1 /* $NetBSD: vfs_bio.c,v 1.260 2016/07/31 04:05:32 dholland 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, and by Wasabi Systems, Inc. 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 * @(#)vfs_bio.c 8.6 (Berkeley) 1/11/94 66 */ 67 68 /*- 69 * Copyright (c) 1994 Christopher G. Demetriou 70 * 71 * Redistribution and use in source and binary forms, with or without 72 * modification, are permitted provided that the following conditions 73 * are met: 74 * 1. Redistributions of source code must retain the above copyright 75 * notice, this list of conditions and the following disclaimer. 76 * 2. Redistributions in binary form must reproduce the above copyright 77 * notice, this list of conditions and the following disclaimer in the 78 * documentation and/or other materials provided with the distribution. 79 * 3. All advertising materials mentioning features or use of this software 80 * must display the following acknowledgement: 81 * This product includes software developed by the University of 82 * California, Berkeley and its contributors. 83 * 4. Neither the name of the University nor the names of its contributors 84 * may be used to endorse or promote products derived from this software 85 * without specific prior written permission. 86 * 87 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 88 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 89 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 90 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 91 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 92 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 93 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 94 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 95 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 96 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 97 * SUCH DAMAGE. 98 * 99 * @(#)vfs_bio.c 8.6 (Berkeley) 1/11/94 100 */ 101 102 /* 103 * The buffer cache subsystem. 104 * 105 * Some references: 106 * Bach: The Design of the UNIX Operating System (Prentice Hall, 1986) 107 * Leffler, et al.: The Design and Implementation of the 4.3BSD 108 * UNIX Operating System (Addison Welley, 1989) 109 * 110 * Locking 111 * 112 * There are three locks: 113 * - bufcache_lock: protects global buffer cache state. 114 * - BC_BUSY: a long term per-buffer lock. 115 * - buf_t::b_objlock: lock on completion (biowait vs biodone). 116 * 117 * For buffers associated with vnodes (a most common case) b_objlock points 118 * to the vnode_t::v_interlock. Otherwise, it points to generic buffer_lock. 119 * 120 * Lock order: 121 * bufcache_lock -> 122 * buf_t::b_objlock 123 */ 124 125 #include <sys/cdefs.h> 126 __KERNEL_RCSID(0, "$NetBSD: vfs_bio.c,v 1.260 2016/07/31 04:05:32 dholland Exp $"); 127 128 #ifdef _KERNEL_OPT 129 #include "opt_bufcache.h" 130 #include "opt_dtrace.h" 131 #endif 132 133 #include <sys/param.h> 134 #include <sys/systm.h> 135 #include <sys/kernel.h> 136 #include <sys/proc.h> 137 #include <sys/buf.h> 138 #include <sys/vnode.h> 139 #include <sys/mount.h> 140 #include <sys/resourcevar.h> 141 #include <sys/sysctl.h> 142 #include <sys/conf.h> 143 #include <sys/kauth.h> 144 #include <sys/fstrans.h> 145 #include <sys/intr.h> 146 #include <sys/cpu.h> 147 #include <sys/wapbl.h> 148 #include <sys/bitops.h> 149 #include <sys/cprng.h> 150 #include <sys/sdt.h> 151 152 #include <uvm/uvm.h> /* extern struct uvm uvm */ 153 154 #include <miscfs/specfs/specdev.h> 155 156 #ifndef