1 /* $NetBSD: svc.c,v 1.21 2000/07/06 03:10:35 christos Exp $ */ 2 3 /*- 4 * Copyright (c) 2009, Sun Microsystems, Inc. 5 * All rights reserved. 6 * 7 * Redistribution and use in source and binary forms, with or without 8 * modification, are permitted provided that the following conditions are met: 9 * - Redistributions of source code must retain the above copyright notice, 10 * this list of conditions and the following disclaimer. 11 * - Redistributions in binary form must reproduce the above copyright notice, 12 * this list of conditions and the following disclaimer in the documentation 13 * and/or other materials provided with the distribution. 14 * - Neither the name of Sun Microsystems, Inc. nor the names of its 15 * contributors may be used to endorse or promote products derived 16 * from this software without specific prior written permission. 17 * 18 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" 19 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 20 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 21 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE 22 * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 23 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 24 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 25 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 26 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 27 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 28 * POSSIBILITY OF SUCH DAMAGE. 29 */ 30 31 #if defined(LIBC_SCCS) && !defined(lint) 32 static char *sccsid2 = "@(#)svc.c 1.44 88/02/08 Copyr 1984 Sun Micro"; 33 static char *sccsid = "@(#)svc.c 2.4 88/08/11 4.0 RPCSRC"; 34 #endif 35 #include <sys/cdefs.h> 36 __FBSDID("$FreeBSD$"); 37 38 /* 39 * svc.c, Server-side remote procedure call interface. 40 * 41 * There are two sets of procedures here. The xprt routines are 42 * for handling transport handles. The svc routines handle the 43 * list of service routines. 44 * 45 * Copyright (C) 1984, Sun Microsystems, Inc. 46 */ 47 48 #include <sys/param.h> 49 #include <sys/lock.h> 50 #include <sys/kernel.h> 51 #include <sys/kthread.h> 52 #include <sys/malloc.h> 53 #include <sys/mbuf.h> 54 #include <sys/mutex.h> 55 #include <sys/proc.h> 56 #include <sys/queue.h> 57 #include <sys/socketvar.h> 58 #include <sys/systm.h> 59 #include <sys/ucred.h> 60 61 #include <rpc/rpc.h> 62 #include <rpc/rpcb_clnt.h> 63 #include <rpc/replay.h> 64 65 #include <rpc/rpc_com.h> 66 67 #define SVC_VERSQUIET 0x0001 /* keep quiet about vers mismatch */ 68 #define version_keepquiet(xp) (SVC_EXT(xp)->xp_flags & SVC_VERSQUIET) 69 70 static struct svc_callout *svc_find(SVCPOOL *pool, rpcprog_t, rpcvers_t, 71 char *); 72 static void svc_new_thread(SVCPOOL *pool); 73 static void xprt_unregister_locked(SVCXPRT *xprt); 74 75 /* *************** SVCXPRT related stuff **************** */ 76 77 static int svcpool_minthread_sysctl(SYSCTL_HANDLER_ARGS); 78 static int svcpool_maxthread_sysctl(SYSCTL_HANDLER_ARGS); 79 80 SVCPOOL* 81 svcpool_create(const char *name, struct sysctl_oid_list *sysctl_base) 82 { 83 SVCPOOL *pool; 84 85 pool = malloc(sizeof(SVCPOOL), M_RPC, M_WAITOK|M_ZERO); 86 87 mtx_init(&pool->sp_lock, "sp_lock", NULL, MTX_DEF); 88 pool->sp_name = name; 89 pool->sp_state = SVCPOOL_INIT; 90 pool->sp_proc = NULL; 91 TAILQ_INIT(&pool->sp_xlist); 92 TAILQ_INIT(&pool->sp_active); 93 TAILQ_INIT(&pool->sp_callouts); 94 LIST_INIT(&pool->sp_threads); 95 LIST_INIT(&pool->sp_idlethreads); 96 pool->sp_minthreads = 1; 97 pool->sp_maxthreads = 1; 98 pool->sp_threadcount = 0; 99 100 /* 101 * Don't use more than a quarter of mbuf clusters or more than 102 * 45Mb buffering requests. 