0) {
2178: m_free(m);
2179: return NULL;
2180: }
1.1 cgd 2181: }
1.176 plunky 2182:
2183: memcpy(mtod(m, void *), sopt->sopt_data, sopt->sopt_size);
2184: m->m_len = sopt->sopt_size;
1.160 ad 2185:
1.171 plunky 2186: return m;
1.1 cgd 2187: }
2188:
1.14 mycroft 2189: void
1.54 lukem 2190: sohasoutofband(struct socket *so)
1.1 cgd 2191: {
1.153 rmind 2192:
1.90 christos 2193: fownsignal(so->so_pgid, SIGURG, POLL_PRI, POLLPRI|POLLRDBAND, so);
1.189 ad 2194: selnotify(&so->so_rcv.sb_sel, POLLPRI | POLLRDBAND, NOTE_SUBMIT);
1.1 cgd 2195: }
1.72 jdolecek 2196:
2197: static void
2198: filt_sordetach(struct knote *kn)
2199: {
2200: struct socket *so;
2201:
1.155 ad 2202: so = ((file_t *)kn->kn_obj)->f_data;
1.160 ad 2203: solock(so);
1.73 christos 2204: SLIST_REMOVE(&so->so_rcv.sb_sel.sel_klist, kn, knote, kn_selnext);
2205: if (SLIST_EMPTY(&so->so_rcv.sb_sel.sel_klist))
1.72 jdolecek 2206: so->so_rcv.sb_flags &= ~SB_KNOTE;
1.160 ad 2207: sounlock(so);
1.72 jdolecek 2208: }
2209:
2210: /*ARGSUSED*/
2211: static int
1.129 yamt 2212: filt_soread(struct knote *kn, long hint)
1.72 jdolecek 2213: {
2214: struct socket *so;
1.160 ad 2215: int rv;
1.72 jdolecek 2216:
1.155 ad 2217: so = ((file_t *)kn->kn_obj)->f_data;
1.160 ad 2218: if (hint != NOTE_SUBMIT)
2219: solock(so);
1.72 jdolecek 2220: kn->kn_data = so->so_rcv.sb_cc;
2221: if (so->so_state & SS_CANTRCVMORE) {
1.108 perry 2222: kn->kn_flags |= EV_EOF;
1.72 jdolecek 2223: kn->kn_fflags = so->so_error;
1.160 ad 2224: rv = 1;
2225: } else if (so->so_error) /* temporary udp error */
2226: rv = 1;
2227: else if (kn->kn_sfflags & NOTE_LOWAT)
2228: rv = (kn->kn_data >= kn->kn_sdata);
2229: else
2230: rv = (kn->kn_data >= so->so_rcv.sb_lowat);
2231: if (hint != NOTE_SUBMIT)
2232: sounlock(so);
2233: return rv;
1.72 jdolecek 2234: }
2235:
2236: static void
2237: filt_sowdetach(struct knote *kn)
2238: {
2239: struct socket *so;
2240:
1.155 ad 2241: so = ((file_t *)kn->kn_obj)->f_data;
1.160 ad 2242: solock(so);
1.73 christos 2243: SLIST_REMOVE(&so->so_snd.sb_sel.sel_klist, kn, knote, kn_selnext);
2244: if (SLIST_EMPTY(&so->so_snd.sb_sel.sel_klist))
1.72 jdolecek 2245: so->so_snd.sb_flags &= ~SB_KNOTE;
1.160 ad 2246: sounlock(so);
1.72 jdolecek 2247: }
2248:
2249: /*ARGSUSED*/
2250: static int
1.129 yamt 2251: filt_sowrite(struct knote *kn, long hint)
1.72 jdolecek 2252: {
2253: struct socket *so;
1.160 ad 2254: int rv;
1.72 jdolecek 2255:
1.155 ad 2256: so = ((file_t *)kn->kn_obj)->f_data;
1.160 ad 2257: if (hint != NOTE_SUBMIT)
2258: solock(so);
1.72 jdolecek 2259: kn->kn_data = sbspace(&so->so_snd);
2260: if (so->so_state & SS_CANTSENDMORE) {
1.108 perry 2261: kn->kn_flags |= EV_EOF;
1.72 jdolecek 2262: kn->kn_fflags = so->so_error;
1.160 ad 2263: rv = 1;
2264: } else if (so->so_error) /* temporary udp error */
2265: rv = 1;
2266: else if (((so->so_state & SS_ISCONNECTED) == 0) &&
1.72 jdolecek 2267: (so->so_proto->pr_flags & PR_CONNREQUIRED))
1.160 ad 2268: rv = 0;