BUFPAGES 157 # define BUFPAGES 0 158 #endif 159 160 #ifdef BUFCACHE 161 # if (BUFCACHE < 5) || (BUFCACHE > 95) 162 # error BUFCACHE is not between 5 and 95 163 # endif 164 #else 165 # define BUFCACHE 15 166 #endif 167 168 u_int nbuf; /* desired number of buffer headers */ 169 u_int bufpages = BUFPAGES; /* optional hardwired count */ 170 u_int bufcache = BUFCACHE; /* max % of RAM to use for buffer cache */ 171 172 /* Function prototypes */ 173 struct bqueue; 174 175 static void buf_setwm(void); 176 static int buf_trim(void); 177 static void *bufpool_page_alloc(struct pool *, int); 178 static void bufpool_page_free(struct pool *, void *); 179 static buf_t *bio_doread(struct vnode *, daddr_t, int, int); 180 static buf_t *getnewbuf(int, int, int); 181 static int buf_lotsfree(void); 182 static int buf_canrelease(void); 183 static u_long buf_mempoolidx(u_long); 184 static u_long buf_roundsize(u_long); 185 static void *buf_alloc(size_t); 186 static void buf_mrelease(void *, size_t); 187 static void binsheadfree(buf_t *, struct bqueue *); 188 static void binstailfree(buf_t *, struct bqueue *); 189 #ifdef DEBUG 190 static int checkfreelist(buf_t *, struct bqueue *, int); 191 #endif 192 static void biointr(void *); 193 static void biodone2(buf_t *); 194 static void bref(buf_t *); 195 static void brele(buf_t *); 196 static void sysctl_kern_buf_setup(void); 197 static void sysctl_vm_buf_setup(void); 198 199 /* 200 * Definitions for the buffer hash lists. 201 */ 202 #define BUFHASH(dvp, lbn) \ 203 (&bufhashtbl[(((long)(dvp) >> 8) + (int)(lbn)) & bufhash]) 204 LIST_HEAD(bufhashhdr, buf) *bufhashtbl, invalhash; 205 u_long bufhash; 206 struct bqueue bufqueues[BQUEUES]; 207 208 static kcondvar_t needbuffer_cv; 209 210 /* 211 * Buffer queue lock. 212 */ 213 kmutex_t bufcache_lock; 214 kmutex_t buffer_lock; 215 216 /* Software ISR for completed transfers. */ 217 static void *biodone_sih; 218 219 /* Buffer pool for I/O buffers. */ 220 static pool_cache_t buf_cache; 221 static pool_cache_t bufio_cache; 222 223 #define MEMPOOL_INDEX_OFFSET (ilog2(DEV_BSIZE)) /* smallest pool is 512 bytes */ 224 #define NMEMPOOLS (ilog2(MAXBSIZE) - MEMPOOL_INDEX_OFFSET + 1) 225 __CTASSERT((1 << (NMEMPOOLS + MEMPOOL_INDEX_OFFSET - 1)) == MAXBSIZE); 226 227 /* Buffer memory pools */ 228 static struct pool bmempools[NMEMPOOLS]; 229 230 static struct vm_map *buf_map; 231 232 /* 233 * Buffer memory pool allocator. 234 */ 235 static void * 236 bufpool_page_alloc(struct pool *pp, int flags) 237 { 238 239 return (void *)uvm_km_alloc(buf_map, 240 MAXBSIZE, MAXBSIZE, 241 ((flags & PR_WAITOK) ? 0 : UVM_KMF_NOWAIT|UVM_KMF_TRYLOCK) 242 | UVM_KMF_WIRED); 243 } 244 245 static void 246 bufpool_page_free(struct pool *pp, void *v) 247 { 248 249 uvm_km_free(buf_map, (vaddr_t)v, MAXBSIZE, UVM_KMF_WIRED); 250 } 251 252 static struct pool_allocator bufmempool_allocator = { 253 .pa_alloc = bufpool_page_alloc, 254 .pa_free = bufpool_page_free, 255 .pa_pagesz = MAXBSIZE, 256 }; 257 258 /* Buffer memory management variables */ 259 u_long bufmem_valimit; 260 u_long bufmem_hiwater; 261 u_long bufmem_lowater; 262 u_long bufmem; 263 264 /* 265 * MD code can call this to set a hard limit on the amount 266 * of virtual memory used by the buffer cache. 267 */ 268 int 269 buf_setvalimit(vsize_t sz) 270 { 271 272 /* We need to accommodate at least NMEMPOOLS of MAXBSIZE each */ 273 if (sz < NMEMPOOLS * MAXBSIZE) 274 return EINVAL; 275 276 bufmem_valimit = sz; 277 return 0; 278 } 279 280 static void 281 buf_setwm(void) 282 { 283 284 bufmem_hiwater = buf_memcalc(); 285 /* lowater is approx. 