103 */ 104 pool->sp_space_high = nmbclusters * MCLBYTES / 4; 105 if (pool->sp_space_high > 45 << 20) 106 pool->sp_space_high = 45 << 20; 107 pool->sp_space_low = 2 * pool->sp_space_high / 3; 108 109 sysctl_ctx_init(&pool->sp_sysctl); 110 if (sysctl_base) { 111 SYSCTL_ADD_PROC(&pool->sp_sysctl, sysctl_base, OID_AUTO, 112 "minthreads", CTLTYPE_INT | CTLFLAG_RW, 113 pool, 0, svcpool_minthread_sysctl, "I", ""); 114 SYSCTL_ADD_PROC(&pool->sp_sysctl, sysctl_base, OID_AUTO, 115 "maxthreads", CTLTYPE_INT | CTLFLAG_RW, 116 pool, 0, svcpool_maxthread_sysctl, "I", ""); 117 SYSCTL_ADD_INT(&pool->sp_sysctl, sysctl_base, OID_AUTO, 118 "threads", CTLFLAG_RD, &pool->sp_threadcount, 0, ""); 119 120 SYSCTL_ADD_UINT(&pool->sp_sysctl, sysctl_base, OID_AUTO, 121 "request_space_used", CTLFLAG_RD, 122 &pool->sp_space_used, 0, 123 "Space in parsed but not handled requests."); 124 125 SYSCTL_ADD_UINT(&pool->sp_sysctl, sysctl_base, OID_AUTO, 126 "request_space_used_highest", CTLFLAG_RD, 127 &pool->sp_space_used_highest, 0, 128 "Highest space used since reboot."); 129 130 SYSCTL_ADD_UINT(&pool->sp_sysctl, sysctl_base, OID_AUTO, 131 "request_space_high", CTLFLAG_RW, 132 &pool->sp_space_high, 0, 133 "Maximum space in parsed but not handled requests."); 134 135 SYSCTL_ADD_UINT(&pool->sp_sysctl, sysctl_base, OID_AUTO, 136 "request_space_low", CTLFLAG_RW, 137 &pool->sp_space_low, 0, 138 "Low water mark for request space."); 139 140 SYSCTL_ADD_INT(&pool->sp_sysctl, sysctl_base, OID_AUTO, 141 "request_space_throttled", CTLFLAG_RD, 142 &pool->sp_space_throttled, 0, 143 "Whether nfs requests are currently throttled"); 144 145 SYSCTL_ADD_INT(&pool->sp_sysctl, sysctl_base, OID_AUTO, 146 "request_space_throttle_count", CTLFLAG_RD, 147 &pool->sp_space_throttle_count, 0, 148 "Count of times throttling based on request space has occurred"); 149 } 150 151 return pool; 152 } 153 154 void 155 svcpool_destroy(SVCPOOL *pool) 156 { 157 SVCXPRT *xprt, *nxprt; 158 struct svc_callout *s; 159 struct svcxprt_list cleanup; 160 161 TAILQ_INIT(&cleanup); 162 mtx_lock(&pool->sp_lock); 163 164 while (TAILQ_FIRST(&pool->sp_xlist)) { 165 xprt = TAILQ_FIRST(&pool->sp_xlist); 166 xprt_unregister_locked(xprt); 167 TAILQ_INSERT_TAIL(&cleanup, xprt, xp_link); 168 } 169 170 while (TAILQ_FIRST(&pool->sp_callouts)) { 171 s = TAILQ_FIRST(&pool->sp_callouts); 172 mtx_unlock(&pool->sp_lock); 173 svc_unreg(pool, s->sc_prog, s->sc_vers); 174 mtx_lock(&pool->sp_lock); 175 } 176 mtx_unlock(&pool->sp_lock); 177 178 TAILQ_FOREACH_SAFE(xprt, &cleanup, xp_link, nxprt) { 179 SVC_RELEASE(xprt); 180 } 181 182 mtx_destroy(&pool->sp_lock); 183 184 if (pool->sp_rcache) 185 replay_freecache(pool->sp_rcache); 186 187 sysctl_ctx_free(&pool->sp_sysctl); 188 free(pool, M_RPC); 189 } 190 191 static bool_t 192 svcpool_active(SVCPOOL *pool) 193 { 194 enum svcpool_state state = pool->sp_state; 195 196 if (state == SVCPOOL_INIT || state == SVCPOOL_CLOSING) 197 return (FALSE); 198 return (TRUE); 199 } 200 201 /* 202 * Sysctl handler to set the minimum thread count on a pool 203 */ 204 static int 205 svcpool_minthread_sysctl(SYSCTL_HANDLER_ARGS) 206 { 207 SVCPOOL *pool; 208 int newminthreads, error, n; 209 210 pool = oidp->oid_arg1; 211 newminthreads = pool->sp_minthreads; 212 error = sysctl_handle_int(oidp, &newminthreads, 0, req); 213 if (error == 0 && newminthreads != pool->sp_minthreads) { 214 if (newminthreads > pool->sp_maxthreads) 215 return (EINVAL); 216 mtx_lock(&pool->sp_lock); 217 if (newminthreads > pool->sp_minthreads 218 && svcpool_active(pool)) { 219 /* 220 * If the pool is running and we are 221 * increasing, create some more threads now. 222 */ 223 n = newminthreads - pool->sp_threadcount; 224 if (n > 0) { 225 mtx_unlock(&pool->sp_lock); 226 while (n--) 227 svc_new_thread(pool); 228 mtx_lock(&pool->sp_lock); 229 } 230 } 231 pool->sp_minthreads = newminthreads; 232 mtx_unlock(&pool->sp_lock); 233 } 234 return (error); 235 } 236 237 /* 238 * Sysctl handler to set the maximum thread count on a pool 239 */ 240 static int 241 svcpool_maxthread_sysctl(SYSCTL_HANDLER_ARGS) 242 { 243 SVCPOOL *pool; 244 SVCTHREAD *st; 245 int newmaxthreads, error; 246 247 pool = oidp->oid_arg1; 248 newmaxthreads = pool->sp_maxthreads; 249 error = sysctl_handle_int(oidp, &newmaxthreads, 0, req); 250 if (error == 0 && newmaxthreads != pool->sp_maxthreads) { 251 if (newmaxthreads < pool->sp_minthreads) 252 return (EINVAL); 253 mtx_lock(&pool->sp_lock); 254 if (newmaxthreads < pool->sp_maxthreads 255 && svcpool_active(pool)) { 256 /* 257 * If the pool is running and we are 258 * decreasing, wake up some idle threads to 259 * encourage them to exit. 260 */ 261 LIST_FOREACH(st, &pool->sp_idlethreads, st_ilink) 262 cv_signal(&st->st_cond); 263 } 264 pool->sp_maxthreads = newmaxthreads; 265 mtx_unlock(&pool->sp_lock); 266 } 267 return (error); 268 } 269 270 /* 271 * Activate a transport handle. 