Annotation of src/sys/kern/uipc_socket.c, Revision 1.177.4.2.2.1
1.177.4.2.2.1! (bouyer 1:: /* $NetBSD$ */
1.64 thorpej 2:
3: /*-
1.177.4.2 snj 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
1.177.4.2 snj 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>
1.177.4.2.2.1! (bouyer 66:: __KERNEL_RCSID(0, "$NetBSD$");
1.64 thorpej 67:
68: #include "opt_sock_counters.h"
69: #include "opt_sosend_loan.h"
1.81 martin 70: #include "opt_mbuftrace.h"
1.84 ragge 71: #include "opt_somaxkva.h"
1.167 ad 72: #include "opt_multiprocessor.h" /* XXX */
1.1 cgd 73:
1.9 mycroft 74: #include <sys/param.h>
75: #include <sys/systm.h>
76: #include <sys/proc.h>
77: #include <sys/file.h>
1.142 dyoung 78: #include <sys/filedesc.h>
1.173 plunky 79: #include <sys/kmem.h>
1.9 mycroft 80: #include <sys/mbuf.h>
81: #include <sys/domain.h>
82: #include <sys/kernel.h>
83: #include <sys/protosw.h>
84: #include <sys/socket.h>
85: #include <sys/socketvar.h>
1.21 christos 86: #include <sys/signalvar.h>
1.9 mycroft 87: #include <sys/resourcevar.h>
1.174 pooka 88: #include <sys/uidinfo.h>
1.72 jdolecek 89: #include <sys/event.h>
1.89 christos 90: #include <sys/poll.h>
1.118 elad 91: #include <sys/kauth.h>
1.136 ad 92: #include <sys/mutex.h>
93: #include <sys/condvar.h>
1.37 thorpej 94:
1.64 thorpej 95: #include <uvm/uvm.h>
96:
1.77 thorpej 97: MALLOC_DEFINE(M_SOOPTS, "soopts", "socket options");
98: MALLOC_DEFINE(M_SONAME, "soname", "socket name");
1.37 thorpej 99:
1.142 dyoung 100: extern const struct fileops socketops;
101:
1.54 lukem 102: extern int somaxconn; /* patchable (XXX sysctl) */
103: int somaxconn = SOMAXCONN;
1.160 ad 104: kmutex_t *softnet_lock;
1.49 jonathan 105:
1.64 thorpej 106: #ifdef SOSEND_COUNTERS
107: #include <sys/device.h>
108:
1.113 thorpej 109: static struct evcnt sosend_loan_big = EVCNT_INITIALIZER(EVCNT_TYPE_MISC,
1.64 thorpej 110: NULL, "sosend", "loan big");
1.113 thorpej 111: static struct evcnt sosend_copy_big = EVCNT_INITIALIZER(EVCNT_TYPE_MISC,
1.64 thorpej 112: NULL, "sosend", "copy big");
1.113 thorpej 113: static struct evcnt sosend_copy_small = EVCNT_INITIALIZER(EVCNT_TYPE_MISC,
1.64 thorpej 114: NULL, "sosend", "copy small");
1.113 thorpej 115: static struct evcnt sosend_kvalimit = EVCNT_INITIALIZER(EVCNT_TYPE_MISC,
1.64 thorpej 116: NULL, "sosend", "kva limit");
117:
118: #define SOSEND_COUNTER_INCR(ev) (ev)->ev_count++
119:
1.101 matt 120: EVCNT_ATTACH_STATIC(sosend_loan_big);
121: EVCNT_ATTACH_STATIC(sosend_copy_big);
122: EVCNT_ATTACH_STATIC(sosend_copy_small);
123: EVCNT_ATTACH_STATIC(sosend_kvalimit);
1.64 thorpej 124: #else
125:
126: #define SOSEND_COUNTER_INCR(ev) /* nothing */
127:
128: #endif /* SOSEND_COUNTERS */
129:
1.119 yamt 130: static struct callback_entry sokva_reclaimerentry;
1.1 cgd 131:
1.167 ad 132: #if defined(SOSEND_NO_LOAN) || defined(MULTIPROCESSOR)
1.121 yamt 133: int sock_loan_thresh = -1;
1.71 thorpej 134: #else
1.121 yamt 135: int sock_loan_thresh = 4096;
1.65 thorpej 136: #endif
1.64 thorpej 137:
1.136 ad 138: static kmutex_t so_pendfree_lock;
1.113 thorpej 139: static struct mbuf *so_pendfree;
1.64 thorpej 140:
1.84 ragge 141: #ifndef SOMAXKVA
142: #define SOMAXKVA (16 * 1024 * 1024)
143: #endif
144: int somaxkva = SOMAXKVA;
1.113 thorpej 145: static int socurkva;
1.136 ad 146: static kcondvar_t socurkva_cv;
1.64 thorpej 147:
148: #define SOCK_LOAN_CHUNK 65536
149:
1.117 yamt 150: static size_t sodopendfree(void);
151: static size_t sodopendfreel(void);
1.93 yamt 152:
1.113 thorpej 153: static vsize_t
1.129 yamt 154: sokvareserve(struct socket *so, vsize_t len)
1.80 yamt 155: {
1.98 christos 156: int error;
1.80 yamt 157:
1.136 ad 158: mutex_enter(&so_pendfree_lock);
1.80 yamt 159: while (socurkva + len > somaxkva) {
1.93 yamt 160: size_t freed;
161:
162: /*
163: * try to do pendfree.
164: */
165:
1.117 yamt 166: freed = sodopendfreel();
1.93 yamt 167:
168: /*
169: * if some kva was freed, try again.
170: */
171:
172: if (freed)
1.80 yamt 173: continue;
1.93 yamt 174:
1.80 yamt 175: SOSEND_COUNTER_INCR(&sosend_kvalimit);
1.136 ad 176: error = cv_wait_sig(&socurkva_cv, &so_pendfree_lock);
1.98 christos 177: if (error) {
178: len = 0;
179: break;
180: }
1.80 yamt 181: }
1.93 yamt 182: socurkva += len;
1.136 ad 183: mutex_exit(&so_pendfree_lock);
1.98 christos 184: return len;
1.95 yamt 185: }
186:
1.113 thorpej 187: static void
1.95 yamt 188: sokvaunreserve(vsize_t len)
189: {
190:
1.136 ad 191: mutex_enter(&so_pendfree_lock);
1.95 yamt 192: socurkva -= len;
1.136 ad 193: cv_broadcast(&socurkva_cv);
194: mutex_exit(&so_pendfree_lock);
1.95 yamt 195: }
196:
197: /*
198: * sokvaalloc: allocate kva for loan.
199: */
200:
201: vaddr_t
202: sokvaalloc(vsize_t len, struct socket *so)
203: {
204: vaddr_t lva;
205:
206: /*
207: * reserve kva.
208: */
209:
1.98 christos 210: if (sokvareserve(so, len) == 0)
211: return 0;
1.93 yamt 212:
213: /*
214: * allocate kva.
215: */
1.80 yamt 216:
1.109 yamt 217: lva = uvm_km_alloc(kernel_map, len, 0, UVM_KMF_VAONLY | UVM_KMF_WAITVA);
1.95 yamt 218: if (lva == 0) {
219: sokvaunreserve(len);
1.80 yamt 220: return (0);
1.95 yamt 221: }
1.80 yamt 222:
223: return lva;
224: }
225:
1.93 yamt 226: /*
227: * sokvafree: free kva for loan.
228: */
229:
1.80 yamt 230: void
231: sokvafree(vaddr_t sva, vsize_t len)
232: {
1.93 yamt 233:
234: /*
235: * free kva.
236: */
1.80 yamt 237:
1.109 yamt 238: uvm_km_free(kernel_map, sva, len, UVM_KMF_VAONLY);
1.93 yamt 239:
240: /*
241: * unreserve kva.
242: */
243:
1.95 yamt 244: sokvaunreserve(len);
1.80 yamt 245: }
246:
1.64 thorpej 247: static void
1.134 christos 248: sodoloanfree(struct vm_page **pgs, void *buf, size_t size)
1.64 thorpej 249: {
1.156 yamt 250: vaddr_t sva, eva;
1.64 thorpej 251: vsize_t len;
1.156 yamt 252: int npgs;
253:
254: KASSERT(pgs != NULL);
1.64 thorpej 255:
256: eva = round_page((vaddr_t) buf + size);
257: sva = trunc_page((vaddr_t) buf);
258: len = eva - sva;
259: npgs = len >> PAGE_SHIFT;
260:
261: pmap_kremove(sva, len);
262: pmap_update(pmap_kernel());
263: uvm_unloan(pgs, npgs, UVM_LOAN_TOPAGE);
1.80 yamt 264: sokvafree(sva, len);
1.64 thorpej 265: }
266:
267: static size_t
1.152 matt 268: sodopendfree(void)
1.64 thorpej 269: {
1.93 yamt 270: size_t rv;
1.64 thorpej 271:
1.160 ad 272: if (__predict_true(so_pendfree == NULL))
273: return 0;
274:
1.136 ad 275: mutex_enter(&so_pendfree_lock);
1.117 yamt 276: rv = sodopendfreel();
1.136 ad 277: mutex_exit(&so_pendfree_lock);
1.93 yamt 278:
279: return rv;
280: }
281:
282: /*
283: * sodopendfreel: free mbufs on "pendfree" list.
1.136 ad 284: * unlock and relock so_pendfree_lock when freeing mbufs.
1.93 yamt 285: *
1.136 ad 286: * => called with so_pendfree_lock held.
1.93 yamt 287: */
288:
289: static size_t
1.152 matt 290: sodopendfreel(void)
1.93 yamt 291: {
1.137 ad 292: struct mbuf *m, *next;
1.93 yamt 293: size_t rv = 0;
294:
1.136 ad 295: KASSERT(mutex_owned(&so_pendfree_lock));
1.64 thorpej 296:
1.137 ad 297: while (so_pendfree != NULL) {
1.64 thorpej 298: m = so_pendfree;
1.93 yamt 299: so_pendfree = NULL;
1.136 ad 300: mutex_exit(&so_pendfree_lock);
1.93 yamt 301:
302: for (; m != NULL; m = next) {
303: next = m->m_next;
1.156 yamt 304: KASSERT((~m->m_flags & (M_EXT|M_EXT_PAGES)) == 0);
305: KASSERT(m->m_ext.ext_refcnt == 0);
1.93 yamt 306:
307: rv += m->m_ext.ext_size;
1.156 yamt 308: sodoloanfree(m->m_ext.ext_pgs, m->m_ext.ext_buf,
1.93 yamt 309: m->m_ext.ext_size);
1.145 ad 310: pool_cache_put(mb_cache, m);
1.93 yamt 311: }
1.64 thorpej 312:
1.136 ad 313: mutex_enter(&so_pendfree_lock);
1.64 thorpej 314: }
315:
316: return (rv);
317: }
318:
1.80 yamt 319: void
1.134 christos 320: soloanfree(struct mbuf *m, void *buf, size_t size, void *arg)
1.64 thorpej 321: {
322:
1.156 yamt 323: KASSERT(m != NULL);
1.64 thorpej 324:
1.93 yamt 325: /*
326: * postpone freeing mbuf.
327: *
328: * we can't do it in interrupt context
329: * because we need to put kva back to kernel_map.