2% of memory (with bufcache = 15) */ 286 #define BUFMEM_WMSHIFT 3 287 #define BUFMEM_HIWMMIN (64 * 1024 << BUFMEM_WMSHIFT) 288 if (bufmem_hiwater < BUFMEM_HIWMMIN) 289 /* Ensure a reasonable minimum value */ 290 bufmem_hiwater = BUFMEM_HIWMMIN; 291 bufmem_lowater = bufmem_hiwater >> BUFMEM_WMSHIFT; 292 } 293 294 #ifdef DEBUG 295 int debug_verify_freelist = 0; 296 static int 297 checkfreelist(buf_t *bp, struct bqueue *dp, int ison) 298 { 299 buf_t *b; 300 301 if (!debug_verify_freelist) 302 return 1; 303 304 TAILQ_FOREACH(b, &dp->bq_queue, b_freelist) { 305 if (b == bp) 306 return ison ? 1 : 0; 307 } 308 309 return ison ? 0 : 1; 310 } 311 #endif 312 313 /* 314 * Insq/Remq for the buffer hash lists. 315 * Call with buffer queue locked. 316 */ 317 static void 318 binsheadfree(buf_t *bp, struct bqueue *dp) 319 { 320 321 KASSERT(mutex_owned(&bufcache_lock)); 322 KASSERT(bp->b_freelistindex == -1); 323 TAILQ_INSERT_HEAD(&dp->bq_queue, bp, b_freelist); 324 dp->bq_bytes += bp->b_bufsize; 325 bp->b_freelistindex = dp - bufqueues; 326 } 327 328 static void 329 binstailfree(buf_t *bp, struct bqueue *dp) 330 { 331 332 KASSERT(mutex_owned(&bufcache_lock)); 333 KASSERTMSG(bp->b_freelistindex == -1, "double free of buffer? " 334 "bp=%p, b_freelistindex=%d\n", bp, bp->b_freelistindex); 335 TAILQ_INSERT_TAIL(&dp->bq_queue, bp, b_freelist); 336 dp->bq_bytes += bp->b_bufsize; 337 bp->b_freelistindex = dp - bufqueues; 338 } 339 340 void 341 bremfree(buf_t *bp) 342 { 343 struct bqueue *dp; 344 int bqidx = bp->b_freelistindex; 345 346 KASSERT(mutex_owned(&bufcache_lock)); 347 348 KASSERT(bqidx != -1); 349 dp = &bufqueues[bqidx]; 350 KDASSERT(checkfreelist(bp, dp, 1)); 351 KASSERT(dp->bq_bytes >= bp->b_bufsize); 352 TAILQ_REMOVE(&dp->bq_queue, bp, b_freelist); 353 dp->bq_bytes -= bp->b_bufsize; 354 355 /* For the sysctl helper. */ 356 if (bp == dp->bq_marker) 357 dp->bq_marker = NULL; 358 359 #if defined(DIAGNOSTIC) 360 bp->b_freelistindex = -1; 361 #endif /* defined(DIAGNOSTIC) */ 362 } 363 364 /* 365 * Add a reference to an buffer structure that came from buf_cache. 366 */ 367 static inline void 368 bref(buf_t *bp) 369 { 370 371 KASSERT(mutex_owned(&bufcache_lock)); 372 KASSERT(bp->b_refcnt > 0); 373 374 bp->b_refcnt++; 375 } 376 377 /* 378 * Free an unused buffer structure that came from buf_cache. 379 */ 380 static inline void 381 brele(buf_t *bp) 382 { 383 384 KASSERT(mutex_owned(&bufcache_lock)); 385 KASSERT(bp->b_refcnt > 0); 386 387 if (bp->b_refcnt-- == 1) { 388 buf_destroy(bp); 389 #ifdef DEBUG 390 memset((char *)bp, 0, sizeof(*bp)); 391 #endif 392 pool_cache_put(buf_cache, bp); 393 } 394 } 395 396 /* 397 * note that for some ports this is used by pmap bootstrap code to 398 * determine kva size. 399 */ 400 u_long 401 buf_memcalc(void) 402 { 403 u_long n; 404 vsize_t mapsz = 0; 405 406 /* 407 * Determine the upper bound of memory to use for buffers. 408 * 409 * - If bufpages is specified, use that as the number 410 * pages. 411 * 412 * - Otherwise, use bufcache as the percentage of 413 * physical memory. 414 */ 415 if (bufpages != 0) { 416 n = bufpages; 417 } else { 418 if (bufcache < 5) { 419 printf("forcing bufcache %d -> 5", bufcache); 420 bufcache = 5; 421 } 422 if (bufcache > 95) { 423 printf("forcing bufcache %d -> 95", bufcache); 424 bufcache = 95; 425 } 426 if (buf_map != NULL) 427 mapsz = vm_map_max(buf_map) - vm_map_min(buf_map); 428 n = calc_cache_size(mapsz, bufcache, 429 (buf_map != kernel_map) ? 