2269: else if (kn->kn_sfflags & NOTE_LOWAT)
2270: rv = (kn->kn_data >= kn->kn_sdata);
2271: else
2272: rv = (kn->kn_data >= so->so_snd.sb_lowat);
2273: if (hint != NOTE_SUBMIT)
2274: sounlock(so);
2275: return rv;
1.72 jdolecek 2276: }
2277:
2278: /*ARGSUSED*/
2279: static int
1.129 yamt 2280: filt_solisten(struct knote *kn, long hint)
1.72 jdolecek 2281: {
2282: struct socket *so;
1.160 ad 2283: int rv;
1.72 jdolecek 2284:
1.155 ad 2285: so = ((file_t *)kn->kn_obj)->f_data;
1.72 jdolecek 2286:
2287: /*
2288: * Set kn_data to number of incoming connections, not
2289: * counting partial (incomplete) connections.
1.108 perry 2290: */
1.160 ad 2291: if (hint != NOTE_SUBMIT)
2292: solock(so);
1.72 jdolecek 2293: kn->kn_data = so->so_qlen;
1.160 ad 2294: rv = (kn->kn_data > 0);
2295: if (hint != NOTE_SUBMIT)
2296: sounlock(so);
2297: return rv;
1.72 jdolecek 2298: }
2299:
2300: static const struct filterops solisten_filtops =
2301: { 1, NULL, filt_sordetach, filt_solisten };
2302: static const struct filterops soread_filtops =
2303: { 1, NULL, filt_sordetach, filt_soread };
2304: static const struct filterops sowrite_filtops =
2305: { 1, NULL, filt_sowdetach, filt_sowrite };
2306:
2307: int
1.129 yamt 2308: soo_kqfilter(struct file *fp, struct knote *kn)
1.72 jdolecek 2309: {
2310: struct socket *so;
2311: struct sockbuf *sb;
2312:
1.155 ad 2313: so = ((file_t *)kn->kn_obj)->f_data;
1.160 ad 2314: solock(so);
1.72 jdolecek 2315: switch (kn->kn_filter) {
2316: case EVFILT_READ:
2317: if (so->so_options & SO_ACCEPTCONN)
2318: kn->kn_fop = &solisten_filtops;
2319: else
2320: kn->kn_fop = &soread_filtops;
2321: sb = &so->so_rcv;
2322: break;
2323: case EVFILT_WRITE:
2324: kn->kn_fop = &sowrite_filtops;
2325: sb = &so->so_snd;
2326: break;
2327: default:
1.160 ad 2328: sounlock(so);
1.149 pooka 2329: return (EINVAL);
1.72 jdolecek 2330: }
1.73 christos 2331: SLIST_INSERT_HEAD(&sb->sb_sel.sel_klist, kn, kn_selnext);
1.72 jdolecek 2332: sb->sb_flags |= SB_KNOTE;
1.160 ad 2333: sounlock(so);
1.72 jdolecek 2334: return (0);
2335: }
2336:
1.154 ad 2337: static int
2338: sodopoll(struct socket *so, int events)
2339: {
2340: int revents;
2341:
2342: revents = 0;
2343:
2344: if (events & (POLLIN | POLLRDNORM))
2345: if (soreadable(so))
2346: revents |= events & (POLLIN | POLLRDNORM);
2347:
2348: if (events & (POLLOUT | POLLWRNORM))
2349: if (sowritable(so))
2350: revents |= events & (POLLOUT | POLLWRNORM);
2351:
2352: if (events & (POLLPRI | POLLRDBAND))
2353: if (so->so_oobmark || (so->so_state & SS_RCVATMARK))
2354: revents |= events & (POLLPRI | POLLRDBAND);
2355:
2356: return revents;
2357: }
2358:
2359: int
2360: sopoll(struct socket *so, int events)
2361: {
2362: int revents = 0;
2363:
1.160 ad 2364: #ifndef DIAGNOSTIC
2365: /*
2366: * Do a quick, unlocked check in expectation that the socket
2367: * will be ready for I/O. Don't do this check if DIAGNOSTIC,
2368: * as the solocked() assertions will fail.