330: */
331:
1.136 ad 332: mutex_enter(&so_pendfree_lock);
1.92 yamt 333: m->m_next = so_pendfree;
334: so_pendfree = m;
1.136 ad 335: cv_broadcast(&socurkva_cv);
336: mutex_exit(&so_pendfree_lock);
1.64 thorpej 337: }
338:
339: static long
340: sosend_loan(struct socket *so, struct uio *uio, struct mbuf *m, long space)
341: {
342: struct iovec *iov = uio->uio_iov;
343: vaddr_t sva, eva;
344: vsize_t len;
1.156 yamt 345: vaddr_t lva;
346: int npgs, error;
347: vaddr_t va;
348: int i;
1.64 thorpej 349:
1.116 yamt 350: if (VMSPACE_IS_KERNEL_P(uio->uio_vmspace))
1.64 thorpej 351: return (0);
352:
353: if (iov->iov_len < (size_t) space)
354: space = iov->iov_len;
355: if (space > SOCK_LOAN_CHUNK)
356: space = SOCK_LOAN_CHUNK;
357:
358: eva = round_page((vaddr_t) iov->iov_base + space);
359: sva = trunc_page((vaddr_t) iov->iov_base);
360: len = eva - sva;
361: npgs = len >> PAGE_SHIFT;
362:
1.79 thorpej 363: KASSERT(npgs <= M_EXT_MAXPAGES);
364:
1.80 yamt 365: lva = sokvaalloc(len, so);
1.64 thorpej 366: if (lva == 0)
1.80 yamt 367: return 0;
1.64 thorpej 368:
1.116 yamt 369: error = uvm_loan(&uio->uio_vmspace->vm_map, sva, len,
1.79 thorpej 370: m->m_ext.ext_pgs, UVM_LOAN_TOPAGE);
1.64 thorpej 371: if (error) {
1.80 yamt 372: sokvafree(lva, len);
1.64 thorpej 373: return (0);
374: }
375:
376: for (i = 0, va = lva; i < npgs; i++, va += PAGE_SIZE)
1.79 thorpej 377: pmap_kenter_pa(va, VM_PAGE_TO_PHYS(m->m_ext.ext_pgs[i]),
378: VM_PROT_READ);
1.64 thorpej 379: pmap_update(pmap_kernel());
380:
381: lva += (vaddr_t) iov->iov_base & PAGE_MASK;
382:
1.134 christos 383: MEXTADD(m, (void *) lva, space, M_MBUF, soloanfree, so);
1.79 thorpej 384: m->m_flags |= M_EXT_PAGES | M_EXT_ROMAP;
1.64 thorpej 385:
386: uio->uio_resid -= space;
387: /* uio_offset not updated, not set/used for write(2) */
1.134 christos 388: uio->uio_iov->iov_base = (char *)uio->uio_iov->iov_base + space;
1.64 thorpej 389: uio->uio_iov->iov_len -= space;
390: if (uio->uio_iov->iov_len == 0) {
391: uio->uio_iov++;
392: uio->uio_iovcnt--;
393: }
394:
395: return (space);
396: }
397:
1.119 yamt 398: static int
1.129 yamt 399: sokva_reclaim_callback(struct callback_entry *ce, void *obj, void *arg)
1.119 yamt 400: {
401:
402: KASSERT(ce == &sokva_reclaimerentry);
403: KASSERT(obj == NULL);
404:
405: sodopendfree();
406: if (!vm_map_starved_p(kernel_map)) {
407: return CALLBACK_CHAIN_ABORT;
408: }
409: return CALLBACK_CHAIN_CONTINUE;
410: }
411:
1.142 dyoung 412: struct mbuf *
1.147 dyoung 413: getsombuf(struct socket *so, int type)
1.142 dyoung 414: {
415: struct mbuf *m;
416:
1.147 dyoung 417: m = m_get(M_WAIT, type);
1.142 dyoung 418: MCLAIM(m, so->so_mowner);
419: return m;
420: }
421:
1.119 yamt 422: void
423: soinit(void)
424: {
425:
1.148 ad 426: mutex_init(&so_pendfree_lock, MUTEX_DEFAULT, IPL_VM);
1.160 ad 427: softnet_lock = mutex_obj_alloc(MUTEX_DEFAULT, IPL_NONE);
1.136 ad 428: cv_init(&socurkva_cv, "sokva");
1.166 ad 429: soinit2();
1.136 ad 430:
1.119 yamt 431: /* Set the initial adjusted socket buffer size. */
432: if (sb_max_set(sb_max))
433: panic("bad initial sb_max value: %lu", sb_max);
434:
435: callback_register(&vm_map_to_kernel(kernel_map)->vmk_reclaim_callback,
436: &sokva_reclaimerentry, NULL, sokva_reclaim_callback);
437: }
438:
1.1 cgd 439: /*
440: * Socket operation routines.
441: * These routines are called by the routines in
442: * sys_socket.c or from a system process, and
443: * implement the semantics of socket operations by
444: * switching out to the protocol specific routines.
445: */
446: /*ARGSUSED*/
1.3 andrew 447: int
1.160 ad 448: socreate(int dom, struct socket **aso, int type, int proto, struct lwp *l,
449: struct socket *lockso)
1.1 cgd 450: {
1.99 matt 451: const struct protosw *prp;
1.54 lukem 452: struct socket *so;
1.115 yamt 453: uid_t uid;
1.160 ad 454: int error;
455: kmutex_t *lock;
1.1 cgd 456:
1.132 elad 457: error = kauth_authorize_network(l->l_cred, KAUTH_NETWORK_SOCKET,
458: KAUTH_REQ_NETWORK_SOCKET_OPEN, KAUTH_ARG(dom), KAUTH_ARG(type),
459: KAUTH_ARG(proto));
1.140 dyoung 460: if (error != 0)
461: return error;
1.127 elad 462:
1.1 cgd 463: if (proto)
464: prp = pffindproto(dom, proto, type);
465: else
466: prp = pffindtype(dom, type);
1.140 dyoung 467: if (prp == NULL) {
1.120 ginsbach 468: /* no support for domain */
469: if (pffinddomain(dom) == 0)
1.140 dyoung 470: return EAFNOSUPPORT;
1.120 ginsbach 471: /* no support for socket type */
472: if (proto == 0 && type != 0)
1.140 dyoung 473: return EPROTOTYPE;
474: return EPROTONOSUPPORT;
1.120 ginsbach 475: }
1.140 dyoung 476: if (prp->pr_usrreq == NULL)
477: return EPROTONOSUPPORT;
1.1 cgd 478: if (prp->pr_type != type)
1.140 dyoung 479: return EPROTOTYPE;
1.160 ad 480:
481: so = soget(true);
1.1 cgd 482: so->so_type = type;
483: so->so_proto = prp;
1.33 matt 484: so->so_send = sosend;
485: so->so_receive = soreceive;
1.78 matt 486: #ifdef MBUFTRACE
487: so->so_rcv.sb_mowner = &prp->pr_domain->dom_mowner;
488: so->so_snd.sb_mowner = &prp->pr_domain->dom_mowner;
489: so->so_mowner = &prp->pr_domain->dom_mowner;
490: #endif
1.138 rmind 491: uid = kauth_cred_geteuid(l->l_cred);
1.115 yamt 492: so->so_uidinfo = uid_find(uid);
1.168 yamt 493: so->so_egid = kauth_cred_getegid(l->l_cred);
494: so->so_cpid = l->l_proc->p_pid;
1.160 ad 495: if (lockso != NULL) {
496: /* Caller wants us to share a lock. */
497: lock = lockso->so_lock;
498: so->so_lock = lock;
499: mutex_obj_hold(lock);
500: mutex_enter(lock);
501: } else {
502: /* Lock assigned and taken during PRU_ATTACH. */
503: }
1.140 dyoung 504: error = (*prp->pr_usrreq)(so, PRU_ATTACH, NULL,
505: (struct mbuf *)(long)proto, NULL, l);
1.160 ad 506: KASSERT(solocked(so));
1.140 dyoung 507: if (error != 0) {
1.1 cgd 508: so->so_state |= SS_NOFDREF;
509: sofree(so);
1.140 dyoung 510: return error;
1.1 cgd 511: }
1.160 ad 512: sounlock(so);
1.1 cgd 513: *aso = so;
1.140 dyoung 514: return 0;
1.1 cgd 515: }
516:
1.142 dyoung 517: /* On success, write file descriptor to fdout and return zero. On
518: * failure, return non-zero; *fdout will be undefined.
519: */
520: int
521: fsocreate(int domain, struct socket **sop, int type, int protocol,
522: struct lwp *l, int *fdout)
523: {
524: struct socket *so;
525: struct file *fp;
526: int fd, error;
527:
1.155 ad 528: if ((error = fd_allocfile(&fp, &fd)) != 0)
1.142 dyoung 529: return (error);
530: fp->f_flag = FREAD|FWRITE;
531: fp->f_type = DTYPE_SOCKET;
532: fp->f_ops = &socketops;
1.160 ad 533: error = socreate(domain, &so, type, protocol, l, NULL);
1.142 dyoung 534: if (error != 0) {
1.155 ad 535: fd_abort(curproc, fp, fd);
1.142 dyoung 536: } else {
537: if (sop != NULL)
538: *sop = so;
539: fp->f_data = so;
1.155 ad 540: fd_affix(curproc, fp, fd);
1.142 dyoung 541: *fdout = fd;
542: }
543: return error;
544: }
545:
1.3 andrew 546: int
1.114 christos 547: sobind(struct socket *so, struct mbuf *nam, struct lwp *l)
1.1 cgd 548: {
1.160 ad 549: int error;
1.1 cgd 550:
1.160 ad 551: solock(so);
1.140 dyoung 552: error = (*so->so_proto->pr_usrreq)(so, PRU_BIND, NULL, nam, NULL, l);
1.160 ad 553: sounlock(so);
1.140 dyoung 554: return error;
1.1 cgd 555: }
556:
1.3 andrew 557: int
1.150 elad 558: solisten(struct socket *so, int backlog, struct lwp *l)
1.1 cgd 559: {
1.160 ad 560: int error;
1.1 cgd 561:
1.160 ad 562: solock(so);
1.158 ad 563: if ((so->so_state & (SS_ISCONNECTED | SS_ISCONNECTING |
1.163 ad 564: SS_ISDISCONNECTING)) != 0) {
565: sounlock(so);
1.158 ad 566: return (EOPNOTSUPP);
1.163 ad 567: }
1.140 dyoung 568: error = (*so->so_proto->pr_usrreq)(so, PRU_LISTEN, NULL,
1.150 elad 569: NULL, NULL, l);
1.140 dyoung 570: if (error != 0) {
1.160 ad 571: sounlock(so);
1.140 dyoung 572: return error;
1.1 cgd 573: }
1.63 matt 574: if (TAILQ_EMPTY(&so->so_q))
1.1 cgd 575: so->so_options |= SO_ACCEPTCONN;
576: if (backlog < 0)
577: backlog = 0;
1.49 jonathan 578: so->so_qlimit = min(backlog, somaxconn);
1.160 ad 579: sounlock(so);
1.140 dyoung 580: return 0;
1.1 cgd 581: }
582:
1.21 christos 583: void
1.54 lukem 584: sofree(struct socket *so)
1.1 cgd 585: {
1.161 ad 586: u_int refs;
1.1 cgd 587:
1.160 ad 588: KASSERT(solocked(so));
589:
590: if (so->so_pcb || (so->so_state & SS_NOFDREF) == 0) {
591: sounlock(so);
1.1 cgd 592: return;
1.160 ad 593: }
1.43 mycroft 594: if (so->so_head) {
595: /*
596: * We must not decommission a socket that's on the accept(2)
597: * queue. If we do, then accept(2) may hang after select(2)
598: * indicated that the listening socket was ready.
599: */
1.160 ad 600: if (!soqremque(so, 0)) {
601: sounlock(so);
1.43 mycroft 602: return;
1.160 ad 603: }
1.43 mycroft 604: }
1.98 christos 605: if (so->so_rcv.sb_hiwat)
1.110 christos 606: (void)chgsbsize(so->so_uidinfo, &so->so_rcv.sb_hiwat, 0,
1.98 christos 607: RLIM_INFINITY);
608: if (so->so_snd.sb_hiwat)
1.110 christos 609: (void)chgsbsize(so->so_uidinfo, &so->so_snd.sb_hiwat, 0,
1.98 christos 610: RLIM_INFINITY);
611: sbrelease(&so->so_snd, so);
1.160 ad 612: KASSERT(!cv_has_waiters(&so->so_cv));
613: KASSERT(!cv_has_waiters(&so->so_rcv.sb_cv));
614: KASSERT(!cv_has_waiters(&so->so_snd.sb_cv));
1.1 cgd 615: sorflush(so);
1.161 ad 616: refs = so->so_aborting; /* XXX */
1.177 ad 617: /* Remove acccept filter if one is present. */
1.170 tls 618: if (so->so_accf != NULL)
1.177 ad 619: (void)accept_filt_clear(so);
1.160 ad 620: sounlock(so);
1.161 ad 621: if (refs == 0) /* XXX */
622: soput(so);
1.1 cgd 623: }
624:
625: /*
626: * Close a socket on last file table reference removal.