100 : BUFCACHE_VA_MAXPCT) 430 / PAGE_SIZE; 431 } 432 433 n <<= PAGE_SHIFT; 434 if (bufmem_valimit != 0 && n > bufmem_valimit) 435 n = bufmem_valimit; 436 437 return (n); 438 } 439 440 /* 441 * Initialize buffers and hash links for buffers. 442 */ 443 void 444 bufinit(void) 445 { 446 struct bqueue *dp; 447 int use_std; 448 u_int i; 449 450 biodone_vfs = biodone; 451 452 mutex_init(&bufcache_lock, MUTEX_DEFAULT, IPL_NONE); 453 mutex_init(&buffer_lock, MUTEX_DEFAULT, IPL_NONE); 454 cv_init(&needbuffer_cv, "needbuf"); 455 456 if (bufmem_valimit != 0) { 457 vaddr_t minaddr = 0, maxaddr; 458 buf_map = uvm_km_suballoc(kernel_map, &minaddr, &maxaddr, 459 bufmem_valimit, 0, false, 0); 460 if (buf_map == NULL) 461 panic("bufinit: cannot allocate submap"); 462 } else 463 buf_map = kernel_map; 464 465 /* 466 * Initialize buffer cache memory parameters. 467 */ 468 bufmem = 0; 469 buf_setwm(); 470 471 /* On "small" machines use small pool page sizes where possible */ 472 use_std = (physmem < atop(16*1024*1024)); 473 474 /* 475 * Also use them on systems that can map the pool pages using 476 * a direct-mapped segment. 477 */ 478 #ifdef PMAP_MAP_POOLPAGE 479 use_std = 1; 480 #endif 481 482 buf_cache = pool_cache_init(sizeof(buf_t), 0, 0, 0, 483 "bufpl", NULL, IPL_SOFTBIO, NULL, NULL, NULL); 484 bufio_cache = pool_cache_init(sizeof(buf_t), 0, 0, 0, 485 "biopl", NULL, IPL_BIO, NULL, NULL, NULL); 486 487 for (i = 0; i < NMEMPOOLS; i++) { 488 struct pool_allocator *pa; 489 struct pool *pp = &bmempools[i]; 490 u_int size = 1 << (i + MEMPOOL_INDEX_OFFSET); 491 char *name = kmem_alloc(8, KM_SLEEP); /* XXX: never freed */ 492 if (__predict_false(size >= 1048576)) 493 (void)snprintf(name, 8, "buf%um", size / 1048576); 494 else if (__predict_true(size >= 1024)) 495 (void)snprintf(name, 8, "buf%uk", size / 1024); 496 else 497 (void)snprintf(name, 8, "buf%ub", size); 498 pa = (size <= PAGE_SIZE && use_std) 499 ? &pool_allocator_nointr 500 : &bufmempool_allocator; 501 pool_init(pp, size, 0, 0, 0, name, pa, IPL_NONE); 502 pool_setlowat(pp, 1); 503 pool_sethiwat(pp, 1); 504 } 505 506 /* Initialize the buffer queues */ 507 for (dp = bufqueues; dp < &bufqueues[BQUEUES]; dp++) { 508 TAILQ_INIT(&dp->bq_queue); 509 dp->bq_bytes = 0; 510 } 511 512 /* 513 * Estimate hash table size based on the amount of memory we 514 * intend to use for the buffer cache. The average buffer 515 * size is dependent on our clients (i.e. filesystems). 516 * 517 * For now, use an empirical 3K per buffer. 518 */ 519 nbuf = (bufmem_hiwater / 1024) / 3; 520 bufhashtbl = hashinit(nbuf, HASH_LIST, true, &bufhash); 521 522 sysctl_kern_buf_setup(); 523 sysctl_vm_buf_setup(); 524 } 525 526 void 527 bufinit2(void) 528 { 529 530 biodone_sih = softint_establish(SOFTINT_BIO | SOFTINT_MPSAFE, biointr, 531 NULL); 532 if (biodone_sih == NULL) 533 panic("bufinit2: can't establish soft interrupt"); 534 } 535 536 static int 537 buf_lotsfree(void) 538 { 539 u_long guess; 540 541 /* Always allocate if less than the low water mark. */ 542 if (bufmem < bufmem_lowater) 543 return 1; 544 545 /* Never allocate if greater than the high water mark. */ 546 if (bufmem > bufmem_hiwater) 547 return 0; 548 549 /* If there's anything on the AGE list, it should be eaten. */ 550 if (TAILQ_FIRST(&bufqueues[BQ_AGE].bq_queue) != NULL) 551 return 0; 552 553 /* 554 * The probabily of getting a new allocation is inversely 555 * proportional to the current size of the cache above 556 * the low water mark. Divide the total first to avoid overflows 557 * in the product. 