2369: */
1.154 ad 2370: if ((revents = sodopoll(so, events)) != 0)
2371: return revents;
1.160 ad 2372: #endif
1.154 ad 2373:
1.160 ad 2374: solock(so);
1.154 ad 2375: if ((revents = sodopoll(so, events)) == 0) {
2376: if (events & (POLLIN | POLLPRI | POLLRDNORM | POLLRDBAND)) {
2377: selrecord(curlwp, &so->so_rcv.sb_sel);
1.160 ad 2378: so->so_rcv.sb_flags |= SB_NOTIFY;
1.154 ad 2379: }
2380:
2381: if (events & (POLLOUT | POLLWRNORM)) {
2382: selrecord(curlwp, &so->so_snd.sb_sel);
1.160 ad 2383: so->so_snd.sb_flags |= SB_NOTIFY;
1.154 ad 2384: }
2385: }
1.160 ad 2386: sounlock(so);
1.154 ad 2387:
2388: return revents;
2389: }
2390:
2391:
1.94 yamt 2392: #include <sys/sysctl.h>
2393:
2394: static int sysctl_kern_somaxkva(SYSCTLFN_PROTO);
2395:
2396: /*
2397: * sysctl helper routine for kern.somaxkva. ensures that the given
2398: * value is not too small.
2399: * (XXX should we maybe make sure it's not too large as well?)
2400: */
2401: static int
2402: sysctl_kern_somaxkva(SYSCTLFN_ARGS)
2403: {
2404: int error, new_somaxkva;
2405: struct sysctlnode node;
2406:
2407: new_somaxkva = somaxkva;
2408: node = *rnode;
2409: node.sysctl_data = &new_somaxkva;
2410: error = sysctl_lookup(SYSCTLFN_CALL(&node));
2411: if (error || newp == NULL)
2412: return (error);
2413:
2414: if (new_somaxkva < (16 * 1024 * 1024)) /* sanity */
2415: return (EINVAL);
2416:
1.136 ad 2417: mutex_enter(&so_pendfree_lock);
1.94 yamt 2418: somaxkva = new_somaxkva;
1.136 ad 2419: cv_broadcast(&socurkva_cv);
2420: mutex_exit(&so_pendfree_lock);
1.94 yamt 2421:
2422: return (error);
2423: }
2424:
1.178 pooka 2425: static void
1.187 cegger 2426: sysctl_kern_somaxkva_setup(void)
1.94 yamt 2427: {
2428:
1.178 pooka 2429: KASSERT(socket_sysctllog == NULL);
2430: sysctl_createv(&socket_sysctllog, 0, NULL, NULL,
1.97 atatat 2431: CTLFLAG_PERMANENT,
2432: CTLTYPE_NODE, "kern", NULL,
2433: NULL, 0, NULL, 0,
2434: CTL_KERN, CTL_EOL);
2435:
1.178 pooka 2436: sysctl_createv(&socket_sysctllog, 0, NULL, NULL,
1.97 atatat 2437: CTLFLAG_PERMANENT|CTLFLAG_READWRITE,
1.103 atatat 2438: CTLTYPE_INT, "somaxkva",
2439: SYSCTL_DESCR("Maximum amount of kernel memory to be "
2440: "used for socket buffers"),
1.94 yamt 2441: sysctl_kern_somaxkva, 0, NULL, 0,
2442: CTL_KERN, KERN_SOMAXKVA, CTL_EOL);
2443: }
CVSweb <webmaster@jp.NetBSD.org>