627: * Initiate disconnect if connected.
628: * Free socket when disconnect complete.
629: */
1.3 andrew 630: int
1.54 lukem 631: soclose(struct socket *so)
1.1 cgd 632: {
1.54 lukem 633: struct socket *so2;
1.160 ad 634: int error;
635: int error2;
1.1 cgd 636:
1.54 lukem 637: error = 0;
1.160 ad 638: solock(so);
1.1 cgd 639: if (so->so_options & SO_ACCEPTCONN) {
1.172 ad 640: for (;;) {
641: if ((so2 = TAILQ_FIRST(&so->so_q0)) != 0) {
1.160 ad 642: KASSERT(solocked2(so, so2));
643: (void) soqremque(so2, 0);
644: /* soabort drops the lock. */
645: (void) soabort(so2);
646: solock(so);
1.172 ad 647: continue;
1.160 ad 648: }
1.172 ad 649: if ((so2 = TAILQ_FIRST(&so->so_q)) != 0) {
1.160 ad 650: KASSERT(solocked2(so, so2));
651: (void) soqremque(so2, 1);
652: /* soabort drops the lock. */
653: (void) soabort(so2);
654: solock(so);
1.172 ad 655: continue;
1.160 ad 656: }
1.172 ad 657: break;
658: }
1.1 cgd 659: }
660: if (so->so_pcb == 0)
661: goto discard;
662: if (so->so_state & SS_ISCONNECTED) {
663: if ((so->so_state & SS_ISDISCONNECTING) == 0) {
664: error = sodisconnect(so);
665: if (error)
666: goto drop;
667: }
668: if (so->so_options & SO_LINGER) {
1.151 ad 669: if ((so->so_state & SS_ISDISCONNECTING) && so->so_nbio)
1.1 cgd 670: goto drop;
1.21 christos 671: while (so->so_state & SS_ISCONNECTED) {
1.177.4.1 snj 672: error = sowait(so, true, so->so_linger * hz);
1.21 christos 673: if (error)
1.1 cgd 674: break;
1.21 christos 675: }
1.1 cgd 676: }
677: }
1.54 lukem 678: drop:
1.1 cgd 679: if (so->so_pcb) {
1.160 ad 680: error2 = (*so->so_proto->pr_usrreq)(so, PRU_DETACH,
1.140 dyoung 681: NULL, NULL, NULL, NULL);
1.1 cgd 682: if (error == 0)
683: error = error2;
684: }
1.54 lukem 685: discard:
1.1 cgd 686: if (so->so_state & SS_NOFDREF)
687: panic("soclose: NOFDREF");
688: so->so_state |= SS_NOFDREF;
689: sofree(so);
690: return (error);
691: }
692:
693: /*
1.160 ad 694: * Must be called with the socket locked.. Will return with it unlocked.
1.1 cgd 695: */
1.3 andrew 696: int
1.54 lukem 697: soabort(struct socket *so)
1.1 cgd 698: {
1.161 ad 699: u_int refs;
1.139 yamt 700: int error;
1.160 ad 701:
702: KASSERT(solocked(so));
703: KASSERT(so->so_head == NULL);
1.1 cgd 704:
1.161 ad 705: so->so_aborting++; /* XXX */
1.140 dyoung 706: error = (*so->so_proto->pr_usrreq)(so, PRU_ABORT, NULL,
707: NULL, NULL, NULL);
1.161 ad 708: refs = --so->so_aborting; /* XXX */
1.164 drochner 709: if (error || (refs == 0)) {
1.139 yamt 710: sofree(so);
1.160 ad 711: } else {
712: sounlock(so);
1.139 yamt 713: }
714: return error;
1.1 cgd 715: }
716:
1.3 andrew 717: int
1.54 lukem 718: soaccept(struct socket *so, struct mbuf *nam)
1.1 cgd 719: {
1.160 ad 720: int error;
721:
722: KASSERT(solocked(so));
1.1 cgd 723:
1.54 lukem 724: error = 0;
1.1 cgd 725: if ((so->so_state & SS_NOFDREF) == 0)
726: panic("soaccept: !NOFDREF");
727: so->so_state &= ~SS_NOFDREF;
1.55 thorpej 728: if ((so->so_state & SS_ISDISCONNECTED) == 0 ||
729: (so->so_proto->pr_flags & PR_ABRTACPTDIS) == 0)
1.41 mycroft 730: error = (*so->so_proto->pr_usrreq)(so, PRU_ACCEPT,
1.140 dyoung 731: NULL, nam, NULL, NULL);
1.41 mycroft 732: else
1.53 itojun 733: error = ECONNABORTED;
1.52 itojun 734:
1.1 cgd 735: return (error);
736: }
737:
1.3 andrew 738: int
1.114 christos 739: soconnect(struct socket *so, struct mbuf *nam, struct lwp *l)
1.1 cgd 740: {
1.160 ad 741: int error;
742:
743: KASSERT(solocked(so));
1.1 cgd 744:
745: if (so->so_options & SO_ACCEPTCONN)
746: return (EOPNOTSUPP);
747: /*
748: * If protocol is connection-based, can only connect once.
749: * Otherwise, if connected, try to disconnect first.
750: * This allows user to disconnect by connecting to, e.g.,
751: * a null address.
752: */
753: if (so->so_state & (SS_ISCONNECTED|SS_ISCONNECTING) &&
754: ((so->so_proto->pr_flags & PR_CONNREQUIRED) ||
755: (error = sodisconnect(so))))
756: error = EISCONN;
757: else
758: error = (*so->so_proto->pr_usrreq)(so, PRU_CONNECT,
1.140 dyoung 759: NULL, nam, NULL, l);
1.1 cgd 760: return (error);
761: }
762:
1.3 andrew 763: int
1.54 lukem 764: soconnect2(struct socket *so1, struct socket *so2)
1.1 cgd 765: {
1.160 ad 766: int error;
767:
768: KASSERT(solocked2(so1, so2));
1.1 cgd 769:
1.22 mycroft 770: error = (*so1->so_proto->pr_usrreq)(so1, PRU_CONNECT2,
1.140 dyoung 771: NULL, (struct mbuf *)so2, NULL, NULL);
1.1 cgd 772: return (error);
773: }
774:
1.3 andrew 775: int
1.54 lukem 776: sodisconnect(struct socket *so)
1.1 cgd 777: {
1.160 ad 778: int error;
779:
780: KASSERT(solocked(so));
1.1 cgd 781:
782: if ((so->so_state & SS_ISCONNECTED) == 0) {
783: error = ENOTCONN;
1.160 ad 784: } else if (so->so_state & SS_ISDISCONNECTING) {
1.1 cgd 785: error = EALREADY;
1.160 ad 786: } else {
787: error = (*so->so_proto->pr_usrreq)(so, PRU_DISCONNECT,
788: NULL, NULL, NULL, NULL);
1.1 cgd 789: }
1.117 yamt 790: sodopendfree();
1.1 cgd 791: return (error);
792: }
793:
1.15 mycroft 794: #define SBLOCKWAIT(f) (((f) & MSG_DONTWAIT) ? M_NOWAIT : M_WAITOK)
1.1 cgd 795: /*
796: * Send on a socket.
797: * If send must go all at once and message is larger than
798: * send buffering, then hard error.
799: * Lock against other senders.
800: * If must go all at once and not enough room now, then
801: * inform user that this would block and do nothing.
802: * Otherwise, if nonblocking, send as much as possible.
803: * The data to be sent is described by "uio" if nonzero,
804: * otherwise by the mbuf chain "top" (which must be null
805: * if uio is not). Data provided in mbuf chain must be small
806: * enough to send all at once.
807: *
808: * Returns nonzero on error, timeout or signal; callers
809: * must check for short counts if EINTR/ERESTART are returned.
810: * Data and control buffers are freed on return.
811: */
1.3 andrew 812: int
1.54 lukem 813: sosend(struct socket *so, struct mbuf *addr, struct uio *uio, struct mbuf *top,
1.114 christos 814: struct mbuf *control, int flags, struct lwp *l)
1.1 cgd 815: {
1.54 lukem 816: struct mbuf **mp, *m;
1.114 christos 817: struct proc *p;
1.58 jdolecek 818: long space, len, resid, clen, mlen;
819: int error, s, dontroute, atomic;
1.54 lukem 820:
1.114 christos 821: p = l->l_proc;
1.117 yamt 822: sodopendfree();
1.160 ad 823: clen = 0;
1.64 thorpej 824:
1.160 ad 825: /*
826: * solock() provides atomicity of access. splsoftnet() prevents
827: * protocol processing soft interrupts from interrupting us and
828: * blocking (expensive).
829: */
830: s = splsoftnet();
831: solock(so);
1.54 lukem 832: atomic = sosendallatonce(so) || top;
1.1 cgd 833: if (uio)
834: resid = uio->uio_resid;
835: else
836: resid = top->m_pkthdr.len;
1.7 cgd 837: /*
838: * In theory resid should be unsigned.
839: * However, space must be signed, as it might be less than 0
840: * if we over-committed, and we must use a signed comparison
841: * of space and resid. On the other hand, a negative resid
842: * causes us to loop sending 0-length segments to the protocol.
843: */
1.29 mycroft 844: if (resid < 0) {
845: error = EINVAL;
846: goto out;
847: }
1.1 cgd 848: dontroute =
849: (flags & MSG_DONTROUTE) && (so->so_options & SO_DONTROUTE) == 0 &&
850: (so->so_proto->pr_flags & PR_ATOMIC);
1.165 christos 851: l->l_ru.ru_msgsnd++;
1.1 cgd 852: if (control)
853: clen = control->m_len;
1.54 lukem 854: restart:
1.21 christos 855: if ((error = sblock(&so->so_snd, SBLOCKWAIT(flags))) != 0)
1.1 cgd 856: goto out;
857: do {
1.160 ad 858: if (so->so_state & SS_CANTSENDMORE) {
859: error = EPIPE;
860: goto release;
861: }
1.48 thorpej 862: if (so->so_error) {
863: error = so->so_error;
864: so->so_error = 0;
865: goto release;
866: }
1.1 cgd 867: if ((so->so_state & SS_ISCONNECTED) == 0) {
868: if (so->so_proto->pr_flags & PR_CONNREQUIRED) {
869: if ((so->so_state & SS_ISCONFIRMING) == 0 &&
1.160 ad 870: !(resid == 0 && clen != 0)) {
871: error = ENOTCONN;
872: goto release;
873: }
874: } else if (addr == 0) {
875: error = EDESTADDRREQ;
876: goto release;
877: }
1.1 cgd 878: }
879: space = sbspace(&so->so_snd);
880: if (flags & MSG_OOB)
881: space += 1024;
1.21 christos 882: if ((atomic && resid > so->so_snd.sb_hiwat) ||
1.160 ad 883: clen > so->so_snd.sb_hiwat) {
884: error = EMSGSIZE;
885: goto release;
886: }
1.96 mycroft 887: if (space < resid + clen &&
1.1 cgd 888: (atomic || space < so->so_snd.sb_lowat || space < clen)) {
1.160 ad 889: if (so->so_nbio) {
890: error = EWOULDBLOCK;
891: goto release;
892: }
1.1 cgd 893: sbunlock(&so->so_snd);
894: error = sbwait(&so->so_snd);
895: if (error)
896: goto out;
897: goto restart;
898: }
899: mp = ⊤
900: space -= clen;
901: do {
1.45 tv 902: if (uio == NULL) {
903: /*
904: * Data is prepackaged in "top".
