1 /* $NetBSD: sys_select.c,v 1.14 2009/03/29 19:21:19 christos Exp $ */ 2 3 /*- 4 * Copyright (c) 2007, 2008, 2009 The NetBSD Foundation, Inc. 5 * All rights reserved. 6 * 7 * This code is derived from software contributed to The NetBSD Foundation 8 * by Andrew Doran. 9 * 10 * Redistribution and use in source and binary forms, with or without 11 * modification, are permitted provided that the following conditions 12 * are met: 13 * 1. Redistributions of source code must retain the above copyright 14 * notice, this list of conditions and the following disclaimer. 15 * 2. Redistributions in binary form must reproduce the above copyright 16 * notice, this list of conditions and the following disclaimer in the 17 * documentation and/or other materials provided with the distribution. 18 * 19 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS 20 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 21 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 22 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS 23 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 24 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 25 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 26 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 27 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 28 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 29 * POSSIBILITY OF SUCH DAMAGE. 30 */ 31 32 /* 33 * Copyright (c) 1982, 1986, 1989, 1993 34 * The Regents of the University of California. All rights reserved. 35 * (c) UNIX System Laboratories, Inc. 36 * All or some portions of this file are derived from material licensed 37 * to the University of California by American Telephone and Telegraph 38 * Co. or Unix System Laboratories, Inc. and are reproduced herein with 39 * the permission of UNIX System Laboratories, Inc. 40 * 41 * Redistribution and use in source and binary forms, with or without 42 * modification, are permitted provided that the following conditions 43 * are met: 44 * 1. Redistributions of source code must retain the above copyright 45 * notice, this list of conditions and the following disclaimer. 46 * 2. Redistributions in binary form must reproduce the above copyright 47 * notice, this list of conditions and the following disclaimer in the 48 * documentation and/or other materials provided with the distribution. 49 * 3. Neither the name of the University nor the names of its contributors 50 * may be used to endorse or promote products derived from this software 51 * without specific prior written permission. 52 * 53 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 54 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 55 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 56 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 57 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 58 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 59 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 60 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 61 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 62 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 63 * SUCH DAMAGE. 64 * 65 * @(#)sys_generic.c 8.9 (Berkeley) 2/14/95 66 */ 67 68 /* 69 * System calls relating to files. 