558 */ 559 guess = cprng_fast32() % 16; 560 561 if ((bufmem_hiwater - bufmem_lowater) / 16 * guess >= 562 (bufmem - bufmem_lowater)) 563 return 1; 564 565 /* Otherwise don't allocate. */ 566 return 0; 567 } 568 569 /* 570 * Return estimate of bytes we think need to be 571 * released to help resolve low memory conditions. 572 * 573 * => called with bufcache_lock held. 574 */ 575 static int 576 buf_canrelease(void) 577 { 578 int pagedemand, ninvalid = 0; 579 580 KASSERT(mutex_owned(&bufcache_lock)); 581 582 if (bufmem < bufmem_lowater) 583 return 0; 584 585 if (bufmem > bufmem_hiwater) 586 return bufmem - bufmem_hiwater; 587 588 ninvalid += bufqueues[BQ_AGE].bq_bytes; 589 590 pagedemand = uvmexp.freetarg - uvmexp.free; 591 if (pagedemand < 0) 592 return ninvalid; 593 return MAX(ninvalid, MIN(2 * MAXBSIZE, 594 MIN((bufmem - bufmem_lowater) / 16, pagedemand * PAGE_SIZE))); 595 } 596 597 /* 598 * Buffer memory allocation helper functions 599 */ 600 static u_long 601 buf_mempoolidx(u_long size) 602 { 603 u_int n = 0; 604 605 size -= 1; 606 size >>= MEMPOOL_INDEX_OFFSET; 607 while (size) { 608 size >>= 1; 609 n += 1; 610 } 611 if (n >= NMEMPOOLS) 612 panic("buf mem pool index %d", n); 613 return n; 614 } 615 616 static u_long 617 buf_roundsize(u_long size) 618 { 619 /* Round up to nearest power of 2 */ 620 return (1 << (buf_mempoolidx(size) + MEMPOOL_INDEX_OFFSET)); 621 } 622 623 static void * 624 buf_alloc(size_t size) 625 { 626 u_int n = buf_mempoolidx(size); 627 void *addr; 628 629 while (1) { 630 addr = pool_get(&bmempools[n], PR_NOWAIT); 631 if (addr != NULL) 632 break; 633 634 /* No memory, see if we can free some. If so, try again */ 635 mutex_enter(&bufcache_lock); 636 if (buf_drain(1) > 0) { 637 mutex_exit(&bufcache_lock); 638 continue; 639 } 640 641 if (curlwp == uvm.pagedaemon_lwp) { 642 mutex_exit(&bufcache_lock); 643 return NULL; 644 } 645 646 /* Wait for buffers to arrive on the LRU queue */ 647 cv_timedwait(&needbuffer_cv, &bufcache_lock, hz / 4); 648 mutex_exit(&bufcache_lock); 649 } 650 651 return addr; 652 } 653 654 static void 655 buf_mrelease(void *addr, size_t size) 656 { 657 658 pool_put(&bmempools[buf_mempoolidx(size)], addr); 659 } 660 661 /* 662 * bread()/breadn() helper. 663 */ 664 static buf_t * 665 bio_doread(struct vnode *vp, daddr_t blkno, int size, int async) 666 { 667 buf_t *bp; 668 struct mount *mp; 669 670 bp = getblk(vp, blkno, size, 0, 0); 671 672 /* 673 * getblk() may return NULL if we are the pagedaemon. 674 */ 675 if (bp == NULL) { 676 KASSERT(curlwp == uvm.pagedaemon_lwp); 677 return NULL; 678 } 679 680 /* 681 * If buffer does not have data valid, start a read. 682 * Note that if buffer is BC_INVAL, getblk() won't return it. 683 * Therefore, it's valid if its I/O has completed or been delayed. 684 */ 685 if (!ISSET(bp->b_oflags, (BO_DONE | BO_DELWRI))) { 686 /* Start I/O for the buffer. */ 687 SET(bp->b_flags, B_READ | async); 688 if (async) 689 BIO_SETPRIO(bp, BPRIO_TIMELIMITED); 690 else 691 BIO_SETPRIO(bp, BPRIO_TIMECRITICAL); 692 VOP_STRATEGY(vp, bp); 693 694 /* Pay for the read. */ 695 curlwp->l_ru.ru_inblock++; 696 } else if (async) 697 brelse(bp, 0); 698 699 if (vp->v_type == VBLK) 700 mp = spec_node_getmountedfs(vp); 701 else 702 mp = vp->v_mount; 703 704 /* 705 * Collect statistics on synchronous and asynchronous reads. 