905: */
906: resid = 0;
907: if (flags & MSG_EOR)
908: top->m_flags |= M_EOR;
909: } else do {
1.160 ad 910: sounlock(so);
911: splx(s);
1.144 dyoung 912: if (top == NULL) {
1.78 matt 913: m = m_gethdr(M_WAIT, MT_DATA);
1.45 tv 914: mlen = MHLEN;
915: m->m_pkthdr.len = 0;
1.140 dyoung 916: m->m_pkthdr.rcvif = NULL;
1.45 tv 917: } else {
1.78 matt 918: m = m_get(M_WAIT, MT_DATA);
1.45 tv 919: mlen = MLEN;
920: }
1.78 matt 921: MCLAIM(m, so->so_snd.sb_mowner);
1.121 yamt 922: if (sock_loan_thresh >= 0 &&
923: uio->uio_iov->iov_len >= sock_loan_thresh &&
924: space >= sock_loan_thresh &&
1.64 thorpej 925: (len = sosend_loan(so, uio, m,
926: space)) != 0) {
927: SOSEND_COUNTER_INCR(&sosend_loan_big);
928: space -= len;
929: goto have_data;
930: }
1.45 tv 931: if (resid >= MINCLSIZE && space >= MCLBYTES) {
1.64 thorpej 932: SOSEND_COUNTER_INCR(&sosend_copy_big);
1.78 matt 933: m_clget(m, M_WAIT);
1.45 tv 934: if ((m->m_flags & M_EXT) == 0)
935: goto nopages;
936: mlen = MCLBYTES;
937: if (atomic && top == 0) {
1.58 jdolecek 938: len = lmin(MCLBYTES - max_hdr,
1.54 lukem 939: resid);
1.45 tv 940: m->m_data += max_hdr;
941: } else
1.58 jdolecek 942: len = lmin(MCLBYTES, resid);
1.45 tv 943: space -= len;
944: } else {
1.64 thorpej 945: nopages:
946: SOSEND_COUNTER_INCR(&sosend_copy_small);
1.58 jdolecek 947: len = lmin(lmin(mlen, resid), space);
1.45 tv 948: space -= len;
949: /*
950: * For datagram protocols, leave room
951: * for protocol headers in first mbuf.
952: */
953: if (atomic && top == 0 && len < mlen)
954: MH_ALIGN(m, len);
955: }
1.144 dyoung 956: error = uiomove(mtod(m, void *), (int)len, uio);
1.64 thorpej 957: have_data:
1.45 tv 958: resid = uio->uio_resid;
959: m->m_len = len;
960: *mp = m;
961: top->m_pkthdr.len += len;
1.160 ad 962: s = splsoftnet();
963: solock(so);
1.144 dyoung 964: if (error != 0)
1.45 tv 965: goto release;
966: mp = &m->m_next;
967: if (resid <= 0) {
968: if (flags & MSG_EOR)
969: top->m_flags |= M_EOR;
970: break;
971: }
972: } while (space > 0 && atomic);
1.108 perry 973:
1.160 ad 974: if (so->so_state & SS_CANTSENDMORE) {
975: error = EPIPE;
976: goto release;
977: }
1.45 tv 978: if (dontroute)
979: so->so_options |= SO_DONTROUTE;
980: if (resid > 0)
981: so->so_state |= SS_MORETOCOME;
1.46 sommerfe 982: error = (*so->so_proto->pr_usrreq)(so,
983: (flags & MSG_OOB) ? PRU_SENDOOB : PRU_SEND,
1.160 ad 984: top, addr, control, curlwp);
1.45 tv 985: if (dontroute)
986: so->so_options &= ~SO_DONTROUTE;
987: if (resid > 0)
988: so->so_state &= ~SS_MORETOCOME;
989: clen = 0;
1.144 dyoung 990: control = NULL;
991: top = NULL;
1.45 tv 992: mp = ⊤
1.144 dyoung 993: if (error != 0)
1.1 cgd 994: goto release;
995: } while (resid && space > 0);
996: } while (resid);
997:
1.54 lukem 998: release:
1.1 cgd 999: sbunlock(&so->so_snd);
1.54 lukem 1000: out:
1.160 ad 1001: sounlock(so);
1002: splx(s);
1.1 cgd 1003: if (top)
1004: m_freem(top);
1005: if (control)
1006: m_freem(control);
1007: return (error);
1008: }
1009:
1010: /*
1.159 ad 1011: * Following replacement or removal of the first mbuf on the first
1012: * mbuf chain of a socket buffer, push necessary state changes back
1013: * into the socket buffer so that other consumers see the values
1014: * consistently. 'nextrecord' is the callers locally stored value of
1015: * the original value of sb->sb_mb->m_nextpkt which must be restored
1016: * when the lead mbuf changes. NOTE: 'nextrecord' may be NULL.
1017: */
1018: static void
1019: sbsync(struct sockbuf *sb, struct mbuf *nextrecord)
1020: {
1021:
1.160 ad 1022: KASSERT(solocked(sb->sb_so));
1023:
1.159 ad 1024: /*
1025: * First, update for the new value of nextrecord. If necessary,
1026: * make it the first record.
1027: */
1028: if (sb->sb_mb != NULL)
1029: sb->sb_mb->m_nextpkt = nextrecord;
1030: else
1031: sb->sb_mb = nextrecord;
1032:
1033: /*
1034: * Now update any dependent socket buffer fields to reflect
1035: * the new state. This is an inline of SB_EMPTY_FIXUP, with
1036: * the addition of a second clause that takes care of the
1037: * case where sb_mb has been updated, but remains the last
1038: * record.
1039: */
1040: if (sb->sb_mb == NULL) {
1041: sb->sb_mbtail = NULL;
1042: sb->sb_lastrecord = NULL;
1043: } else if (sb->sb_mb->m_nextpkt == NULL)
1044: sb->sb_lastrecord = sb->sb_mb;
1045: }
1046:
1047: /*
1.1 cgd 1048: * Implement receive operations on a socket.
1049: * We depend on the way that records are added to the sockbuf
1050: * by sbappend*. In particular, each record (mbufs linked through m_next)
1051: * must begin with an address if the protocol so specifies,
1052: * followed by an optional mbuf or mbufs containing ancillary data,
1053: * and then zero or more mbufs of data.
1054: * In order to avoid blocking network interrupts for the entire time here,
1055: * we splx() while doing the actual copy to user space.
1056: * Although the sockbuf is locked, new data may still be appended,
1057: * and thus we must maintain consistency of the sockbuf during that time.
1058: *
1059: * The caller may receive the data as a single mbuf chain by supplying
1060: * an mbuf **mp0 for use in returning the chain. The uio is then used
1061: * only for the count in uio_resid.
1062: */
1.3 andrew 1063: int
1.54 lukem 1064: soreceive(struct socket *so, struct mbuf **paddr, struct uio *uio,
1065: struct mbuf **mp0, struct mbuf **controlp, int *flagsp)
1.1 cgd 1066: {
1.116 yamt 1067: struct lwp *l = curlwp;
1.160 ad 1068: struct mbuf *m, **mp, *mt;
1.146 dyoung 1069: int atomic, flags, len, error, s, offset, moff, type, orig_resid;
1.99 matt 1070: const struct protosw *pr;
1.54 lukem 1071: struct mbuf *nextrecord;
1.67 he 1072: int mbuf_removed = 0;
1.146 dyoung 1073: const struct domain *dom;
1.64 thorpej 1074:
1.54 lukem 1075: pr = so->so_proto;
1.146 dyoung 1076: atomic = pr->pr_flags & PR_ATOMIC;
1077: dom = pr->pr_domain;
1.1 cgd 1078: mp = mp0;
1.54 lukem 1079: type = 0;
1080: orig_resid = uio->uio_resid;
1.102 jonathan 1081:
1.144 dyoung 1082: if (paddr != NULL)
1083: *paddr = NULL;
1084: if (controlp != NULL)
1085: *controlp = NULL;
1086: if (flagsp != NULL)
1.1 cgd 1087: flags = *flagsp &~ MSG_EOR;
1088: else
1089: flags = 0;
1.66 enami 1090:
1091: if ((flags & MSG_DONTWAIT) == 0)
1.117 yamt 1092: sodopendfree();
1.66 enami 1093:
1.1 cgd 1094: if (flags & MSG_OOB) {
1095: m = m_get(M_WAIT, MT_DATA);
1.160 ad 1096: solock(so);
1.17 cgd 1097: error = (*pr->pr_usrreq)(so, PRU_RCVOOB, m,
1.140 dyoung 1098: (struct mbuf *)(long)(flags & MSG_PEEK), NULL, l);
1.160 ad 1099: sounlock(so);
1.1 cgd 1100: if (error)
1101: goto bad;
1102: do {
1.134 christos 1103: error = uiomove(mtod(m, void *),
1.1 cgd 1104: (int) min(uio->uio_resid, m->m_len), uio);
1105: m = m_free(m);
1.144 dyoung 1106: } while (uio->uio_resid > 0 && error == 0 && m);
1.54 lukem 1107: bad:
1.144 dyoung 1108: if (m != NULL)
1.1 cgd 1109: m_freem(m);
1.144 dyoung 1110: return error;
1.1 cgd 1111: }
1.144 dyoung 1112: if (mp != NULL)
1.140 dyoung 1113: *mp = NULL;
1.160 ad 1114:
1115: /*
1116: * solock() provides atomicity of access. splsoftnet() prevents
1117: * protocol processing soft interrupts from interrupting us and
1118: * blocking (expensive).
1119: */
1120: s = splsoftnet();
1121: solock(so);
1.1 cgd 1122: if (so->so_state & SS_ISCONFIRMING && uio->uio_resid)
1.140 dyoung 1123: (*pr->pr_usrreq)(so, PRU_RCVD, NULL, NULL, NULL, l);
1.1 cgd 1124:
1.54 lukem 1125: restart:
1.160 ad 1126: if ((error = sblock(&so->so_rcv, SBLOCKWAIT(flags))) != 0) {
1127: sounlock(so);
1128: splx(s);
1.144 dyoung 1129: return error;
1.160 ad 1130: }
1.1 cgd 1131:
1132: m = so->so_rcv.sb_mb;
1133: /*
1134: * If we have less data than requested, block awaiting more
1135: * (subject to any timeout) if:
1.15 mycroft 1136: * 1. the current count is less than the low water mark,
1.1 cgd 1137: * 2. MSG_WAITALL is set, and it is possible to do the entire
1.15 mycroft 1138: * receive operation at once if we block (resid <= hiwat), or
1139: * 3. MSG_DONTWAIT is not set.
1.1 cgd 1140: * If MSG_WAITALL is set but resid is larger than the receive buffer,
1141: * we have to do the receive in sections, and thus risk returning
1142: * a short count if a timeout or signal occurs after we start.