70 */ 71 72 #include <sys/cdefs.h> 73 __KERNEL_RCSID(0, "$NetBSD: sys_select.c,v 1.14 2009/03/29 19:21:19 christos Exp $"); 74 75 #include <sys/param.h> 76 #include <sys/systm.h> 77 #include <sys/filedesc.h> 78 #include <sys/ioctl.h> 79 #include <sys/file.h> 80 #include <sys/proc.h> 81 #include <sys/socketvar.h> 82 #include <sys/signalvar.h> 83 #include <sys/uio.h> 84 #include <sys/kernel.h> 85 #include <sys/stat.h> 86 #include <sys/poll.h> 87 #include <sys/vnode.h> 88 #include <sys/mount.h> 89 #include <sys/syscallargs.h> 90 #include <sys/cpu.h> 91 #include <sys/atomic.h> 92 #include <sys/socketvar.h> 93 #include <sys/sleepq.h> 94 95 /* Flags for lwp::l_selflag. */ 96 #define SEL_RESET 0 /* awoken, interrupted, or not yet polling */ 97 #define SEL_SCANNING 1 /* polling descriptors */ 98 #define SEL_BLOCKING 2 /* about to block on select_cv */ 99 100 /* Per-CPU state for select()/poll(). */ 101 #if MAXCPUS > 32 102 #error adjust this code 103 #endif 104 typedef struct selcpu { 105 kmutex_t *sc_lock; 106 sleepq_t sc_sleepq; 107 int sc_ncoll; 108 uint32_t sc_mask; 109 } selcpu_t; 110 111 static int selscan(lwp_t *, fd_mask *, fd_mask *, int, register_t *); 112 static int pollscan(lwp_t *, struct pollfd *, int, register_t *); 113 static void selclear(void); 114 115 static syncobj_t select_sobj = { 116 SOBJ_SLEEPQ_FIFO, 117 sleepq_unsleep, 118 sleepq_changepri, 119 sleepq_lendpri, 120 syncobj_noowner, 121 }; 122 123 /* 124 * Select system call. 125 */ 126 int 127 sys___pselect50(struct lwp *l, const struct sys___pselect50_args *uap, 128 register_t *retval) 129 { 130 /* { 131 syscallarg(int) nd; 132 syscallarg(fd_set *) in; 133 syscallarg(fd_set *) ou; 134 syscallarg(fd_set *) ex; 135 syscallarg(const struct timespec *) ts; 136 syscallarg(sigset_t *) mask; 137 } */ 138 struct timespec ats, *ts = NULL; 139 sigset_t amask, *mask = NULL; 140 int error; 141 142 if (SCARG(uap, ts)) { 143 error = copyin(SCARG(uap, ts), &ats, sizeof(ats)); 144 if (error) 145 return error; 146 ts = &ats; 147 } 148 if (SCARG(uap, mask) != NULL) { 149 error = copyin(SCARG(uap, mask), &amask, sizeof(amask)); 150 if (error) 151 return error; 152 mask = &amask; 153 } 154 155 return selcommon(l, retval, SCARG(uap, nd), SCARG(uap, in), 156 SCARG(uap, ou), SCARG(uap, ex), ts, mask); 157 } 158 159 int 160 inittimeleft(struct timespec *ts, struct timespec *sleepts) 161 { 162 if (itimespecfix(ts)) 163 return -1; 164 getnanouptime(sleepts); 165 return 0; 166 } 167 168 int 169 gettimeleft(struct timespec *ts, struct timespec *sleepts) 170 { 171 /* 172 * We have to recalculate the timeout on every retry. 173 */ 174 struct timespec sleptts; 175 /* 176 * reduce ts by elapsed time 177 * based on monotonic time scale 178 */ 179 getnanouptime(&sleptts); 180 timespecadd(ts, sleepts, ts); 181 timespecsub(ts, &sleptts, ts); 182 *sleepts = sleptts; 183 return tstohz(ts); 184 } 185 186 int 187 sys___select50(struct lwp *l, const struct sys___select50_args *uap, 188 register_t *retval) 189 { 190 /* { 191 syscallarg(int) nd; 192 syscallarg(fd_set *) in; 193 syscallarg(fd_set *) ou; 194 syscallarg(fd_set *) ex; 195 syscallarg(struct timeval *) tv; 196 } */ 197 struct timeval atv; 198 struct timespec ats, *ts = NULL; 199 int error; 200 201 if (SCARG(uap, tv)) { 202 error = copyin(SCARG(uap, tv), (void *)&atv, sizeof(atv)); 203 if (error) 204 return error; 205 TIMEVAL_TO_TIMESPEC(&atv, &ats); 206 ts = &ats; 207 } 208 209 return selcommon(l, retval, SCARG(uap, nd), SCARG(uap, in), 210 SCARG(uap, ou), SCARG(uap, ex), ts, NULL); 