272 */ 273 void 274 xprt_register(SVCXPRT *xprt) 275 { 276 SVCPOOL *pool = xprt->xp_pool; 277 278 SVC_ACQUIRE(xprt); 279 mtx_lock(&pool->sp_lock); 280 xprt->xp_registered = TRUE; 281 xprt->xp_active = FALSE; 282 TAILQ_INSERT_TAIL(&pool->sp_xlist, xprt, xp_link); 283 mtx_unlock(&pool->sp_lock); 284 } 285 286 /* 287 * De-activate a transport handle. Note: the locked version doesn't 288 * release the transport - caller must do that after dropping the pool 289 * lock. 290 */ 291 static void 292 xprt_unregister_locked(SVCXPRT *xprt) 293 { 294 SVCPOOL *pool = xprt->xp_pool; 295 296 KASSERT(xprt->xp_registered == TRUE, 297 ("xprt_unregister_locked: not registered")); 298 if (xprt->xp_active) { 299 TAILQ_REMOVE(&pool->sp_active, xprt, xp_alink); 300 xprt->xp_active = FALSE; 301 } 302 TAILQ_REMOVE(&pool->sp_xlist, xprt, xp_link); 303 xprt->xp_registered = FALSE; 304 } 305 306 void 307 xprt_unregister(SVCXPRT *xprt) 308 { 309 SVCPOOL *pool = xprt->xp_pool; 310 311 mtx_lock(&pool->sp_lock); 312 if (xprt->xp_registered == FALSE) { 313 /* Already unregistered by another thread */ 314 mtx_unlock(&pool->sp_lock); 315 return; 316 } 317 xprt_unregister_locked(xprt); 318 mtx_unlock(&pool->sp_lock); 319 320 SVC_RELEASE(xprt); 321 } 322 323 static void 324 xprt_assignthread(SVCXPRT *xprt) 325 { 326 SVCPOOL *pool = xprt->xp_pool; 327 SVCTHREAD *st; 328 329 /* 330 * Attempt to assign a service thread to this 331 * transport. 332 */ 333 LIST_FOREACH(st, &pool->sp_idlethreads, st_ilink) { 334 if (st->st_xprt == NULL && STAILQ_EMPTY(&st->st_reqs)) 335 break; 336 } 337 if (st) { 338 SVC_ACQUIRE(xprt); 339 xprt->xp_thread = st; 340 st->st_xprt = xprt; 341 cv_signal(&st->st_cond); 342 } else { 343 /* 344 * See if we can create a new thread. The 345 * actual thread creation happens in 346 * svc_run_internal because our locking state 347 * is poorly defined (we are typically called 348 * from a socket upcall). Don't create more 349 * than one thread per second. 350 */ 351 if (pool->sp_state == SVCPOOL_ACTIVE 352 && pool->sp_lastcreatetime < time_uptime 353 && pool->sp_threadcount < pool->sp_maxthreads) { 354 pool->sp_state = SVCPOOL_THREADWANTED; 355 } 356 } 357 } 358 359 void 360 xprt_active(SVCXPRT *xprt) 361 { 362 SVCPOOL *pool = xprt->xp_pool; 363 364 mtx_lock(&pool->sp_lock); 365 366 if (!xprt->xp_registered) { 367 /* 368 * Race with xprt_unregister - we lose. 369 */ 370 mtx_unlock(&pool->sp_lock); 371 return; 372 } 373 374 if (!xprt->xp_active) { 375 TAILQ_INSERT_TAIL(&pool->sp_active, xprt, xp_alink); 376 xprt->xp_active = TRUE; 377 xprt_assignthread(xprt); 378 } 379 380 mtx_unlock(&pool->sp_lock); 381 } 382 383 void 384 xprt_inactive_locked(SVCXPRT *xprt) 385 { 386 SVCPOOL *pool = xprt->xp_pool; 387 388 if (xprt->xp_active) { 389 TAILQ_REMOVE(&pool->sp_active, xprt, xp_alink); 390 xprt->xp_active = FALSE; 391 } 392 } 393 394 void 395 xprt_inactive(SVCXPRT *xprt) 396 { 397 SVCPOOL *pool = xprt->xp_pool; 398 399 mtx_lock(&pool->sp_lock); 400 xprt_inactive_locked(xprt); 401 mtx_unlock(&pool->sp_lock); 402 } 403 404 /* 405 * Add a service program to the callout list. 406 * The dispatch routine will be called when a rpc request for this 407 * program number comes in. 408 */ 409 bool_t 410 svc_reg(SVCXPRT *xprt, const rpcprog_t prog, const rpcvers_t vers, 411 void (*dispatch)(struct svc_req *, SVCXPRT *), 412 const struct netconfig *nconf) 413 { 414 SVCPOOL *pool = xprt->xp_pool; 415 struct svc_callout *s; 416 char *netid = NULL; 417 int flag = 0; 418 419 /* VARIABLES PROTECTED BY svc_lock: s, svc_head */ 420 421 if (xprt->xp_netid) { 422 netid = strdup(xprt->xp_netid, M_RPC); 423 flag = 1; 424 } else if (nconf && nconf->nc_netid) { 425 netid = strdup(nconf->nc_netid, M_RPC); 426 flag = 1; 427 } /* must have been created with svc_raw_create */ 428 if ((netid == NULL) && (flag == 1)) { 429 return (FALSE); 430 } 431 432 mtx_lock(&pool->sp_lock); 433 if ((s = svc_find(pool, prog, vers, netid)) != NULL) { 434 if (netid) 435 free(netid, M_RPC); 436 if (s->sc_dispatch == dispatch) 437 goto rpcb_it; /* he is registering another xptr */ 438 mtx_unlock(&pool->sp_lock); 439 return (FALSE); 440 } 441 s = malloc(sizeof (struct svc_callout), M_RPC, M_NOWAIT); 442 if (s == NULL) { 443 if (netid) 444 free(netid, M_RPC); 445 mtx_unlock(&pool->sp_lock); 446 return (FALSE); 447 } 448 449 s->sc_prog = prog; 450 s->sc_vers = vers; 