1143: */
1.144 dyoung 1144: if (m == NULL ||
1145: ((flags & MSG_DONTWAIT) == 0 &&
1146: so->so_rcv.sb_cc < uio->uio_resid &&
1147: (so->so_rcv.sb_cc < so->so_rcv.sb_lowat ||
1148: ((flags & MSG_WAITALL) &&
1149: uio->uio_resid <= so->so_rcv.sb_hiwat)) &&
1.146 dyoung 1150: m->m_nextpkt == NULL && !atomic)) {
1.1 cgd 1151: #ifdef DIAGNOSTIC
1.144 dyoung 1152: if (m == NULL && so->so_rcv.sb_cc)
1.1 cgd 1153: panic("receive 1");
1154: #endif
1155: if (so->so_error) {
1.144 dyoung 1156: if (m != NULL)
1.15 mycroft 1157: goto dontblock;
1.1 cgd 1158: error = so->so_error;
1159: if ((flags & MSG_PEEK) == 0)
1160: so->so_error = 0;
1161: goto release;
1162: }
1163: if (so->so_state & SS_CANTRCVMORE) {
1.144 dyoung 1164: if (m != NULL)
1.15 mycroft 1165: goto dontblock;
1.1 cgd 1166: else
1167: goto release;
1168: }
1.144 dyoung 1169: for (; m != NULL; m = m->m_next)
1.1 cgd 1170: if (m->m_type == MT_OOBDATA || (m->m_flags & M_EOR)) {
1171: m = so->so_rcv.sb_mb;
1172: goto dontblock;
1173: }
1174: if ((so->so_state & (SS_ISCONNECTED|SS_ISCONNECTING)) == 0 &&
1175: (so->so_proto->pr_flags & PR_CONNREQUIRED)) {
1176: error = ENOTCONN;
1177: goto release;
1178: }
1179: if (uio->uio_resid == 0)
1180: goto release;
1.151 ad 1181: if (so->so_nbio || (flags & MSG_DONTWAIT)) {
1.1 cgd 1182: error = EWOULDBLOCK;
1183: goto release;
1184: }
1.69 thorpej 1185: SBLASTRECORDCHK(&so->so_rcv, "soreceive sbwait 1");
1186: SBLASTMBUFCHK(&so->so_rcv, "soreceive sbwait 1");
1.1 cgd 1187: sbunlock(&so->so_rcv);
1188: error = sbwait(&so->so_rcv);
1.160 ad 1189: if (error != 0) {
1190: sounlock(so);
1191: splx(s);
1.144 dyoung 1192: return error;
1.160 ad 1193: }
1.1 cgd 1194: goto restart;
1195: }
1.54 lukem 1196: dontblock:
1.69 thorpej 1197: /*
1198: * On entry here, m points to the first record of the socket buffer.
1.159 ad 1199: * From this point onward, we maintain 'nextrecord' as a cache of the
1200: * pointer to the next record in the socket buffer. We must keep the
1201: * various socket buffer pointers and local stack versions of the
1202: * pointers in sync, pushing out modifications before dropping the
1.160 ad 1203: * socket lock, and re-reading them when picking it up.
1.159 ad 1204: *
1205: * Otherwise, we will race with the network stack appending new data
1206: * or records onto the socket buffer by using inconsistent/stale
1207: * versions of the field, possibly resulting in socket buffer
1208: * corruption.
1209: *
1210: * By holding the high-level sblock(), we prevent simultaneous
1211: * readers from pulling off the front of the socket buffer.
1.69 thorpej 1212: */
1.144 dyoung 1213: if (l != NULL)
1.157 ad 1214: l->l_ru.ru_msgrcv++;
1.69 thorpej 1215: KASSERT(m == so->so_rcv.sb_mb);
1216: SBLASTRECORDCHK(&so->so_rcv, "soreceive 1");
1217: SBLASTMBUFCHK(&so->so_rcv, "soreceive 1");
1.1 cgd 1218: nextrecord = m->m_nextpkt;
1219: if (pr->pr_flags & PR_ADDR) {
1220: #ifdef DIAGNOSTIC
1221: if (m->m_type != MT_SONAME)
1222: panic("receive 1a");
1223: #endif
1.3 andrew 1224: orig_resid = 0;
1.1 cgd 1225: if (flags & MSG_PEEK) {
1226: if (paddr)
1227: *paddr = m_copy(m, 0, m->m_len);
1228: m = m->m_next;
1229: } else {
1230: sbfree(&so->so_rcv, m);
1.67 he 1231: mbuf_removed = 1;
1.144 dyoung 1232: if (paddr != NULL) {
1.1 cgd 1233: *paddr = m;
1234: so->so_rcv.sb_mb = m->m_next;
1.144 dyoung 1235: m->m_next = NULL;
1.1 cgd 1236: m = so->so_rcv.sb_mb;
1237: } else {
1238: MFREE(m, so->so_rcv.sb_mb);
1239: m = so->so_rcv.sb_mb;
1240: }
1.159 ad 1241: sbsync(&so->so_rcv, nextrecord);
1.1 cgd 1242: }
1243: }
1.159 ad 1244:
1245: /*
1246: * Process one or more MT_CONTROL mbufs present before any data mbufs
1247: * in the first mbuf chain on the socket buffer. If MSG_PEEK, we
1248: * just copy the data; if !MSG_PEEK, we call into the protocol to
1249: * perform externalization (or freeing if controlp == NULL).
1250: */
1251: if (__predict_false(m != NULL && m->m_type == MT_CONTROL)) {
1252: struct mbuf *cm = NULL, *cmn;
1253: struct mbuf **cme = &cm;
1254:
1255: do {
1256: if (flags & MSG_PEEK) {
1257: if (controlp != NULL) {
1258: *controlp = m_copy(m, 0, m->m_len);
1259: controlp = &(*controlp)->m_next;
1260: }
1261: m = m->m_next;
1262: } else {
1263: sbfree(&so->so_rcv, m);
1.1 cgd 1264: so->so_rcv.sb_mb = m->m_next;
1.144 dyoung 1265: m->m_next = NULL;
1.159 ad 1266: *cme = m;
1267: cme = &(*cme)->m_next;
1.1 cgd 1268: m = so->so_rcv.sb_mb;
1.159 ad 1269: }
1270: } while (m != NULL && m->m_type == MT_CONTROL);
1271: if ((flags & MSG_PEEK) == 0)
1272: sbsync(&so->so_rcv, nextrecord);
1273: for (; cm != NULL; cm = cmn) {
1274: cmn = cm->m_next;
1275: cm->m_next = NULL;
1276: type = mtod(cm, struct cmsghdr *)->cmsg_type;
1277: if (controlp != NULL) {
1278: if (dom->dom_externalize != NULL &&
1279: type == SCM_RIGHTS) {
1.160 ad 1280: sounlock(so);
1.159 ad 1281: splx(s);
1282: error = (*dom->dom_externalize)(cm, l);
1283: s = splsoftnet();
1.160 ad 1284: solock(so);
1.159 ad 1285: }
1286: *controlp = cm;
1287: while (*controlp != NULL)
1288: controlp = &(*controlp)->m_next;
1.1 cgd 1289: } else {
1.106 itojun 1290: /*
1291: * Dispose of any SCM_RIGHTS message that went
1292: * through the read path rather than recv.
1293: */
1.159 ad 1294: if (dom->dom_dispose != NULL &&
1295: type == SCM_RIGHTS) {
1.160 ad 1296: sounlock(so);
1.159 ad 1297: (*dom->dom_dispose)(cm);
1.160 ad 1298: solock(so);
1.159 ad 1299: }
1300: m_freem(cm);
1.1 cgd 1301: }
1302: }
1.159 ad 1303: if (m != NULL)
1304: nextrecord = so->so_rcv.sb_mb->m_nextpkt;
1305: else
1306: nextrecord = so->so_rcv.sb_mb;
1307: orig_resid = 0;
1.1 cgd 1308: }
1.69 thorpej 1309:
1.159 ad 1310: /* If m is non-NULL, we have some data to read. */
1311: if (__predict_true(m != NULL)) {
1.1 cgd 1312: type = m->m_type;
1313: if (type == MT_OOBDATA)
1314: flags |= MSG_OOB;
1315: }
1.69 thorpej 1316: SBLASTRECORDCHK(&so->so_rcv, "soreceive 2");
1317: SBLASTMBUFCHK(&so->so_rcv, "soreceive 2");
1318:
1.1 cgd 1319: moff = 0;
1320: offset = 0;
1.144 dyoung 1321: while (m != NULL && uio->uio_resid > 0 && error == 0) {
1.1 cgd 1322: if (m->m_type == MT_OOBDATA) {
1323: if (type != MT_OOBDATA)
1324: break;
1325: } else if (type == MT_OOBDATA)
1326: break;
1327: #ifdef DIAGNOSTIC
1328: else if (m->m_type != MT_DATA && m->m_type != MT_HEADER)
1329: panic("receive 3");
1330: #endif
1331: so->so_state &= ~SS_RCVATMARK;
1332: len = uio->uio_resid;
1333: if (so->so_oobmark && len > so->so_oobmark - offset)
1334: len = so->so_oobmark - offset;
1335: if (len > m->m_len - moff)
1336: len = m->m_len - moff;
1337: /*
1338: * If mp is set, just pass back the mbufs.
1339: * Otherwise copy them out via the uio, then free.
1340: * Sockbuf must be consistent here (points to current mbuf,
1341: * it points to next record) when we drop priority;
1342: * we must note any additions to the sockbuf when we
1343: * block interrupts again.
1344: */
1.144 dyoung 1345: if (mp == NULL) {
1.69 thorpej 1346: SBLASTRECORDCHK(&so->so_rcv, "soreceive uiomove");
1347: SBLASTMBUFCHK(&so->so_rcv, "soreceive uiomove");
1.160 ad 1348: sounlock(so);
1.1 cgd 1349: splx(s);
1.134 christos 1350: error = uiomove(mtod(m, char *) + moff, (int)len, uio);
1.20 mycroft 1351: s = splsoftnet();
1.160 ad 1352: solock(so);
1.144 dyoung 1353: if (error != 0) {
1.67 he 1354: /*
1355: * If any part of the record has been removed
1356: * (such as the MT_SONAME mbuf, which will
1357: * happen when PR_ADDR, and thus also
1358: * PR_ATOMIC, is set), then drop the entire
1359: * record to maintain the atomicity of the
1360: * receive operation.
1361: *
1362: * This avoids a later panic("receive 1a")
1363: * when compiled with DIAGNOSTIC.
1364: */
1.146 dyoung 1365: if (m && mbuf_removed && atomic)
1.67 he 1366: (void) sbdroprecord(&so->so_rcv);
1367:
1.57 jdolecek 1368: goto release;
1.67 he 1369: }
1.1 cgd 1370: } else
1371: uio->uio_resid -= len;
1372: if (len == m->m_len - moff) {
1373: if (m->m_flags & M_EOR)
1374: flags |= MSG_EOR;
1375: if (flags & MSG_PEEK) {
1376: m = m->m_next;
1377: moff = 0;
1378: } else {
1379: nextrecord = m->m_nextpkt;
1380: sbfree(&so->so_rcv, m);
1381: if (mp) {
1382: *mp = m;
1383: mp = &m->m_next;
1384: so->so_rcv.sb_mb = m = m->m_next;
1.140 dyoung 1385: *mp = NULL;
1.1 cgd 1386: } else {
1387: MFREE(m, so->so_rcv.sb_mb);
1388: m = so->so_rcv.sb_mb;
1389: }
1.69 thorpej 1390: /*
1391: * If m != NULL, we also know that
1392: * so->so_rcv.sb_mb != NULL.
1393: */
1394: KASSERT(so->so_rcv.sb_mb == m);
1395: if (m) {
1.1 cgd 1396: m->m_nextpkt = nextrecord;
1.69 thorpej 1397: if (nextrecord == NULL)
1398: so->so_rcv.sb_lastrecord = m;
1399: } else {
1400: so->so_rcv.sb_mb = nextrecord;
1.70 thorpej 1401: SB_EMPTY_FIXUP(&so->so_rcv);
1.69 thorpej 1402: }
1403: SBLASTRECORDCHK(&so->so_rcv, "soreceive 3");
1404: SBLASTMBUFCHK(&so->so_rcv, "soreceive 3");
1.1 cgd 1405: }
1.144 dyoung 1406: } else if (flags & MSG_PEEK)
1407: moff += len;
1408: else {
1.160 ad 1409: if (mp != NULL) {
1410: mt = m_copym(m, 0, len, M_NOWAIT);
1411: if (__predict_false(mt == NULL)) {
1412: sounlock(so);
1413: mt = m_copym(m, 0, len, M_WAIT);
1414: solock(so);
1415: }
1416: *mp = mt;
1417: }
1.144 dyoung 1418: m->m_data += len;
1419: m->m_len -= len;
1420: so->so_rcv.sb_cc -= len;
1.1 cgd 1421: }
1422: if (so->so_oobmark) {
1423: if ((flags & MSG_PEEK) == 0) {
1424: so->so_oobmark -= len;
1425: if (so->so_oobmark == 0) {
1426: so->so_state |= SS_RCVATMARK;
1427: break;
1428: }
1.7 cgd 1429: } else {
1.1 cgd 1430: offset += len;
1.7 cgd 1431: if (offset == so->so_oobmark)
1432: break;
1433: }
1.1 cgd 1434: }
1435: if (flags & MSG_EOR)
1436: break;
1437: /*
1438: * If the MSG_WAITALL flag is set (for non-atomic socket),
1439: * we must not quit until "uio->uio_resid == 0" or an error
1440: * termination. If a signal/timeout occurs, return
1441: * with a short count but without error.