211 } 212 213 int 214 selcommon(lwp_t *l, register_t *retval, int nd, fd_set *u_in, 215 fd_set *u_ou, fd_set *u_ex, struct timespec *ts, sigset_t *mask) 216 { 217 char smallbits[howmany(FD_SETSIZE, NFDBITS) * 218 sizeof(fd_mask) * 6]; 219 proc_t * const p = l->l_proc; 220 char *bits; 221 int ncoll, error, timo; 222 size_t ni; 223 sigset_t oldmask; 224 struct timespec sleepts; 225 selcpu_t *sc; 226 kmutex_t *lock; 227 228 error = 0; 229 if (nd < 0) 230 return (EINVAL); 231 if (nd > p->p_fd->fd_nfiles) { 232 /* forgiving; slightly wrong */ 233 nd = p->p_fd->fd_nfiles; 234 } 235 ni = howmany(nd, NFDBITS) * sizeof(fd_mask); 236 if (ni * 6 > sizeof(smallbits)) { 237 bits = kmem_alloc(ni * 6, KM_SLEEP); 238 if (bits == NULL) 239 return ENOMEM; 240 } else 241 bits = smallbits; 242 243 #define getbits(name, x) \ 244 if (u_ ## name) { \ 245 error = copyin(u_ ## name, bits + ni * x, ni); \ 246 if (error) \ 247 goto done; \ 248 } else \ 249 memset(bits + ni * x, 0, ni); 250 getbits(in, 0); 251 getbits(ou, 1); 252 getbits(ex, 2); 253 #undef getbits 254 255 timo = 0; 256 if (ts && inittimeleft(ts, &sleepts) == -1) { 257 error = EINVAL; 258 goto done; 259 } 260 261 if (mask) { 262 sigminusset(&sigcantmask, mask); 263 mutex_enter(p->p_lock); 264 oldmask = l->l_sigmask; 265 l->l_sigmask = *mask; 266 mutex_exit(p->p_lock); 267 } else 268 oldmask = l->l_sigmask; /* XXXgcc */ 269 270 sc = curcpu()->ci_data.cpu_selcpu; 271 lock = sc->sc_lock; 272 l->l_selcpu = sc; 273 SLIST_INIT(&l->l_selwait); 274 for (;;) { 275 /* 276 * No need to lock. If this is overwritten by another 277 * value while scanning, we will retry below. We only 278 * need to see exact state from the descriptors that 279 * we are about to poll, and lock activity resulting 280 * from fo_poll is enough to provide an up to date value 281 * for new polling activity. 282 */ 283 l->l_selflag = SEL_SCANNING; 284 ncoll = sc->sc_ncoll; 285 286 error = selscan(l, (fd_mask *)(bits + ni * 0), 287 (fd_mask *)(bits + ni * 3), nd, retval); 288 289 if (error || *retval) 290 break; 291 if (ts && (timo = gettimeleft(ts, &sleepts)) <= 0) 292 break; 293 mutex_spin_enter(lock); 294 if (l->l_selflag != SEL_SCANNING || sc->sc_ncoll != ncoll) { 295 mutex_spin_exit(lock); 296 continue; 297 } 298 l->l_selflag = SEL_BLOCKING; 299 l->l_kpriority = true; 300 sleepq_enter(&sc->sc_sleepq, l, lock); 301 sleepq_enqueue(&sc->sc_sleepq, sc, "select", &select_sobj); 302 error = sleepq_block(timo, true); 303 if (error != 0) 304 break; 305 } 306 selclear(); 307 308 if (mask) { 309 mutex_enter(p->p_lock); 310 l->l_sigmask = oldmask; 311 mutex_exit(p->p_lock); 312 } 313 314 done: 315 /* select is not restarted after signals... */ 316 if (error == ERESTART) 317 error = EINTR; 318 if (error == EWOULDBLOCK) 319 error = 0; 320 if (error == 0 && u_in != NULL) 321 error = copyout(bits + ni * 3, u_in, ni); 322 if (error == 0 && u_ou != NULL) 323 error = copyout(bits + ni * 4, u_ou, ni); 324 if (error == 0 && u_ex != NULL) 325 error = copyout(bits + ni * 5, u_ex, ni); 326 if (bits != smallbits) 327 kmem_free(bits, ni * 6); 328 return (error); 329 } 330 331 int 332 selscan(lwp_t *l, fd_mask *ibitp, fd_mask *obitp, int nfd, 333 register_t *retval) 334 { 335 static const int flag[3] = { POLLRDNORM | POLLHUP | POLLERR, 336 POLLWRNORM | POLLHUP | POLLERR, 337 POLLRDBAND }; 338 int msk, i, j, fd, n; 339 fd_mask ibits, obits; 