451 s->sc_dispatch = dispatch; 452 s->sc_netid = netid; 453 TAILQ_INSERT_TAIL(&pool->sp_callouts, s, sc_link); 454 455 if ((xprt->xp_netid == NULL) && (flag == 1) && netid) 456 ((SVCXPRT *) xprt)->xp_netid = strdup(netid, M_RPC); 457 458 rpcb_it: 459 mtx_unlock(&pool->sp_lock); 460 /* now register the information with the local binder service */ 461 if (nconf) { 462 bool_t dummy; 463 struct netconfig tnc; 464 struct netbuf nb; 465 tnc = *nconf; 466 nb.buf = &xprt->xp_ltaddr; 467 nb.len = xprt->xp_ltaddr.ss_len; 468 dummy = rpcb_set(prog, vers, &tnc, &nb); 469 return (dummy); 470 } 471 return (TRUE); 472 } 473 474 /* 475 * Remove a service program from the callout list. 476 */ 477 void 478 svc_unreg(SVCPOOL *pool, const rpcprog_t prog, const rpcvers_t vers) 479 { 480 struct svc_callout *s; 481 482 /* unregister the information anyway */ 483 (void) rpcb_unset(prog, vers, NULL); 484 mtx_lock(&pool->sp_lock); 485 while ((s = svc_find(pool, prog, vers, NULL)) != NULL) { 486 TAILQ_REMOVE(&pool->sp_callouts, s, sc_link); 487 if (s->sc_netid) 488 mem_free(s->sc_netid, sizeof (s->sc_netid) + 1); 489 mem_free(s, sizeof (struct svc_callout)); 490 } 491 mtx_unlock(&pool->sp_lock); 492 } 493 494 /* ********************** CALLOUT list related stuff ************* */ 495 496 /* 497 * Search the callout list for a program number, return the callout 498 * struct. 499 */ 500 static struct svc_callout * 501 svc_find(SVCPOOL *pool, rpcprog_t prog, rpcvers_t vers, char *netid) 502 { 503 struct svc_callout *s; 504 505 mtx_assert(&pool->sp_lock, MA_OWNED); 506 TAILQ_FOREACH(s, &pool->sp_callouts, sc_link) { 507 if (s->sc_prog == prog && s->sc_vers == vers 508 && (netid == NULL || s->sc_netid == NULL || 509 strcmp(netid, s->sc_netid) == 0)) 510 break; 511 } 512 513 return (s); 514 } 515 516 /* ******************* REPLY GENERATION ROUTINES ************ */ 517 518 static bool_t 519 svc_sendreply_common(struct svc_req *rqstp, struct rpc_msg *rply, 520 struct mbuf *body) 521 { 522 SVCXPRT *xprt = rqstp->rq_xprt; 523 bool_t ok; 524 525 if (rqstp->rq_args) { 526 m_freem(rqstp->rq_args); 527 rqstp->rq_args = NULL; 528 } 529 530 if (xprt->xp_pool->sp_rcache) 531 replay_setreply(xprt->xp_pool->sp_rcache, 532 rply, svc_getrpccaller(rqstp), body); 533 534 if (!SVCAUTH_WRAP(&rqstp->rq_auth, &body)) 535 return (FALSE); 536 537 ok = SVC_REPLY(xprt, rply, rqstp->rq_addr, body); 538 if (rqstp->rq_addr) { 539 free(rqstp->rq_addr, M_SONAME); 540 rqstp->rq_addr = NULL; 541 } 542 543 return (ok); 544 } 545 546 /* 547 * Send a reply to an rpc request 548 */ 549 bool_t 550 svc_sendreply(struct svc_req *rqstp, xdrproc_t xdr_results, void * xdr_location) 551 { 552 struct rpc_msg rply; 553 struct mbuf *m; 554 XDR xdrs; 555 bool_t ok; 556 557 rply.rm_xid = rqstp->rq_xid; 558 rply.rm_direction = REPLY; 559 rply.rm_reply.rp_stat = MSG_ACCEPTED; 560 rply.acpted_rply.ar_verf = rqstp->rq_verf; 561 rply.acpted_rply.ar_stat = SUCCESS; 562 rply.acpted_rply.ar_results.where = NULL; 563 rply.acpted_rply.ar_results.proc = (xdrproc_t) xdr_void; 564 565 m = m_getcl(M_WAITOK, MT_DATA, 0); 566 xdrmbuf_create(&xdrs, m, XDR_ENCODE); 567 ok = xdr_results(&xdrs, xdr_location); 568 XDR_DESTROY(&xdrs); 569 570 if (ok) { 571 return (svc_sendreply_common(rqstp, &rply, m)); 572 } else { 573 m_freem(m); 574 return (FALSE); 575 } 576 } 577 578 bool_t 579 svc_sendreply_mbuf(struct svc_req *rqstp, struct mbuf *m) 580 { 581 struct rpc_msg rply; 582 583 rply.rm_xid = rqstp->rq_xid; 584 rply.rm_direction = REPLY; 585 rply.rm_reply.rp_stat = MSG_ACCEPTED; 586 rply.acpted_rply.ar_verf = rqstp->rq_verf; 587 rply.acpted_rply.ar_stat = SUCCESS; 588 rply.acpted_rply.ar_results.where = NULL; 589 rply.acpted_rply.ar_results.proc = (xdrproc_t) xdr_void; 590 591 return (svc_sendreply_common(rqstp, &rply, m)); 592 } 593 594 /* 595 * No procedure error reply 596 */ 597 void 598 svcerr_noproc(struct svc_req *rqstp) 599 { 600 SVCXPRT *xprt = rqstp->rq_xprt; 601 struct rpc_msg rply; 602 603 rply.rm_xid = rqstp->rq_xid; 604 rply.rm_direction = REPLY; 605 rply.rm_reply.rp_stat = MSG_ACCEPTED; 606 rply.acpted_rply.ar_verf = rqstp->rq_verf; 607 rply.acpted_rply.ar_stat = PROC_UNAVAIL; 608 609 if (xprt->xp_pool->sp_rcache) 610 replay_setreply(xprt->xp_pool->sp_rcache, 611 &rply, svc_getrpccaller(rqstp), NULL); 612 613 