706 * Reads from block devices are charged to their associated 707 * filesystem (if any). 708 */ 709 if (mp != NULL) { 710 if (async == 0) 711 mp->mnt_stat.f_syncreads++; 712 else 713 mp->mnt_stat.f_asyncreads++; 714 } 715 716 return (bp); 717 } 718 719 /* 720 * Read a disk block. 721 * This algorithm described in Bach (p.54). 722 */ 723 int 724 bread(struct vnode *vp, daddr_t blkno, int size, int flags, buf_t **bpp) 725 { 726 buf_t *bp; 727 int error; 728 729 /* Get buffer for block. */ 730 bp = *bpp = bio_doread(vp, blkno, size, 0); 731 if (bp == NULL) 732 return ENOMEM; 733 734 /* Wait for the read to complete, and return result. */ 735 error = biowait(bp); 736 if (error == 0 && (flags & B_MODIFY) != 0) 737 error = fscow_run(bp, true); 738 if (error) { 739 brelse(bp, 0); 740 *bpp = NULL; 741 } 742 743 return error; 744 } 745 746 /* 747 * Read-ahead multiple disk blocks. The first is sync, the rest async. 748 * Trivial modification to the breada algorithm presented in Bach (p.55). 749 */ 750 int 751 breadn(struct vnode *vp, daddr_t blkno, int size, daddr_t *rablks, 752 int *rasizes, int nrablks, int flags, buf_t **bpp) 753 { 754 buf_t *bp; 755 int error, i; 756 757 bp = *bpp = bio_doread(vp, blkno, size, 0); 758 if (bp == NULL) 759 return ENOMEM; 760 761 /* 762 * For each of the read-ahead blocks, start a read, if necessary. 763 */ 764 mutex_enter(&bufcache_lock); 765 for (i = 0; i < nrablks; i++) { 766 /* If it's in the cache, just go on to next one. */ 767 if (incore(vp, rablks[i])) 768 continue; 769 770 /* Get a buffer for the read-ahead block */ 771 mutex_exit(&bufcache_lock); 772 (void) bio_doread(vp, rablks[i], rasizes[i], B_ASYNC); 773 mutex_enter(&bufcache_lock); 774 } 775 mutex_exit(&bufcache_lock); 776 777 /* Otherwise, we had to start a read for it; wait until it's valid. */ 778 error = biowait(bp); 779 if (error == 0 && (flags & B_MODIFY) != 0) 780 error = fscow_run(bp, true); 781 if (error) { 782 brelse(bp, 0); 783 *bpp = NULL; 784 } 785 786 return error; 787 } 788 789 /* 790 * Block write. Described in Bach (p.56) 791 */ 792 int 793 bwrite(buf_t *bp) 794 { 795 int rv, sync, wasdelayed; 796 struct vnode *vp; 797 struct mount *mp; 798 799 KASSERT(ISSET(bp->b_cflags, BC_BUSY)); 800 KASSERT(!cv_has_waiters(&bp->b_done)); 801 802 vp = bp->b_vp; 803 804 /* 805 * dholland 20160728 AFAICT vp==NULL must be impossible as it 806 * will crash upon reaching VOP_STRATEGY below... see further 807 * analysis on tech-kern. 808 */ 809 KASSERTMSG(vp != NULL, "bwrite given buffer with null vnode"); 810 811 if (vp != NULL) { 812 KASSERT(bp->b_objlock == vp->v_interlock); 813 if (vp->v_type == VBLK) 814 mp = spec_node_getmountedfs(vp); 815 else 816 mp = vp->v_mount; 817 } else { 818 mp = NULL; 819 } 820 821 if (mp && mp->mnt_wapbl) { 822 if (bp->b_iodone != mp->mnt_wapbl_op->wo_wapbl_biodone) { 823 bdwrite(bp); 824 return 0; 825 } 826 } 827 828 /* 829 * Remember buffer type, to switch on it later. If the write was 830 * synchronous, but the file system was mounted with MNT_ASYNC, 831 * convert it to a delayed write. 832 * XXX note that this relies on delayed tape writes being converted 833 * to async, not sync writes (which is safe, but ugly). 834 */ 835 sync = !ISSET(bp->b_flags, B_ASYNC); 836 if (sync && mp != NULL && ISSET(mp->mnt_flag, MNT_ASYNC)) { 837 bdwrite(bp); 838 return (0); 839 } 840 841 /* 842 * Collect statistics on synchronous and asynchronous writes. 