340 file_t *fp; 341 342 n = 0; 343 for (msk = 0; msk < 3; msk++) { 344 for (i = 0; i < nfd; i += NFDBITS) { 345 ibits = *ibitp++; 346 obits = 0; 347 while ((j = ffs(ibits)) && (fd = i + --j) < nfd) { 348 ibits &= ~(1 << j); 349 if ((fp = fd_getfile(fd)) == NULL) 350 return (EBADF); 351 if ((*fp->f_ops->fo_poll)(fp, flag[msk])) { 352 obits |= (1 << j); 353 n++; 354 } 355 fd_putfile(fd); 356 } 357 *obitp++ = obits; 358 } 359 } 360 *retval = n; 361 return (0); 362 } 363 364 /* 365 * Poll system call. 366 */ 367 int 368 sys_poll(struct lwp *l, const struct sys_poll_args *uap, register_t *retval) 369 { 370 /* { 371 syscallarg(struct pollfd *) fds; 372 syscallarg(u_int) nfds; 373 syscallarg(int) timeout; 374 } */ 375 struct timespec ats, *ts = NULL; 376 377 if (SCARG(uap, timeout) != INFTIM) { 378 ats.tv_sec = SCARG(uap, timeout) / 1000; 379 ats.tv_nsec = (SCARG(uap, timeout) % 1000) * 1000000; 380 ts = &ats; 381 } 382 383 return pollcommon(l, retval, SCARG(uap, fds), SCARG(uap, nfds), 384 ts, NULL); 385 } 386 387 /* 388 * Poll system call. 389 */ 390 int 391 sys___pollts50(struct lwp *l, const struct sys___pollts50_args *uap, 392 register_t *retval) 393 { 394 /* { 395 syscallarg(struct pollfd *) fds; 396 syscallarg(u_int) nfds; 397 syscallarg(const struct timespec *) ts; 398 syscallarg(const sigset_t *) mask; 399 } */ 400 struct timespec ats, *ts = NULL; 401 sigset_t amask, *mask = NULL; 402 int error; 403 404 if (SCARG(uap, ts)) { 405 error = copyin(SCARG(uap, ts), &ats, sizeof(ats)); 406 if (error) 407 return error; 408 ts = &ats; 409 } 410 if (SCARG(uap, mask)) { 411 error = copyin(SCARG(uap, mask), &amask, sizeof(amask)); 412 if (error) 413 return error; 414 mask = &amask; 415 } 416 417 return pollcommon(l, retval, SCARG(uap, fds), SCARG(uap, nfds), 418 ts, mask); 419 } 420 421 int 422 pollcommon(lwp_t *l, register_t *retval, struct pollfd *u_fds, u_int nfds, 423 struct timespec *ts, sigset_t *mask) 424 { 425 struct pollfd smallfds[32]; 426 struct pollfd *fds; 427 proc_t * const p = l->l_proc; 428 sigset_t oldmask; 429 int ncoll, error, timo; 430 size_t ni; 431 struct timespec sleepts; 432 selcpu_t *sc; 433 kmutex_t *lock; 434 435 if (nfds > p->p_fd->fd_nfiles) { 436 /* forgiving; slightly wrong */ 437 nfds = p->p_fd->fd_nfiles; 438 } 439 ni = nfds * sizeof(struct pollfd); 440 if (ni > sizeof(smallfds)) { 441 fds = kmem_alloc(ni, KM_SLEEP); 442 if (fds == NULL) 443 return ENOMEM; 444 } else 445 fds = smallfds; 446 447 error = copyin(u_fds, fds, ni); 448 if (error) 449 goto done; 450 451 timo = 0; 452 if (ts && inittimeleft(ts, &sleepts) == -1) { 453 error = EINVAL; 454 goto done; 455 } 456 457 if (mask) { 458 sigminusset(&sigcantmask, mask); 459 mutex_enter(p->p_lock); 460 oldmask = l->l_sigmask; 461 l->l_sigmask = *mask; 462 mutex_exit(p->p_lock); 463 } else 464 oldmask = l->l_sigmask; /* XXXgcc */ 465 466 sc = curcpu()->ci_data.cpu_selcpu; 467 lock = sc->sc_lock; 468 l->l_selcpu = sc; 469 SLIST_INIT(&l->l_selwait); 470 for (;;) { 471 /* 472 * No need to lock. If this is overwritten by another 473 * value while scanning, we will retry below. We only 474 * need to see exact state from the descriptors that 475 * we are about to poll, and lock activity resulting 476 * from fo_poll is enough to provide an up to date value 477 * for new polling activity. 