svc_sendreply_common(rqstp, &rply, NULL); 614 } 615 616 /* 617 * Can't decode args error reply 618 */ 619 void 620 svcerr_decode(struct svc_req *rqstp) 621 { 622 SVCXPRT *xprt = rqstp->rq_xprt; 623 struct rpc_msg rply; 624 625 rply.rm_xid = rqstp->rq_xid; 626 rply.rm_direction = REPLY; 627 rply.rm_reply.rp_stat = MSG_ACCEPTED; 628 rply.acpted_rply.ar_verf = rqstp->rq_verf; 629 rply.acpted_rply.ar_stat = GARBAGE_ARGS; 630 631 if (xprt->xp_pool->sp_rcache) 632 replay_setreply(xprt->xp_pool->sp_rcache, 633 &rply, (struct sockaddr *) &xprt->xp_rtaddr, NULL); 634 635 svc_sendreply_common(rqstp, &rply, NULL); 636 } 637 638 /* 639 * Some system error 640 */ 641 void 642 svcerr_systemerr(struct svc_req *rqstp) 643 { 644 SVCXPRT *xprt = rqstp->rq_xprt; 645 struct rpc_msg rply; 646 647 rply.rm_xid = rqstp->rq_xid; 648 rply.rm_direction = REPLY; 649 rply.rm_reply.rp_stat = MSG_ACCEPTED; 650 rply.acpted_rply.ar_verf = rqstp->rq_verf; 651 rply.acpted_rply.ar_stat = SYSTEM_ERR; 652 653 if (xprt->xp_pool->sp_rcache) 654 replay_setreply(xprt->xp_pool->sp_rcache, 655 &rply, svc_getrpccaller(rqstp), NULL); 656 657 svc_sendreply_common(rqstp, &rply, NULL); 658 } 659 660 /* 661 * Authentication error reply 662 */ 663 void 664 svcerr_auth(struct svc_req *rqstp, enum auth_stat why) 665 { 666 SVCXPRT *xprt = rqstp->rq_xprt; 667 struct rpc_msg rply; 668 669 rply.rm_xid = rqstp->rq_xid; 670 rply.rm_direction = REPLY; 671 rply.rm_reply.rp_stat = MSG_DENIED; 672 rply.rjcted_rply.rj_stat = AUTH_ERROR; 673 rply.rjcted_rply.rj_why = why; 674 675 if (xprt->xp_pool->sp_rcache) 676 replay_setreply(xprt->xp_pool->sp_rcache, 677 &rply, svc_getrpccaller(rqstp), NULL); 678 679 svc_sendreply_common(rqstp, &rply, NULL); 680 } 681 682 /* 683 * Auth too weak error reply 684 */ 685 void 686 svcerr_weakauth(struct svc_req *rqstp) 687 { 688 689 svcerr_auth(rqstp, AUTH_TOOWEAK); 690 } 691 692 /* 693 * Program unavailable error reply 694 */ 695 void 696 svcerr_noprog(struct svc_req *rqstp) 697 { 698 SVCXPRT *xprt = rqstp->rq_xprt; 699 struct rpc_msg rply; 700 701 rply.rm_xid = rqstp->rq_xid; 702 rply.rm_direction = REPLY; 703 rply.rm_reply.rp_stat = MSG_ACCEPTED; 704 rply.acpted_rply.ar_verf = rqstp->rq_verf; 705 rply.acpted_rply.ar_stat = PROG_UNAVAIL; 706 707 if (xprt->xp_pool->sp_rcache) 708 replay_setreply(xprt->xp_pool->sp_rcache, 709 &rply, svc_getrpccaller(rqstp), NULL); 710 711 svc_sendreply_common(rqstp, &rply, NULL); 712 } 713 714 /* 715 * Program version mismatch error reply 716 */ 717 void 718 svcerr_progvers(struct svc_req *rqstp, rpcvers_t low_vers, rpcvers_t high_vers) 719 { 720 SVCXPRT *xprt = rqstp->rq_xprt; 721 struct rpc_msg rply; 722 723 rply.rm_xid = rqstp->rq_xid; 724 rply.rm_direction = REPLY; 725 rply.rm_reply.rp_stat = MSG_ACCEPTED; 726 rply.acpted_rply.ar_verf = rqstp->rq_verf; 727 rply.acpted_rply.ar_stat = PROG_MISMATCH; 728 rply.acpted_rply.ar_vers.low = (uint32_t)low_vers; 729 rply.acpted_rply.ar_vers.high = (uint32_t)high_vers; 730 731 if (xprt->xp_pool->sp_rcache) 732 replay_setreply(xprt->xp_pool->sp_rcache, 733 &rply, svc_getrpccaller(rqstp), NULL); 734 735 svc_sendreply_common(rqstp, &rply, NULL); 736 } 737 738 /* 739 * Allocate a new server transport structure. All fields are 740 * initialized to zero and xp_p3 is initialized to point at an 741 * extension structure to hold various flags and authentication 742 * parameters. 743 */ 744 SVCXPRT * 745 svc_xprt_alloc() 746 { 747 SVCXPRT *xprt; 748 SVCXPRT_EXT *ext; 749 750 xprt = mem_alloc(sizeof(SVCXPRT)); 751 memset(xprt, 0, sizeof(SVCXPRT)); 752 ext = mem_alloc(sizeof(SVCXPRT_EXT)); 753 memset(ext, 0, sizeof(SVCXPRT_EXT)); 754 xprt->xp_p3 = ext; 755 refcount_init(&xprt->xp_refs, 1); 756 757 return (xprt); 758 } 759 760 /* 761 * Free a server transport structure. 762 */ 763 void 764 svc_xprt_free(xprt) 765 SVCXPRT *xprt; 766 { 767 768 mem_free(xprt->xp_p3, sizeof(SVCXPRT_EXT)); 769 mem_free(xprt, sizeof(SVCXPRT)); 770 } 771 772 /* ******************* SERVER INPUT STUFF ******************* */ 773 774 /* 775 * Read RPC requests from a transport and queue them to be 776 * executed. We handle authentication and replay cache replies here. 777 * Actually dispatching the RPC is deferred till svc_executereq. 