1442: * Keep sockbuf locked against other readers.
1443: */
1.144 dyoung 1444: while (flags & MSG_WAITALL && m == NULL && uio->uio_resid > 0 &&
1.3 andrew 1445: !sosendallatonce(so) && !nextrecord) {
1.1 cgd 1446: if (so->so_error || so->so_state & SS_CANTRCVMORE)
1447: break;
1.68 matt 1448: /*
1449: * If we are peeking and the socket receive buffer is
1450: * full, stop since we can't get more data to peek at.
1451: */
1452: if ((flags & MSG_PEEK) && sbspace(&so->so_rcv) <= 0)
1453: break;
1454: /*
1455: * If we've drained the socket buffer, tell the
1456: * protocol in case it needs to do something to
1457: * get it filled again.
1458: */
1459: if ((pr->pr_flags & PR_WANTRCVD) && so->so_pcb)
1460: (*pr->pr_usrreq)(so, PRU_RCVD,
1.140 dyoung 1461: NULL, (struct mbuf *)(long)flags, NULL, l);
1.69 thorpej 1462: SBLASTRECORDCHK(&so->so_rcv, "soreceive sbwait 2");
1463: SBLASTMBUFCHK(&so->so_rcv, "soreceive sbwait 2");
1.1 cgd 1464: error = sbwait(&so->so_rcv);
1.144 dyoung 1465: if (error != 0) {
1.1 cgd 1466: sbunlock(&so->so_rcv);
1.160 ad 1467: sounlock(so);
1.1 cgd 1468: splx(s);
1.144 dyoung 1469: return 0;
1.1 cgd 1470: }
1.21 christos 1471: if ((m = so->so_rcv.sb_mb) != NULL)
1.1 cgd 1472: nextrecord = m->m_nextpkt;
1473: }
1474: }
1.3 andrew 1475:
1.146 dyoung 1476: if (m && atomic) {
1.3 andrew 1477: flags |= MSG_TRUNC;
1478: if ((flags & MSG_PEEK) == 0)
1479: (void) sbdroprecord(&so->so_rcv);
1480: }
1.1 cgd 1481: if ((flags & MSG_PEEK) == 0) {
1.144 dyoung 1482: if (m == NULL) {
1.69 thorpej 1483: /*
1.70 thorpej 1484: * First part is an inline SB_EMPTY_FIXUP(). Second
1.69 thorpej 1485: * part makes sure sb_lastrecord is up-to-date if
1486: * there is still data in the socket buffer.
1487: */
1.1 cgd 1488: so->so_rcv.sb_mb = nextrecord;
1.69 thorpej 1489: if (so->so_rcv.sb_mb == NULL) {
1490: so->so_rcv.sb_mbtail = NULL;
1491: so->so_rcv.sb_lastrecord = NULL;
1492: } else if (nextrecord->m_nextpkt == NULL)
1493: so->so_rcv.sb_lastrecord = nextrecord;
1494: }
1495: SBLASTRECORDCHK(&so->so_rcv, "soreceive 4");
1496: SBLASTMBUFCHK(&so->so_rcv, "soreceive 4");
1.1 cgd 1497: if (pr->pr_flags & PR_WANTRCVD && so->so_pcb)
1.140 dyoung 1498: (*pr->pr_usrreq)(so, PRU_RCVD, NULL,
1499: (struct mbuf *)(long)flags, NULL, l);
1.1 cgd 1500: }
1.3 andrew 1501: if (orig_resid == uio->uio_resid && orig_resid &&
1502: (flags & MSG_EOR) == 0 && (so->so_state & SS_CANTRCVMORE) == 0) {
1503: sbunlock(&so->so_rcv);
1504: goto restart;
1505: }
1.108 perry 1506:
1.144 dyoung 1507: if (flagsp != NULL)
1.1 cgd 1508: *flagsp |= flags;
1.54 lukem 1509: release:
1.1 cgd 1510: sbunlock(&so->so_rcv);
1.160 ad 1511: sounlock(so);
1.1 cgd 1512: splx(s);
1.144 dyoung 1513: return error;
1.1 cgd 1514: }
1515:
1.14 mycroft 1516: int
1.54 lukem 1517: soshutdown(struct socket *so, int how)
1.1 cgd 1518: {
1.99 matt 1519: const struct protosw *pr;
1.160 ad 1520: int error;
1521:
1522: KASSERT(solocked(so));
1.34 kleink 1523:
1.54 lukem 1524: pr = so->so_proto;
1.34 kleink 1525: if (!(how == SHUT_RD || how == SHUT_WR || how == SHUT_RDWR))
1526: return (EINVAL);
1.1 cgd 1527:
1.160 ad 1528: if (how == SHUT_RD || how == SHUT_RDWR) {
1.1 cgd 1529: sorflush(so);
1.160 ad 1530: error = 0;
1531: }
1.34 kleink 1532: if (how == SHUT_WR || how == SHUT_RDWR)
1.160 ad 1533: error = (*pr->pr_usrreq)(so, PRU_SHUTDOWN, NULL,
1.140 dyoung 1534: NULL, NULL, NULL);
1.160 ad 1535:
1536: return error;
1.1 cgd 1537: }
1538:
1.177.4.2 snj 1539: int
1540: sodrain(struct socket *so)
1541: {
1542: int error;
1543:
1544: solock(so);
1545: so->so_state |= SS_ISDRAINING;
1546: cv_broadcast(&so->so_cv);
1547: error = soshutdown(so, SHUT_RDWR);
1548: sounlock(so);
1549:
1550: return error;
1551: }
1552:
1.14 mycroft 1553: void
1.54 lukem 1554: sorflush(struct socket *so)
1.1 cgd 1555: {
1.54 lukem 1556: struct sockbuf *sb, asb;
1.99 matt 1557: const struct protosw *pr;
1.160 ad 1558:
1559: KASSERT(solocked(so));
1.1 cgd 1560:
1.54 lukem 1561: sb = &so->so_rcv;
1562: pr = so->so_proto;
1.160 ad 1563: socantrcvmore(so);
1.1 cgd 1564: sb->sb_flags |= SB_NOINTR;
1.160 ad 1565: (void )sblock(sb, M_WAITOK);
1.1 cgd 1566: sbunlock(sb);
1567: asb = *sb;
1.86 wrstuden 1568: /*
1569: * Clear most of the sockbuf structure, but leave some of the
1570: * fields valid.
1571: */
1572: memset(&sb->sb_startzero, 0,
1573: sizeof(*sb) - offsetof(struct sockbuf, sb_startzero));
1.160 ad 1574: if (pr->pr_flags & PR_RIGHTS && pr->pr_domain->dom_dispose) {
1575: sounlock(so);
1.1 cgd 1576: (*pr->pr_domain->dom_dispose)(asb.sb_mb);
1.160 ad 1577: solock(so);
1578: }
1.98 christos 1579: sbrelease(&asb, so);
1.1 cgd 1580: }
1581:
1.171 plunky 1582: /*
1583: * internal set SOL_SOCKET options
1584: */
1.142 dyoung 1585: static int
1.171 plunky 1586: sosetopt1(struct socket *so, const struct sockopt *sopt)
1.1 cgd 1587: {
1.171 plunky 1588: int error, optval;
1589: struct linger l;
1590: struct timeval tv;
1.142 dyoung 1591:
1.171 plunky 1592: switch (sopt->sopt_name) {
1.142 dyoung 1593:
1.170 tls 1594: case SO_ACCEPTFILTER:
1.177 ad 1595: error = accept_filt_setopt(so, sopt);
1596: KASSERT(solocked(so));
1.170 tls 1597: break;
1598:
1.171 plunky 1599: case SO_LINGER:
1600: error = sockopt_get(sopt, &l, sizeof(l));
1.177 ad 1601: solock(so);
1.171 plunky 1602: if (error)
1.177 ad 1603: break;
1.171 plunky 1604: if (l.l_linger < 0 || l.l_linger > USHRT_MAX ||
1.177 ad 1605: l.l_linger > (INT_MAX / hz)) {
1606: error = EDOM;
1607: break;
1608: }
1.171 plunky 1609: so->so_linger = l.l_linger;
1610: if (l.l_onoff)
1611: so->so_options |= SO_LINGER;
1612: else
1613: so->so_options &= ~SO_LINGER;
1.177 ad 1614: break;
1.1 cgd 1615:
1.142 dyoung 1616: case SO_DEBUG:
1617: case SO_KEEPALIVE:
1618: case SO_DONTROUTE:
1619: case SO_USELOOPBACK:
1620: case SO_BROADCAST:
1621: case SO_REUSEADDR:
1622: case SO_REUSEPORT:
1623: case SO_OOBINLINE:
1624: case SO_TIMESTAMP:
1.171 plunky 1625: error = sockopt_getint(sopt, &optval);
1.177 ad 1626: solock(so);
1.171 plunky 1627: if (error)
1.177 ad 1628: break;
1.171 plunky 1629: if (optval)
1630: so->so_options |= sopt->sopt_name;
1.142 dyoung 1631: else
1.171 plunky 1632: so->so_options &= ~sopt->sopt_name;
1.142 dyoung 1633: break;
1634:
1635: case SO_SNDBUF:
1636: case SO_RCVBUF:
1637: case SO_SNDLOWAT:
1638: case SO_RCVLOWAT:
1.171 plunky 1639: error = sockopt_getint(sopt, &optval);
1.177 ad 1640: solock(so);
1.171 plunky 1641: if (error)
1.177 ad 1642: break;
1.1 cgd 1643:
1.142 dyoung 1644: /*
1645: * Values < 1 make no sense for any of these
1646: * options, so disallow them.
1647: */
1.177 ad 1648: if (optval < 1) {
1649: error = EINVAL;
1650: break;
1651: }
1.1 cgd 1652:
1.171 plunky 1653: switch (sopt->sopt_name) {
1654: case SO_SNDBUF:
1.177 ad 1655: if (sbreserve(&so->so_snd, (u_long)optval, so) == 0) {
1656: error = ENOBUFS;
1657: break;
1658: }
1.171 plunky 1659: so->so_snd.sb_flags &= ~SB_AUTOSIZE;
1660: break;
1.1 cgd 1661:
1662: case SO_RCVBUF:
1.177 ad 1663: if (sbreserve(&so->so_rcv, (u_long)optval, so) == 0) {
1664: error = ENOBUFS;
1665: break;
1666: }
1.171 plunky 1667: so->so_rcv.sb_flags &= ~SB_AUTOSIZE;
1.142 dyoung 1668: break;
1669:
1670: /*
1671: * Make sure the low-water is never greater than
1672: * the high-water.