478 */ 479 ncoll = sc->sc_ncoll; 480 l->l_selflag = SEL_SCANNING; 481 482 error = pollscan(l, fds, nfds, retval); 483 484 if (error || *retval) 485 break; 486 if (ts && (timo = gettimeleft(ts, &sleepts)) <= 0) 487 break; 488 mutex_spin_enter(lock); 489 if (l->l_selflag != SEL_SCANNING || sc->sc_ncoll != ncoll) { 490 mutex_spin_exit(lock); 491 continue; 492 } 493 l->l_selflag = SEL_BLOCKING; 494 l->l_kpriority = true; 495 sleepq_enter(&sc->sc_sleepq, l, lock); 496 sleepq_enqueue(&sc->sc_sleepq, sc, "select", &select_sobj); 497 error = sleepq_block(timo, true); 498 if (error != 0) 499 break; 500 } 501 selclear(); 502 503 if (mask) { 504 mutex_enter(p->p_lock); 505 l->l_sigmask = oldmask; 506 mutex_exit(p->p_lock); 507 } 508 done: 509 /* poll is not restarted after signals... */ 510 if (error == ERESTART) 511 error = EINTR; 512 if (error == EWOULDBLOCK) 513 error = 0; 514 if (error == 0) 515 error = copyout(fds, u_fds, ni); 516 if (fds != smallfds) 517 kmem_free(fds, ni); 518 return (error); 519 } 520 521 int 522 pollscan(lwp_t *l, struct pollfd *fds, int nfd, register_t *retval) 523 { 524 int i, n; 525 file_t *fp; 526 527 n = 0; 528 for (i = 0; i < nfd; i++, fds++) { 529 if (fds->fd < 0) { 530 fds->revents = 0; 531 } else if ((fp = fd_getfile(fds->fd)) == NULL) { 532 fds->revents = POLLNVAL; 533 n++; 534 } else { 535 fds->revents = (*fp->f_ops->fo_poll)(fp, 536 fds->events | POLLERR | POLLHUP); 537 if (fds->revents != 0) 538 n++; 539 fd_putfile(fds->fd); 540 } 541 } 542 *retval = n; 543 return (0); 544 } 545 546 /*ARGSUSED*/ 547 int 548 seltrue(dev_t dev, int events, lwp_t *l) 549 { 550 551 return (events & (POLLIN | POLLOUT | POLLRDNORM | POLLWRNORM)); 552 } 553 554 /* 555 * Record a select request. Concurrency issues: 556 * 557 * The caller holds the same lock across calls to selrecord() and 558 * selnotify(), so we don't need to consider a concurrent wakeup 559 * while in this routine. 560 * 561 * The only activity we need to guard against is selclear(), called by 562 * another thread that is exiting selcommon() or pollcommon(). 563 * `sel_lwp' can only become non-NULL while the caller's lock is held, 564 * so it cannot become non-NULL due to a change made by another thread 565 * while we are in this routine. It can only become _NULL_ due to a 566 * call to selclear(). 567 * 568 * If it is non-NULL and != selector there is the potential for 569 * selclear() to be called by another thread. If either of those 570 * conditions are true, we're not interested in touching the `named 571 * waiter' part of the selinfo record because we need to record a 572 * collision. Hence there is no need for additional locking in this 573 * routine. 574 */ 575 void 576 selrecord(lwp_t *selector, struct selinfo *sip) 577 { 578 selcpu_t *sc; 579 lwp_t *other; 580 581 KASSERT(selector == curlwp); 582 583 sc = selector->l_selcpu; 584 other = sip->sel_lwp; 585 586 if (other == selector) { 587 /* `selector' has already claimed it. */ 588 KASSERT(sip->sel_cpu = sc); 589 } else if (other == NULL) { 590 /* 591 * First named waiter, although there may be unnamed 592 * waiters (collisions). Issue a memory barrier to 593 * ensure that we access sel_lwp (above) before other 594 * fields - this guards against a call to selclear(). 