778 */ 779 static enum xprt_stat 780 svc_getreq(SVCXPRT *xprt, struct svc_req **rqstp_ret) 781 { 782 SVCPOOL *pool = xprt->xp_pool; 783 struct svc_req *r; 784 struct rpc_msg msg; 785 struct mbuf *args; 786 enum xprt_stat stat; 787 788 /* now receive msgs from xprtprt (support batch calls) */ 789 r = malloc(sizeof(*r), M_RPC, M_WAITOK|M_ZERO); 790 791 msg.rm_call.cb_cred.oa_base = r->rq_credarea; 792 msg.rm_call.cb_verf.oa_base = &r->rq_credarea[MAX_AUTH_BYTES]; 793 r->rq_clntcred = &r->rq_credarea[2*MAX_AUTH_BYTES]; 794 if (SVC_RECV(xprt, &msg, &r->rq_addr, &args)) { 795 enum auth_stat why; 796 797 /* 798 * Handle replays and authenticate before queuing the 799 * request to be executed. 800 */ 801 SVC_ACQUIRE(xprt); 802 r->rq_xprt = xprt; 803 if (pool->sp_rcache) { 804 struct rpc_msg repmsg; 805 struct mbuf *repbody; 806 enum replay_state rs; 807 rs = replay_find(pool->sp_rcache, &msg, 808 svc_getrpccaller(r), &repmsg, &repbody); 809 switch (rs) { 810 case RS_NEW: 811 break; 812 case RS_DONE: 813 SVC_REPLY(xprt, &repmsg, r->rq_addr, 814 repbody); 815 if (r->rq_addr) { 816 free(r->rq_addr, M_SONAME); 817 r->rq_addr = NULL; 818 } 819 m_freem(args); 820 goto call_done; 821 822 default: 823 m_freem(args); 824 goto call_done; 825 } 826 } 827 828 r->rq_xid = msg.rm_xid; 829 r->rq_prog = msg.rm_call.cb_prog; 830 r->rq_vers = msg.rm_call.cb_vers; 831 r->rq_proc = msg.rm_call.cb_proc; 832 r->rq_size = sizeof(*r) + m_length(args, NULL); 833 r->rq_args = args; 834 if ((why = _authenticate(r, &msg)) != AUTH_OK) { 835 /* 836 * RPCSEC_GSS uses this return code 837 * for requests that form part of its 838 * context establishment protocol and 839 * should not be dispatched to the 840 * application. 841 */ 842 if (why != RPCSEC_GSS_NODISPATCH) 843 svcerr_auth(r, why); 844 goto call_done; 845 } 846 847 if (!SVCAUTH_UNWRAP(&r->rq_auth, &r->rq_args)) { 848 svcerr_decode(r); 849 goto call_done; 850 } 851 852 /* 853 * Everything checks out, return request to caller. 854 */ 855 *rqstp_ret = r; 856 r = NULL; 857 } 858 call_done: 859 if (r) { 860 svc_freereq(r); 861 r = NULL; 862 } 863 if ((stat = SVC_STAT(xprt)) == XPRT_DIED) { 864 xprt_unregister(xprt); 865 } 866 867 return (stat); 868 } 869 870 static void 871 svc_executereq(struct svc_req *rqstp) 872 { 873 SVCXPRT *xprt = rqstp->rq_xprt; 874 SVCPOOL *pool = xprt->xp_pool; 875 int prog_found; 876 rpcvers_t low_vers; 877 rpcvers_t high_vers; 878 struct svc_callout *s; 879 880 /* now match message with a registered service*/ 881 prog_found = FALSE; 882 low_vers = (rpcvers_t) -1L; 883 high_vers = (rpcvers_t) 0L; 884 TAILQ_FOREACH(s, &pool->sp_callouts, sc_link) { 885 if (s->sc_prog == rqstp->rq_prog) { 886 if (s->sc_vers == rqstp->rq_vers) { 887 /* 888 * We hand ownership of r to the 889 * dispatch method - they must call 890 * svc_freereq. 891 */ 892 (*s->sc_dispatch)(rqstp, xprt); 893 return; 894 } /* found correct version */ 895 prog_found = TRUE; 896 if (s->sc_vers < low_vers) 897 low_vers = s->sc_vers; 898 if (s->sc_vers > high_vers) 899 high_vers = s->sc_vers; 900 } /* found correct program */ 901 } 902 903 /* 904 * if we got here, the program or version 905 * is not served ... 906 */ 907 if (prog_found) 908 svcerr_progvers(rqstp, low_vers, high_vers); 909 else 910 svcerr_noprog(rqstp); 911 912 svc_freereq(rqstp); 913 } 914 915 static void 916 svc_checkidle(SVCPOOL *pool) 917 { 918 SVCXPRT *xprt, *nxprt; 919 time_t timo; 920 struct svcxprt_list cleanup; 921 922 TAILQ_INIT(&cleanup); 923 TAILQ_FOREACH_SAFE(xprt, &pool->sp_xlist, xp_link, nxprt) { 924 /* 925 * Only some transports have idle timers. Don't time 926 * something out which is just waking up. 927 */ 928 if (!xprt->xp_idletimeout || xprt->xp_thread) 929 continue; 930 931 timo = xprt->xp_lastactive + xprt->xp_idletimeout; 932 if (time_uptime > timo) { 933 xprt_unregister_locked(xprt); 934 TAILQ_INSERT_TAIL(&cleanup, xprt, xp_link); 935 } 936 } 937 938 mtx_unlock(&pool->sp_lock); 939 TAILQ_FOREACH_SAFE(xprt, &cleanup, xp_link, nxprt) { 940 SVC_RELEASE(xprt); 941 } 942 mtx_lock(&pool->sp_lock); 943 944 } 945 946 static void 947 svc_assign_waiting_sockets(SVCPOOL *pool) 948 { 949 SVCXPRT *xprt; 950 951 TAILQ_FOREACH(xprt, &pool->sp_active, xp_alink) { 952 if (!xprt->xp_thread) { 953 xprt_assignthread(xprt); 954 } 955 } 956 } 957 958 static bool_t 959 svc_request_space_available(SVCPOOL *pool) 960 { 961 962 mtx_assert(&pool->sp_lock, MA_OWNED); 963 964 if (pool->sp_space_throttled) { 965 /* 966 * Below the low-water yet? If so, assign any waiting sockets. 