843 * Writes to block devices are charged to their associated 844 * filesystem (if any). 845 */ 846 if (mp != NULL) { 847 if (sync) 848 mp->mnt_stat.f_syncwrites++; 849 else 850 mp->mnt_stat.f_asyncwrites++; 851 } 852 853 /* 854 * Pay for the I/O operation and make sure the buf is on the correct 855 * vnode queue. 856 */ 857 bp->b_error = 0; 858 wasdelayed = ISSET(bp->b_oflags, BO_DELWRI); 859 CLR(bp->b_flags, B_READ); 860 if (wasdelayed) { 861 mutex_enter(&bufcache_lock); 862 mutex_enter(bp->b_objlock); 863 CLR(bp->b_oflags, BO_DONE | BO_DELWRI); 864 reassignbuf(bp, bp->b_vp); 865 mutex_exit(&bufcache_lock); 866 } else { 867 curlwp->l_ru.ru_oublock++; 868 mutex_enter(bp->b_objlock); 869 CLR(bp->b_oflags, BO_DONE | BO_DELWRI); 870 } 871 if (vp != NULL) 872 vp->v_numoutput++; 873 mutex_exit(bp->b_objlock); 874 875 /* Initiate disk write. */ 876 if (sync) 877 BIO_SETPRIO(bp, BPRIO_TIMECRITICAL); 878 else 879 BIO_SETPRIO(bp, BPRIO_TIMELIMITED); 880 881 VOP_STRATEGY(vp, bp); 882 883 if (sync) { 884 /* If I/O was synchronous, wait for it to complete. */ 885 rv = biowait(bp); 886 887 /* Release the buffer. */ 888 brelse(bp, 0); 889 890 return (rv); 891 } else { 892 return (0); 893 } 894 } 895 896 int 897 vn_bwrite(void *v) 898 { 899 struct vop_bwrite_args *ap = v; 900 901 return (bwrite(ap->a_bp)); 902 } 903 904 /* 905 * Delayed write. 906 * 907 * The buffer is marked dirty, but is not queued for I/O. 908 * This routine should be used when the buffer is expected 909 * to be modified again soon, typically a small write that 910 * partially fills a buffer. 911 * 912 * NB: magnetic tapes cannot be delayed; they must be 913 * written in the order that the writes are requested. 914 * 915 * Described in Leffler, et al. (pp. 208-213). 916 */ 917 void 918 bdwrite(buf_t *bp) 919 { 920 921 KASSERT(bp->b_vp == NULL || bp->b_vp->v_tag != VT_UFS || 922 bp->b_vp->v_type == VBLK || ISSET(bp->b_flags, B_COWDONE)); 923 KASSERT(ISSET(bp->b_cflags, BC_BUSY)); 924 KASSERT(!cv_has_waiters(&bp->b_done)); 925 926 /* If this is a tape block, write the block now. */ 927 if (bdev_type(bp->b_dev) == D_TAPE) { 928 bawrite(bp); 929 return; 930 } 931 932 if (wapbl_vphaswapbl(bp->b_vp)) { 933 struct mount *mp = wapbl_vptomp(bp->b_vp); 934 935 if (bp->b_iodone != mp->mnt_wapbl_op->wo_wapbl_biodone) { 936 WAPBL_ADD_BUF(mp, bp); 937 } 938 } 939 940 /* 941 * If the block hasn't been seen before: 942 * (1) Mark it as having been seen, 943 * (2) Charge for the write, 944 * (3) Make sure it's on its vnode's correct block list. 945 */ 946 KASSERT(bp->b_vp == NULL || bp->b_objlock == bp->b_vp->v_interlock); 947 948 if (!ISSET(bp->b_oflags, BO_DELWRI)) { 949 mutex_enter(&bufcache_lock); 950 mutex_enter(bp->b_objlock); 951 SET(bp->b_oflags, BO_DELWRI); 952 curlwp->l_ru.ru_oublock++; 953 reassignbuf(bp, bp->b_vp); 954 mutex_exit(&bufcache_lock); 955 } else { 956 mutex_enter(bp->b_objlock); 957 } 958 /* Otherwise, the "write" is done, so mark and release the buffer. */ 959 CLR(bp->b_oflags, BO_DONE); 960 mutex_exit(bp->b_objlock); 961 962 brelse(bp, 0); 963 } 964 965 /* 966 * Asynchronous block write; just an asynchronous bwrite(). 967 */ 968 void 969 bawrite(buf_t *bp) 970 { 971 972 KASSERT(ISSET(bp->b_cflags, BC_BUSY)); 973 KASSERT(bp->b_vp != NULL); 974 975 SET(bp->b_flags, B_ASYNC); 976 VOP_BWRITE(bp->b_vp, bp); 977 } 978 979 /* 980 * Release a buffer on to the free lists. 981 * Described in Bach (p. 46). 