595 */ 596 membar_enter(); 597 sip->sel_lwp = selector; 598 SLIST_INSERT_HEAD(&selector->l_selwait, sip, sel_chain); 599 /* Replace selinfo's lock with our chosen CPU's lock. */ 600 sip->sel_cpu = sc; 601 } else { 602 /* Multiple waiters: record a collision. */ 603 sip->sel_collision |= sc->sc_mask; 604 KASSERT(sip->sel_cpu != NULL); 605 } 606 } 607 608 /* 609 * Do a wakeup when a selectable event occurs. Concurrency issues: 610 * 611 * As per selrecord(), the caller's object lock is held. If there 612 * is a named waiter, we must acquire the associated selcpu's lock 613 * in order to synchronize with selclear() and pollers going to sleep 614 * in selcommon() and/or pollcommon(). 615 * 616 * sip->sel_cpu cannot change at this point, as it is only changed 617 * in selrecord(), and concurrent calls to selrecord() are locked 618 * out by the caller. 619 */ 620 void 621 selnotify(struct selinfo *sip, int events, long knhint) 622 { 623 selcpu_t *sc; 624 uint32_t mask; 625 int index, oflag, swapin; 626 lwp_t *l; 627 kmutex_t *lock; 628 629 KNOTE(&sip->sel_klist, knhint); 630 631 if (sip->sel_lwp != NULL) { 632 /* One named LWP is waiting. */ 633 swapin = 0; 634 sc = sip->sel_cpu; 635 lock = sc->sc_lock; 636 mutex_spin_enter(lock); 637 /* Still there? */ 638 if (sip->sel_lwp != NULL) { 639 l = sip->sel_lwp; 640 /* 641 * If thread is sleeping, wake it up. If it's not 642 * yet asleep, it will notice the change in state 643 * and will re-poll the descriptors. 644 */ 645 oflag = l->l_selflag; 646 l->l_selflag = SEL_RESET; 647 if (oflag == SEL_BLOCKING && l->l_mutex == lock) { 648 KASSERT(l->l_wchan == sc); 649 swapin = sleepq_unsleep(l, false); 650 } 651 } 652 mutex_spin_exit(lock); 653 if (swapin) 654 uvm_kick_scheduler(); 655 } 656 657 if ((mask = sip->sel_collision) != 0) { 658 /* 659 * There was a collision (multiple waiters): we must 660 * inform all potentially interested waiters. 661 */ 662 sip->sel_collision = 0; 663 do { 664 index = ffs(mask) - 1; 665 mask &= ~(1 << index); 666 sc = cpu_lookup(index)->ci_data.cpu_selcpu; 667 lock = sc->sc_lock; 668 mutex_spin_enter(lock); 669 sc->sc_ncoll++; 670 sleepq_wake(&sc->sc_sleepq, sc, (u_int)-1, lock); 671 } while (__predict_false(mask != 0)); 672 } 673 } 674 675 /* 676 * Remove an LWP from all objects that it is waiting for. Concurrency 677 * issues: 678 * 679 * The object owner's (e.g. device driver) lock is not held here. Calls 680 * can be made to selrecord() and we do not synchronize against those 681 * directly using locks. However, we use `sel_lwp' to lock out changes. 682 * Before clearing it we must use memory barriers to ensure that we can 683 * safely traverse the list of selinfo records. 684 */ 685 static void 686 selclear(void) 687 { 688 struct selinfo *sip, *next; 689 selcpu_t *sc; 690 lwp_t *l; 691 kmutex_t *lock; 692 693 l = curlwp; 694 sc = l->l_selcpu; 695 lock = sc->sc_lock; 696 697 mutex_spin_enter(lock); 698 for (sip = SLIST_FIRST(&l->l_selwait); sip != NULL; sip = next) { 699 KASSERT(sip->sel_lwp == l); 700 KASSERT(sip->sel_cpu == l->l_selcpu); 701 /* 702 * Read link to next selinfo record, if any. 703 * It's no longer safe to touch `sip' after clearing 704 * `sel_lwp', so ensure that the read of `sel_chain' 705 * completes before the clearing of sel_lwp becomes 706 * globally visible. 707 */ 708 next = SLIST_NEXT(sip, sel_chain); 709 membar_exit(); 710 /* Release the record for another named waiter to use. */ 711 sip->sel_lwp = NULL; 712 } 713 mutex_spin_exit(lock); 714 } 715 716 /* 717 * Initialize the select/poll system calls. Called once for each 718 * CPU in the system, as they are attached. 