1673: */
1.1 cgd 1674: case SO_SNDLOWAT:
1.171 plunky 1675: if (optval > so->so_snd.sb_hiwat)
1676: optval = so->so_snd.sb_hiwat;
1677:
1678: so->so_snd.sb_lowat = optval;
1.142 dyoung 1679: break;
1.171 plunky 1680:
1.1 cgd 1681: case SO_RCVLOWAT:
1.171 plunky 1682: if (optval > so->so_rcv.sb_hiwat)
1683: optval = so->so_rcv.sb_hiwat;
1684:
1685: so->so_rcv.sb_lowat = optval;
1.142 dyoung 1686: break;
1687: }
1688: break;
1.28 thorpej 1689:
1.142 dyoung 1690: case SO_SNDTIMEO:
1691: case SO_RCVTIMEO:
1.171 plunky 1692: error = sockopt_get(sopt, &tv, sizeof(tv));
1.177 ad 1693: solock(so);
1.171 plunky 1694: if (error)
1.177 ad 1695: break;
1.171 plunky 1696:
1.177 ad 1697: if (tv.tv_sec > (INT_MAX - tv.tv_usec / tick) / hz) {
1698: error = EDOM;
1699: break;
1700: }
1.28 thorpej 1701:
1.171 plunky 1702: optval = tv.tv_sec * hz + tv.tv_usec / tick;
1703: if (optval == 0 && tv.tv_usec != 0)
1704: optval = 1;
1.28 thorpej 1705:
1.171 plunky 1706: switch (sopt->sopt_name) {
1.142 dyoung 1707: case SO_SNDTIMEO:
1.171 plunky 1708: so->so_snd.sb_timeo = optval;
1.1 cgd 1709: break;
1710: case SO_RCVTIMEO:
1.171 plunky 1711: so->so_rcv.sb_timeo = optval;
1.142 dyoung 1712: break;
1713: }
1714: break;
1.1 cgd 1715:
1.142 dyoung 1716: default:
1.177 ad 1717: solock(so);
1718: error = ENOPROTOOPT;
1719: break;
1.142 dyoung 1720: }
1.177 ad 1721: KASSERT(solocked(so));
1722: return error;
1.142 dyoung 1723: }
1.1 cgd 1724:
1.142 dyoung 1725: int
1.171 plunky 1726: sosetopt(struct socket *so, struct sockopt *sopt)
1.142 dyoung 1727: {
1728: int error, prerr;
1.1 cgd 1729:
1.177 ad 1730: if (sopt->sopt_level == SOL_SOCKET) {
1.171 plunky 1731: error = sosetopt1(so, sopt);
1.177 ad 1732: KASSERT(solocked(so));
1733: } else {
1.142 dyoung 1734: error = ENOPROTOOPT;
1.177 ad 1735: solock(so);
1736: }
1.1 cgd 1737:
1.142 dyoung 1738: if ((error == 0 || error == ENOPROTOOPT) &&
1739: so->so_proto != NULL && so->so_proto->pr_ctloutput != NULL) {
1740: /* give the protocol stack a shot */
1.171 plunky 1741: prerr = (*so->so_proto->pr_ctloutput)(PRCO_SETOPT, so, sopt);
1.142 dyoung 1742: if (prerr == 0)
1743: error = 0;
1744: else if (prerr != ENOPROTOOPT)
1745: error = prerr;
1.171 plunky 1746: }
1.160 ad 1747: sounlock(so);
1.142 dyoung 1748: return error;
1.1 cgd 1749: }
1750:
1.171 plunky 1751: /*
1752: * so_setsockopt() is a wrapper providing a sockopt structure for sosetopt()
1753: */
1754: int
1755: so_setsockopt(struct lwp *l, struct socket *so, int level, int name,
1756: const void *val, size_t valsize)
1757: {
1758: struct sockopt sopt;
1759: int error;
1760:
1761: KASSERT(valsize == 0 || val != NULL);
1762:
1763: sockopt_init(&sopt, level, name, valsize);
1764: sockopt_set(&sopt, val, valsize);
1765:
1766: error = sosetopt(so, &sopt);
1767:
1768: sockopt_destroy(&sopt);
1769:
1770: return error;
1771: }
1772:
1773: /*
1774: * internal get SOL_SOCKET options
1775: */
1776: static int
1777: sogetopt1(struct socket *so, struct sockopt *sopt)
1778: {
1779: int error, optval;
1780: struct linger l;
1781: struct timeval tv;
1782:
1783: switch (sopt->sopt_name) {
1784:
1785: case SO_ACCEPTFILTER:
1.177 ad 1786: error = accept_filt_getopt(so, sopt);
1.171 plunky 1787: break;
1788:
1789: case SO_LINGER:
1790: l.l_onoff = (so->so_options & SO_LINGER) ? 1 : 0;
1791: l.l_linger = so->so_linger;
1792:
1793: error = sockopt_set(sopt, &l, sizeof(l));
1794: break;
1795:
1796: case SO_USELOOPBACK:
1797: case SO_DONTROUTE:
1798: case SO_DEBUG:
1799: case SO_KEEPALIVE:
1800: case SO_REUSEADDR:
1801: case SO_REUSEPORT:
1802: case SO_BROADCAST:
1803: case SO_OOBINLINE:
1804: case SO_TIMESTAMP:
1805: error = sockopt_setint(sopt,
1806: (so->so_options & sopt->sopt_name) ? 1 : 0);
1807: break;
1808:
1809: case SO_TYPE:
1810: error = sockopt_setint(sopt, so->so_type);
1811: break;
1812:
1813: case SO_ERROR:
1814: error = sockopt_setint(sopt, so->so_error);
1815: so->so_error = 0;
1816: break;
1817:
1818: case SO_SNDBUF:
1819: error = sockopt_setint(sopt, so->so_snd.sb_hiwat);
1820: break;
1821:
1822: case SO_RCVBUF:
1823: error = sockopt_setint(sopt, so->so_rcv.sb_hiwat);
1824: break;
1825:
1826: case SO_SNDLOWAT:
1827: error = sockopt_setint(sopt, so->so_snd.sb_lowat);
1828: break;
1829:
1830: case SO_RCVLOWAT:
1831: error = sockopt_setint(sopt, so->so_rcv.sb_lowat);
1832: break;
1833:
1834: case SO_SNDTIMEO:
1835: case SO_RCVTIMEO:
1836: optval = (sopt->sopt_name == SO_SNDTIMEO ?
1837: so->so_snd.sb_timeo : so->so_rcv.sb_timeo);
1838:
1839: tv.tv_sec = optval / hz;
1840: tv.tv_usec = (optval % hz) * tick;
1841:
1842: error = sockopt_set(sopt, &tv, sizeof(tv));
1843: break;
1844:
1845: case SO_OVERFLOWED:
1846: error = sockopt_setint(sopt, so->so_rcv.sb_overflowed);
1847: break;
1848:
1849: default:
1850: error = ENOPROTOOPT;
1851: break;
1852: }
1853:
1854: return (error);
1855: }
1856:
1.14 mycroft 1857: int
1.171 plunky 1858: sogetopt(struct socket *so, struct sockopt *sopt)
1.1 cgd 1859: {
1.160 ad 1860: int error;
1.1 cgd 1861:
1.160 ad 1862: solock(so);
1.171 plunky 1863: if (sopt->sopt_level != SOL_SOCKET) {
1.1 cgd 1864: if (so->so_proto && so->so_proto->pr_ctloutput) {
1.160 ad 1865: error = ((*so->so_proto->pr_ctloutput)
1.171 plunky 1866: (PRCO_GETOPT, so, sopt));
1.1 cgd 1867: } else
1.160 ad 1868: error = (ENOPROTOOPT);
1.1 cgd 1869: } else {
1.171 plunky 1870: error = sogetopt1(so, sopt);
1871: }
1872: sounlock(so);
1873: return (error);
1874: }
1875:
1876: /*
1877: * alloc sockopt data buffer buffer
1878: * - will be released at destroy
1879: */
1.176 plunky 1880: static int
1881: sockopt_alloc(struct sockopt *sopt, size_t len, km_flag_t kmflag)
1.171 plunky 1882: {
1883:
1884: KASSERT(sopt->sopt_size == 0);
1885:
1.176 plunky 1886: if (len > sizeof(sopt->sopt_buf)) {
1887: sopt->sopt_data = kmem_zalloc(len, kmflag);
1888: if (sopt->sopt_data == NULL)
1889: return ENOMEM;
1890: } else
1.171 plunky 1891: sopt->sopt_data = sopt->sopt_buf;
1892:
1893: sopt->sopt_size = len;
1.176 plunky 1894: return 0;
1.171 plunky 1895: }
1896:
1897: /*
1898: * initialise sockopt storage
1.176 plunky 1899: * - MAY sleep during allocation
1.171 plunky 1900: */
1901: void
1902: sockopt_init(struct sockopt *sopt, int level, int name, size_t size)
1903: {
1.1 cgd 1904:
1.171 plunky 1905: memset(sopt, 0, sizeof(*sopt));
1.1 cgd 1906:
1.171 plunky 1907: sopt->sopt_level = level;
1908: sopt->sopt_name = name;
1.176 plunky 1909: (void)sockopt_alloc(sopt, size, KM_SLEEP);
1.171 plunky 1910: }
1911:
1912: /*
1913: * destroy sockopt storage
1914: * - will release any held memory references
1915: */
1916: void
1917: sockopt_destroy(struct sockopt *sopt)
1918: {
1919:
1920: if (sopt->sopt_data != sopt->sopt_buf)
1.173 plunky 1921: kmem_free(sopt->sopt_data, sopt->sopt_size);
1.171 plunky 1922:
1923: memset(sopt, 0, sizeof(*sopt));
1924: }
1925:
1926: /*
1927: * set sockopt value
1928: * - value is copied into sockopt
1.176 plunky 1929: * - memory is allocated when necessary, will not sleep
1.171 plunky 1930: */
1931: int
1932: sockopt_set(struct sockopt *sopt, const void *buf, size_t len)
1933: {
1.176 plunky 1934: int error;
1.171 plunky 1935:
1.176 plunky 1936: if (sopt->sopt_size == 0) {
1937: error = sockopt_alloc(sopt, len, KM_NOSLEEP);
1938: if (error)
1939: return error;
1940: }
1.171 plunky 1941:
1942: KASSERT(sopt->sopt_size == len);
1943: memcpy(sopt->sopt_data, buf, len);
1944: return 0;
1945: }
1946:
1947: /*
1948: * common case of set sockopt integer value
1949: */
1950: int
1951: sockopt_setint(struct sockopt *sopt, int val)
1952: {
1953:
1954: return sockopt_set(sopt, &val, sizeof(int));
1955: }
1956:
1957: /*
1958: * get sockopt value
1959: * - correct size must be given
1960: */
1961: int
1962: sockopt_get(const struct sockopt *sopt, void *buf, size_t len)
1963: {
1.170 tls 1964:
1.171 plunky 1965: if (sopt->sopt_size != len)
1966: return EINVAL;
1.1 cgd 1967:
1.171 plunky 1968: memcpy(buf, sopt->sopt_data, len);
1969: return 0;
1970: }
1.1 cgd 1971:
1.171 plunky 1972: /*
1973: * common case of get sockopt integer value
1974: */
1975: int
1976: sockopt_getint(const struct sockopt *sopt, int *valp)
1977: {
1.1 cgd 1978:
1.171 plunky 1979: return sockopt_get(sopt, valp, sizeof(int));
1980: }
1.1 cgd 1981:
1.171 plunky 1982: /*
1983: * set sockopt value from mbuf
1984: * - ONLY for legacy code
1985: * - mbuf is released by sockopt
1.176 plunky 1986: * - will not sleep
1.171 plunky 1987: */
1988: int
1989: sockopt_setmbuf(struct sockopt *sopt, struct mbuf *m)
1990: {
1991: size_t len;
1.176 plunky 1992: int error;
1.1 cgd 1993:
1.171 plunky 1994: len = m_length(m);
1.1 cgd 1995:
1.176 plunky 1996: if (sopt->sopt_size == 0) {
1997: error = sockopt_alloc(sopt, len, KM_NOSLEEP);
1998: if (error)
1999: return error;
2000: }
1.1 cgd 2001:
1.171 plunky 2002: KASSERT(sopt->sopt_size == len);