967 */ 968 if (pool->sp_space_used < pool->sp_space_low) { 969 pool->sp_space_throttled = FALSE; 970 svc_assign_waiting_sockets(pool); 971 return TRUE; 972 } 973 974 return FALSE; 975 } else { 976 if (pool->sp_space_used 977 >= pool->sp_space_high) { 978 pool->sp_space_throttled = TRUE; 979 pool->sp_space_throttle_count++; 980 return FALSE; 981 } 982 983 return TRUE; 984 } 985 } 986 987 static void 988 svc_run_internal(SVCPOOL *pool, bool_t ismaster) 989 { 990 SVCTHREAD *st, *stpref; 991 SVCXPRT *xprt; 992 enum xprt_stat stat; 993 struct svc_req *rqstp; 994 int error; 995 996 st = mem_alloc(sizeof(*st)); 997 st->st_xprt = NULL; 998 STAILQ_INIT(&st->st_reqs); 999 cv_init(&st->st_cond, "rpcsvc"); 1000 1001 mtx_lock(&pool->sp_lock); 1002 LIST_INSERT_HEAD(&pool->sp_threads, st, st_link); 1003 1004 /* 1005 * If we are a new thread which was spawned to cope with 1006 * increased load, set the state back to SVCPOOL_ACTIVE. 1007 */ 1008 if (pool->sp_state == SVCPOOL_THREADSTARTING) 1009 pool->sp_state = SVCPOOL_ACTIVE; 1010 1011 while (pool->sp_state != SVCPOOL_CLOSING) { 1012 /* 1013 * Create new thread if requested. 1014 */ 1015 if (pool->sp_state == SVCPOOL_THREADWANTED) { 1016 pool->sp_state = SVCPOOL_THREADSTARTING; 1017 pool->sp_lastcreatetime = time_uptime; 1018 mtx_unlock(&pool->sp_lock); 1019 svc_new_thread(pool); 1020 mtx_lock(&pool->sp_lock); 1021 continue; 1022 } 1023 1024 /* 1025 * Check for idle transports once per second. 1026 */ 1027 if (time_uptime > pool->sp_lastidlecheck) { 1028 pool->sp_lastidlecheck = time_uptime; 1029 svc_checkidle(pool); 1030 } 1031 1032 xprt = st->st_xprt; 1033 if (!xprt && STAILQ_EMPTY(&st->st_reqs)) { 1034 /* 1035 * Enforce maxthreads count. 1036 */ 1037 if (pool->sp_threadcount > pool->sp_maxthreads) 1038 break; 1039 1040 /* 1041 * Before sleeping, see if we can find an 1042 * active transport which isn't being serviced 1043 * by a thread. 1044 */ 1045 if (svc_request_space_available(pool)) { 1046 TAILQ_FOREACH(xprt, &pool->sp_active, 1047 xp_alink) { 1048 if (!xprt->xp_thread) { 1049 SVC_ACQUIRE(xprt); 1050 xprt->xp_thread = st; 1051 st->st_xprt = xprt; 1052 break; 1053 } 1054 } 1055 } 1056 if (st->st_xprt) 1057 continue; 1058 1059 LIST_INSERT_HEAD(&pool->sp_idlethreads, st, st_ilink); 1060 if (ismaster || (!ismaster && 1061 pool->sp_threadcount > pool->sp_minthreads)) 1062 error = cv_timedwait_sig(&st->st_cond, 1063 &pool->sp_lock, 5 * hz); 1064 else 1065 error = cv_wait_sig(&st->st_cond, 1066 &pool->sp_lock); 1067 LIST_REMOVE(st, st_ilink); 1068 1069 /* 1070 * Reduce worker thread count when idle. 1071 */ 1072 if (error == EWOULDBLOCK) { 1073 if (!ismaster 1074 && (pool->sp_threadcount 1075 > pool->sp_minthreads) 1076 && !st->st_xprt 1077 && STAILQ_EMPTY(&st->st_reqs)) 1078 break; 1079 } else if (error) { 1080 mtx_unlock(&pool->sp_lock); 1081 svc_exit(pool); 1082 mtx_lock(&pool->sp_lock); 1083 break; 1084 } 1085 continue; 1086 } 1087 1088 if (xprt) { 1089 /* 1090 * Drain the transport socket and queue up any 1091 * RPCs. 1092 */ 1093 xprt->xp_lastactive = time_uptime; 1094 stat = XPRT_IDLE; 1095 do { 1096 if (!svc_request_space_available(pool)) 1097 break; 1098 rqstp = NULL; 1099 mtx_unlock(&pool->sp_lock); 1100 stat = svc_getreq(xprt, &rqstp); 1101 mtx_lock(&pool->sp_lock); 1102 if (rqstp) { 1103 /* 1104 * See if the application has 1105 * a preference for some other 1106 * thread. 1107 */ 1108 stpref = st; 1109 if (pool->sp_assign) 1110 stpref = pool->sp_assign(st, 1111 rqstp); 1112 1113 pool->sp_space_used += 1114 rqstp->rq_size; 1115 if (pool->sp_space_used 1116 > pool->sp_space_used_highest) 1117 pool->sp_space_used_highest = 1118 pool->sp_space_used; 1119 rqstp->rq_thread = stpref; 1120 STAILQ_INSERT_TAIL(&stpref->st_reqs, 1121 rqstp, rq_link); 1122 stpref->st_reqcount++; 1123 1124 /* 1125 * If we assigned the request 1126 * to another thread, make 1127 * sure its awake and continue 1128 * reading from the 1129 * socket. Otherwise, try to 1130 * find some other thread to 1131 * read from the socket and 1132 * execute the request 1133 * immediately. 