719 */ 720 void 721 selsysinit(struct cpu_info *ci) 722 { 723 selcpu_t *sc; 724 725 sc = kmem_alloc(roundup2(sizeof(selcpu_t), coherency_unit) + 726 coherency_unit, KM_SLEEP); 727 sc = (void *)roundup2((uintptr_t)sc, coherency_unit); 728 sc->sc_lock = mutex_obj_alloc(MUTEX_DEFAULT, IPL_SCHED); 729 sleepq_init(&sc->sc_sleepq); 730 sc->sc_ncoll = 0; 731 sc->sc_mask = (1 << cpu_index(ci)); 732 ci->ci_data.cpu_selcpu = sc; 733 } 734 735 /* 736 * Initialize a selinfo record. 737 */ 738 void 739 selinit(struct selinfo *sip) 740 { 741 742 memset(sip, 0, sizeof(*sip)); 743 } 744 745 /* 746 * Destroy a selinfo record. The owning object must not gain new 747 * references while this is in progress: all activity on the record 748 * must be stopped. 749 * 750 * Concurrency issues: we only need guard against a call to selclear() 751 * by a thread exiting selcommon() and/or pollcommon(). The caller has 752 * prevented further references being made to the selinfo record via 753 * selrecord(), and it won't call selwakeup() again. 754 */ 755 void 756 seldestroy(struct selinfo *sip) 757 { 758 selcpu_t *sc; 759 kmutex_t *lock; 760 lwp_t *l; 761 762 if (sip->sel_lwp == NULL) 763 return; 764 765 /* 766 * Lock out selclear(). The selcpu pointer can't change while 767 * we are here since it is only ever changed in selrecord(), 768 * and that will not be entered again for this record because 769 * it is dying. 770 */ 771 KASSERT(sip->sel_cpu != NULL); 772 sc = sip->sel_cpu; 773 lock = sc->sc_lock; 774 mutex_spin_enter(lock); 775 if ((l = sip->sel_lwp) != NULL) { 776 /* 777 * This should rarely happen, so although SLIST_REMOVE() 778 * is slow, using it here is not a problem. 779 */ 780 KASSERT(l->l_selcpu == sc); 781 SLIST_REMOVE(&l->l_selwait, sip, selinfo, sel_chain); 782 sip->sel_lwp = NULL; 783 } 784 mutex_spin_exit(lock); 785 } 786 787 int 788 pollsock(struct socket *so, const struct timespec *tsp, int events) 789 { 790 int ncoll, error, timo; 791 struct timespec sleepts, ts; 792 selcpu_t *sc; 793 lwp_t *l; 794 kmutex_t *lock; 795 796 timo = 0; 797 if (tsp != NULL) { 798 ts = *tsp; 799 if (inittimeleft(&ts, &sleepts) == -1) 800 return EINVAL; 801 } 802 803 l = curlwp; 804 sc = l->l_cpu->ci_data.cpu_selcpu; 805 lock = sc->sc_lock; 806 l->l_selcpu = sc; 807 SLIST_INIT(&l->l_selwait); 808 error = 0; 809 for (;;) { 810 /* 811 * No need to lock. If this is overwritten by another 812 * value while scanning, we will retry below. We only 813 * need to see exact state from the descriptors that 814 * we are about to poll, and lock activity resulting 815 * from fo_poll is enough to provide an up to date value 816 * for new polling activity. 817 */ 818 ncoll = sc->sc_ncoll; 819 l->l_selflag = SEL_SCANNING; 820 if (sopoll(so, events) != 0) 821 break; 822 if (tsp && (timo = gettimeleft(&ts, &sleepts)) <= 0) 823 break; 824 mutex_spin_enter(lock); 825 if (l->l_selflag != SEL_SCANNING || sc->sc_ncoll != ncoll) { 826 mutex_spin_exit(lock); 827 continue; 828 } 829 l->l_selflag = SEL_BLOCKING; 830 sleepq_enter(&sc->sc_sleepq, l, lock); 831 sleepq_enqueue(&sc->sc_sleepq, sc, "pollsock", &select_sobj); 832 error = sleepq_block(timo, true); 833 if (error != 0) 834 break; 835 } 836 selclear(); 837 /* poll is not restarted after signals... */ 838 if (error == ERESTART) 839 error = EINTR; 840 if (error == EWOULDBLOCK) 841 error = 0; 842 return (error); 843 } 844