2003: m_copydata(m, 0, len, sopt->sopt_data);
2004: m_freem(m);
1.1 cgd 2005:
1.171 plunky 2006: return 0;
2007: }
1.1 cgd 2008:
1.171 plunky 2009: /*
2010: * get sockopt value into mbuf
2011: * - ONLY for legacy code
2012: * - mbuf to be released by the caller
1.176 plunky 2013: * - will not sleep
1.171 plunky 2014: */
2015: struct mbuf *
2016: sockopt_getmbuf(const struct sockopt *sopt)
2017: {
2018: struct mbuf *m;
1.107 darrenr 2019:
1.176 plunky 2020: if (sopt->sopt_size > MCLBYTES)
2021: return NULL;
2022:
2023: m = m_get(M_DONTWAIT, MT_SOOPTS);
1.171 plunky 2024: if (m == NULL)
2025: return NULL;
2026:
1.176 plunky 2027: if (sopt->sopt_size > MLEN) {
2028: MCLGET(m, M_DONTWAIT);
2029: if ((m->m_flags & M_EXT) == 0) {
2030: m_free(m);
2031: return NULL;
2032: }
1.1 cgd 2033: }
1.176 plunky 2034:
2035: memcpy(mtod(m, void *), sopt->sopt_data, sopt->sopt_size);
2036: m->m_len = sopt->sopt_size;
1.160 ad 2037:
1.171 plunky 2038: return m;
1.1 cgd 2039: }
2040:
1.14 mycroft 2041: void
1.54 lukem 2042: sohasoutofband(struct socket *so)
1.1 cgd 2043: {
1.153 rmind 2044:
1.90 christos 2045: fownsignal(so->so_pgid, SIGURG, POLL_PRI, POLLPRI|POLLRDBAND, so);
1.177.4.2.2.1! (bouyer 2046:: selnotify(&so->so_rcv.sb_sel, POLLPRI | POLLRDBAND, NOTE_SUBMIT);
1.1 cgd 2047: }
1.72 jdolecek 2048:
2049: static void
2050: filt_sordetach(struct knote *kn)
2051: {
2052: struct socket *so;
2053:
1.155 ad 2054: so = ((file_t *)kn->kn_obj)->f_data;
1.160 ad 2055: solock(so);
1.73 christos 2056: SLIST_REMOVE(&so->so_rcv.sb_sel.sel_klist, kn, knote, kn_selnext);
2057: if (SLIST_EMPTY(&so->so_rcv.sb_sel.sel_klist))
1.72 jdolecek 2058: so->so_rcv.sb_flags &= ~SB_KNOTE;
1.160 ad 2059: sounlock(so);
1.72 jdolecek 2060: }
2061:
2062: /*ARGSUSED*/
2063: static int
1.129 yamt 2064: filt_soread(struct knote *kn, long hint)
1.72 jdolecek 2065: {
2066: struct socket *so;
1.160 ad 2067: int rv;
1.72 jdolecek 2068:
1.155 ad 2069: so = ((file_t *)kn->kn_obj)->f_data;
1.160 ad 2070: if (hint != NOTE_SUBMIT)
2071: solock(so);
1.72 jdolecek 2072: kn->kn_data = so->so_rcv.sb_cc;
2073: if (so->so_state & SS_CANTRCVMORE) {
1.108 perry 2074: kn->kn_flags |= EV_EOF;
1.72 jdolecek 2075: kn->kn_fflags = so->so_error;
1.160 ad 2076: rv = 1;
2077: } else if (so->so_error) /* temporary udp error */
2078: rv = 1;
2079: else if (kn->kn_sfflags & NOTE_LOWAT)
2080: rv = (kn->kn_data >= kn->kn_sdata);
2081: else
2082: rv = (kn->kn_data >= so->so_rcv.sb_lowat);
2083: if (hint != NOTE_SUBMIT)
2084: sounlock(so);
2085: return rv;
1.72 jdolecek 2086: }
2087:
2088: static void
2089: filt_sowdetach(struct knote *kn)
2090: {
2091: struct socket *so;
2092:
1.155 ad 2093: so = ((file_t *)kn->kn_obj)->f_data;
1.160 ad 2094: solock(so);
1.73 christos 2095: SLIST_REMOVE(&so->so_snd.sb_sel.sel_klist, kn, knote, kn_selnext);
2096: if (SLIST_EMPTY(&so->so_snd.sb_sel.sel_klist))
1.72 jdolecek 2097: so->so_snd.sb_flags &= ~SB_KNOTE;
1.160 ad 2098: sounlock(so);
1.72 jdolecek 2099: }
2100:
2101: /*ARGSUSED*/
2102: static int
1.129 yamt 2103: filt_sowrite(struct knote *kn, long hint)
1.72 jdolecek 2104: {
2105: struct socket *so;
1.160 ad 2106: int rv;
1.72 jdolecek 2107:
1.155 ad 2108: so = ((file_t *)kn->kn_obj)->f_data;
1.160 ad 2109: if (hint != NOTE_SUBMIT)
2110: solock(so);
1.72 jdolecek 2111: kn->kn_data = sbspace(&so->so_snd);
2112: if (so->so_state & SS_CANTSENDMORE) {
1.108 perry 2113: kn->kn_flags |= EV_EOF;
1.72 jdolecek 2114: kn->kn_fflags = so->so_error;
1.160 ad 2115: rv = 1;
2116: } else if (so->so_error) /* temporary udp error */
2117: rv = 1;
2118: else if (((so->so_state & SS_ISCONNECTED) == 0) &&
1.72 jdolecek 2119: (so->so_proto->pr_flags & PR_CONNREQUIRED))
1.160 ad 2120: rv = 0;
2121: else if (kn->kn_sfflags & NOTE_LOWAT)
2122: rv = (kn->kn_data >= kn->kn_sdata);
2123: else
2124: rv = (kn->kn_data >= so->so_snd.sb_lowat);
2125: if (hint != NOTE_SUBMIT)
2126: sounlock(so);
2127: return rv;
1.72 jdolecek 2128: }
2129:
2130: /*ARGSUSED*/
2131: static int
1.129 yamt 2132: filt_solisten(struct knote *kn, long hint)
1.72 jdolecek 2133: {
2134: struct socket *so;
1.160 ad 2135: int rv;
1.72 jdolecek 2136:
1.155 ad 2137: so = ((file_t *)kn->kn_obj)->f_data;
1.72 jdolecek 2138:
2139: /*
2140: * Set kn_data to number of incoming connections, not
2141: * counting partial (incomplete) connections.
1.108 perry 2142: */
1.160 ad 2143: if (hint != NOTE_SUBMIT)
2144: solock(so);
1.72 jdolecek 2145: kn->kn_data = so->so_qlen;
1.160 ad 2146: rv = (kn->kn_data > 0);
2147: if (hint != NOTE_SUBMIT)
2148: sounlock(so);
2149: return rv;
1.72 jdolecek 2150: }
2151:
2152: static const struct filterops solisten_filtops =
2153: { 1, NULL, filt_sordetach, filt_solisten };
2154: static const struct filterops soread_filtops =
2155: { 1, NULL, filt_sordetach, filt_soread };
2156: static const struct filterops sowrite_filtops =
2157: { 1, NULL, filt_sowdetach, filt_sowrite };
2158:
2159: int
1.129 yamt 2160: soo_kqfilter(struct file *fp, struct knote *kn)
1.72 jdolecek 2161: {
2162: struct socket *so;
2163: struct sockbuf *sb;
2164:
1.155 ad 2165: so = ((file_t *)kn->kn_obj)->f_data;
1.160 ad 2166: solock(so);
1.72 jdolecek 2167: switch (kn->kn_filter) {
2168: case EVFILT_READ:
2169: if (so->so_options & SO_ACCEPTCONN)
2170: kn->kn_fop = &solisten_filtops;
2171: else
2172: kn->kn_fop = &soread_filtops;
2173: sb = &so->so_rcv;
2174: break;
2175: case EVFILT_WRITE:
2176: kn->kn_fop = &sowrite_filtops;
2177: sb = &so->so_snd;
2178: break;
2179: default:
1.160 ad 2180: sounlock(so);
1.149 pooka 2181: return (EINVAL);
1.72 jdolecek 2182: }
1.73 christos 2183: SLIST_INSERT_HEAD(&sb->sb_sel.sel_klist, kn, kn_selnext);
1.72 jdolecek 2184: sb->sb_flags |= SB_KNOTE;
1.160 ad 2185: sounlock(so);
1.72 jdolecek 2186: return (0);
2187: }
2188:
1.154 ad 2189: static int
2190: sodopoll(struct socket *so, int events)
2191: {
2192: int revents;
2193:
2194: revents = 0;
2195:
2196: if (events & (POLLIN | POLLRDNORM))
2197: if (soreadable(so))
2198: revents |= events & (POLLIN | POLLRDNORM);
2199:
2200: if (events & (POLLOUT | POLLWRNORM))
2201: if (sowritable(so))
2202: revents |= events & (POLLOUT | POLLWRNORM);
2203:
2204: if (events & (POLLPRI | POLLRDBAND))
2205: if (so->so_oobmark || (so->so_state & SS_RCVATMARK))
2206: revents |= events & (POLLPRI | POLLRDBAND);
2207:
2208: return revents;
2209: }
2210:
2211: int
2212: sopoll(struct socket *so, int events)
2213: {
2214: int revents = 0;
2215:
1.160 ad 2216: #ifndef DIAGNOSTIC
2217: /*
2218: * Do a quick, unlocked check in expectation that the socket
2219: * will be ready for I/O. Don't do this check if DIAGNOSTIC,
2220: * as the solocked() assertions will fail.
2221: */
1.154 ad 2222: if ((revents = sodopoll(so, events)) != 0)
2223: return revents;
1.160 ad 2224: #endif
1.154 ad 2225:
1.160 ad 2226: solock(so);
1.154 ad 2227: if ((revents = sodopoll(so, events)) == 0) {
2228: if (events & (POLLIN | POLLPRI | POLLRDNORM | POLLRDBAND)) {
2229: selrecord(curlwp, &so->so_rcv.sb_sel);
1.160 ad 2230: so->so_rcv.sb_flags |= SB_NOTIFY;
1.154 ad 2231: }
2232:
2233: if (events & (POLLOUT | POLLWRNORM)) {
2234: selrecord(curlwp, &so->so_snd.sb_sel);
1.160 ad 2235: so->so_snd.sb_flags |= SB_NOTIFY;
1.154 ad 2236: }
2237: }
1.160 ad 2238: sounlock(so);
1.154 ad 2239:
2240: return revents;
2241: }
2242:
2243:
1.94 yamt 2244: #include <sys/sysctl.h>
2245:
2246: static int sysctl_kern_somaxkva(SYSCTLFN_PROTO);
2247:
2248: /*
2249: * sysctl helper routine for kern.somaxkva. ensures that the given
2250: * value is not too small.
2251: * (XXX should we maybe make sure it's not too large as well?)
2252: */
2253: static int
2254: sysctl_kern_somaxkva(SYSCTLFN_ARGS)
2255: {
2256: int error, new_somaxkva;
2257: struct sysctlnode node;
2258:
2259: new_somaxkva = somaxkva;
2260: node = *rnode;
2261: node.sysctl_data = &new_somaxkva;
2262: error = sysctl_lookup(SYSCTLFN_CALL(&node));
2263: if (error || newp == NULL)
2264: return (error);
2265:
2266: if (new_somaxkva < (16 * 1024 * 1024)) /* sanity */
2267: return (EINVAL);
2268:
1.136 ad 2269: mutex_enter(&so_pendfree_lock);
1.94 yamt 2270: somaxkva = new_somaxkva;
1.136 ad 2271: cv_broadcast(&socurkva_cv);
2272: mutex_exit(&so_pendfree_lock);
1.94 yamt 2273:
2274: return (error);
2275: }
2276:
2277: SYSCTL_SETUP(sysctl_kern_somaxkva_setup, "sysctl kern.somaxkva setup")
2278: {
2279:
1.97 atatat 2280: sysctl_createv(clog, 0, NULL, NULL,
2281: CTLFLAG_PERMANENT,
2282: CTLTYPE_NODE, "kern", NULL,
2283: NULL, 0, NULL, 0,
2284: CTL_KERN, CTL_EOL);
2285:
2286: sysctl_createv(clog, 0, NULL, NULL,
2287: CTLFLAG_PERMANENT|CTLFLAG_READWRITE,
1.103 atatat 2288: CTLTYPE_INT, "somaxkva",
2289: SYSCTL_DESCR("Maximum amount of kernel memory to be "
2290: "used for socket buffers"),
1.94 yamt 2291: sysctl_kern_somaxkva, 0, NULL, 0,
2292: CTL_KERN, KERN_SOMAXKVA, CTL_EOL);
2293: }
CVSweb <webmaster@jp.NetBSD.org>