1134 */ 1135 if (stpref != st) { 1136 cv_signal(&stpref->st_cond); 1137 continue; 1138 } else { 1139 break; 1140 } 1141 } 1142 } while (stat == XPRT_MOREREQS 1143 && pool->sp_state != SVCPOOL_CLOSING); 1144 1145 /* 1146 * Move this transport to the end of the 1147 * active list to ensure fairness when 1148 * multiple transports are active. If this was 1149 * the last queued request, svc_getreq will 1150 * end up calling xprt_inactive to remove from 1151 * the active list. 1152 */ 1153 xprt->xp_thread = NULL; 1154 st->st_xprt = NULL; 1155 if (xprt->xp_active) { 1156 xprt_assignthread(xprt); 1157 TAILQ_REMOVE(&pool->sp_active, xprt, xp_alink); 1158 TAILQ_INSERT_TAIL(&pool->sp_active, xprt, 1159 xp_alink); 1160 } 1161 mtx_unlock(&pool->sp_lock); 1162 SVC_RELEASE(xprt); 1163 mtx_lock(&pool->sp_lock); 1164 } 1165 1166 /* 1167 * Execute what we have queued. 1168 */ 1169 while ((rqstp = STAILQ_FIRST(&st->st_reqs)) != NULL) { 1170 size_t sz = rqstp->rq_size; 1171 mtx_unlock(&pool->sp_lock); 1172 svc_executereq(rqstp); 1173 mtx_lock(&pool->sp_lock); 1174 pool->sp_space_used -= sz; 1175 } 1176 } 1177 1178 if (st->st_xprt) { 1179 xprt = st->st_xprt; 1180 st->st_xprt = NULL; 1181 SVC_RELEASE(xprt); 1182 } 1183 1184 KASSERT(STAILQ_EMPTY(&st->st_reqs), ("stray reqs on exit")); 1185 LIST_REMOVE(st, st_link); 1186 pool->sp_threadcount--; 1187 1188 mtx_unlock(&pool->sp_lock); 1189 1190 cv_destroy(&st->st_cond); 1191 mem_free(st, sizeof(*st)); 1192 1193 if (!ismaster) 1194 wakeup(pool); 1195 } 1196 1197 static void 1198 svc_thread_start(void *arg) 1199 { 1200 1201 svc_run_internal((SVCPOOL *) arg, FALSE); 1202 kthread_exit(); 1203 } 1204 1205 static void 1206 svc_new_thread(SVCPOOL *pool) 1207 { 1208 struct thread *td; 1209 1210 pool->sp_threadcount++; 1211 kthread_add(svc_thread_start, pool, 1212 pool->sp_proc, &td, 0, 0, 1213 "%s: service", pool->sp_name); 1214 } 1215 1216 void 1217 svc_run(SVCPOOL *pool) 1218 { 1219 int i; 1220 struct proc *p; 1221 struct thread *td; 1222 1223 p = curproc; 1224 td = curthread; 1225 snprintf(td->td_name, sizeof(td->td_name), 1226 "%s: master", pool->sp_name); 1227 pool->sp_state = SVCPOOL_ACTIVE; 1228 pool->sp_proc = p; 1229 pool->sp_lastcreatetime = time_uptime; 1230 pool->sp_threadcount = 1; 1231 1232 for (i = 1; i < pool->sp_minthreads; i++) { 1233 svc_new_thread(pool); 1234 } 1235 1236 svc_run_internal(pool, TRUE); 1237 1238 mtx_lock(&pool->sp_lock); 1239 while (pool->sp_threadcount > 0) 1240 msleep(pool, &pool->sp_lock, 0, "svcexit", 0); 1241 mtx_unlock(&pool->sp_lock); 1242 } 1243 1244 void 1245 svc_exit(SVCPOOL *pool) 1246 { 1247 SVCTHREAD *st; 1248 1249 mtx_lock(&pool->sp_lock); 1250 1251 if (pool->sp_state != SVCPOOL_CLOSING) { 1252 pool->sp_state = SVCPOOL_CLOSING; 1253 LIST_FOREACH(st, &pool->sp_idlethreads, st_ilink) 1254 cv_signal(&st->st_cond); 1255 } 1256 1257 mtx_unlock(&pool->sp_lock); 1258 } 1259 1260 bool_t 1261 svc_getargs(struct svc_req *rqstp, xdrproc_t xargs, void *args) 1262 { 1263 struct mbuf *m; 1264 XDR xdrs; 1265 bool_t stat; 1266 1267 m = rqstp->rq_args; 1268 rqstp->rq_args = NULL; 1269 1270 xdrmbuf_create(&xdrs, m, XDR_DECODE); 1271 stat = xargs(&xdrs, args); 1272 XDR_DESTROY(&xdrs); 1273 1274 return (stat); 1275 } 1276 1277 bool_t 1278 svc_freeargs(struct svc_req *rqstp, xdrproc_t xargs, void *args) 1279 { 1280 XDR xdrs; 1281 1282 if (rqstp->rq_addr) { 1283 free(rqstp->rq_addr, M_SONAME); 1284 rqstp->rq_addr = NULL; 1285 } 1286 1287 xdrs.x_op = XDR_FREE; 1288 return (xargs(&xdrs, args)); 1289 } 1290 1291 void 1292 svc_freereq(struct svc_req *rqstp) 1293 { 1294 SVCTHREAD *st; 1295 SVCXPRT *xprt; 1296 SVCPOOL *pool; 1297 1298 st = rqstp->rq_thread; 1299 xprt = rqstp->rq_xprt; 1300 if (xprt) 1301 pool = xprt->xp_pool; 1302 else 1303 pool = NULL; 1304 if (st) { 1305 mtx_lock(&pool->sp_lock); 1306 KASSERT(rqstp == STAILQ_FIRST(&st->st_reqs), 1307 ("Freeing request out of order")); 1308 STAILQ_REMOVE_HEAD(&st->st_reqs, rq_link); 1309 st->st_reqcount--; 1310 if (pool->sp_done) 1311 pool->sp_done(st, rqstp); 1312 mtx_unlock(&pool->sp_lock); 1313 } 1314 1315 if (rqstp->rq_auth.svc_ah_ops) 1316 SVCAUTH_RELEASE(&rqstp->rq_auth); 1317 1318 if (rqstp->rq_xprt) { 1319 SVC_RELEASE(rqstp->rq_xprt); 1320 } 1321 1322 if (rqstp->rq_addr) 1323 free(rqstp->rq_addr, M_SONAME); 1324 1325 if (rqstp->rq_args) 1326 m_freem(rqstp->rq_args); 1327 1328 free(rqstp, M_RPC); 1329 } 1330