982 */ 983 void 984 brelsel(buf_t *bp, int set) 985 { 986 struct bqueue *bufq; 987 struct vnode *vp; 988 989 KASSERT(bp != NULL); 990 KASSERT(mutex_owned(&bufcache_lock)); 991 KASSERT(!cv_has_waiters(&bp->b_done)); 992 KASSERT(bp->b_refcnt > 0); 993 994 SET(bp->b_cflags, set); 995 996 KASSERT(ISSET(bp->b_cflags, BC_BUSY)); 997 KASSERT(bp->b_iodone == NULL); 998 999 /* Wake up any processes waiting for any buffer to become free. */ 1000 cv_signal(&needbuffer_cv); 1001 1002 /* Wake up any proceeses waiting for _this_ buffer to become */ 1003 if (ISSET(bp->b_cflags, BC_WANTED)) 1004 CLR(bp->b_cflags, BC_WANTED|BC_AGE); 1005 1006 /* If it's clean clear the copy-on-write flag. */ 1007 if (ISSET(bp->b_flags, B_COWDONE)) { 1008 mutex_enter(bp->b_objlock); 1009 if (!ISSET(bp->b_oflags, BO_DELWRI)) 1010 CLR(bp->b_flags, B_COWDONE); 1011 mutex_exit(bp->b_objlock); 1012 } 1013 1014 /* 1015 * Determine which queue the buffer should be on, then put it there. 1016 */ 1017 1018 /* If it's locked, don't report an error; try again later. */ 1019 if (ISSET(bp->b_flags, B_LOCKED)) 1020 bp->b_error = 0; 1021 1022 /* If it's not cacheable, or an error, mark it invalid. */ 1023 if (ISSET(bp->b_cflags, BC_NOCACHE) || bp->b_error != 0) 1024 SET(bp->b_cflags, BC_INVAL); 1025 1026 if (ISSET(bp->b_cflags, BC_VFLUSH)) { 1027 /* 1028 * This is a delayed write buffer that was just flushed to 1029 * disk. It is still on the LRU queue. If it's become 1030 * invalid, then we need to move it to a different queue; 1031 * otherwise leave it in its current position. 1032 */ 1033 CLR(bp->b_cflags, BC_VFLUSH); 1034 if (!ISSET(bp->b_cflags, BC_INVAL|BC_AGE) && 1035 !ISSET(bp->b_flags, B_LOCKED) && bp->b_error == 0) { 1036 KDASSERT(checkfreelist(bp, &bufqueues[BQ_LRU], 1)); 1037 goto already_queued; 1038 } else { 1039 bremfree(bp); 1040 } 1041 } 1042 1043 KDASSERT(checkfreelist(bp, &bufqueues[BQ_AGE], 0)); 1044 KDASSERT(checkfreelist(bp, &bufqueues[BQ_LRU], 0)); 1045 KDASSERT(checkfreelist(bp, &bufqueues[BQ_LOCKED], 0)); 1046 1047 if ((bp->b_bufsize <= 0) || ISSET(bp->b_cflags, BC_INVAL)) { 1048 /* 1049 * If it's invalid or empty, dissociate it from its vnode 1050 * and put on the head of the appropriate queue. 1051 */ 1052 if (ISSET(bp->b_flags, B_LOCKED)) { 1053 if (wapbl_vphaswapbl(vp = bp->b_vp)) { 1054 struct mount *mp = wapbl_vptomp(vp); 1055 1056 KASSERT(bp->b_iodone 1057 != mp->mnt_wapbl_op->wo_wapbl_biodone); 1058 WAPBL_REMOVE_BUF(mp, bp); 1059 } 1060 } 1061 1062 mutex_enter(bp->b_objlock); 1063 CLR(bp->b_oflags, BO_DONE|BO_DELWRI); 1064 if ((vp = bp->b_vp) != NULL) { 1065 KASSERT(bp->b_objlock == vp->v_interlock); 1066 reassignbuf(bp, bp->b_vp); 1067 brelvp(bp); 1068 mutex_exit(vp->v_interlock); 1069 } else { 1070 KASSERT(bp->b_objlock == &buffer_lock); 1071 mutex_exit(bp->b_objlock); 1072 } 1073 1074 if (bp->b_bufsize <= 0) 1075 /* no data */ 1076 goto already_queued; 1077 else 1078 /* invalid data */ 1079 bufq = &bufqueues[BQ_AGE]; 1080 binsheadfree(bp, bufq); 1081 } else { 1082 /* 1083 * It has valid data. Put it on the end of the appropriate 1084 * queue, so that it'll stick around for as long as possible. 1085 * If buf is AGE, but has dependencies, must put it on last 1086 * bufqueue to be scanned, ie LRU. This protects against the 1087 * livelock where BQ_AGE only has buffers with dependencies, 1088 * and we thus never get to the dependent buffers in BQ_LRU. 1089 */ 1090 if (ISSET(bp->b_flags, B_LOCKED)) { 1091 /* locked in core */ 1092 bufq = &bufqueues[BQ_LOCKED]; 1093 } else if (!ISSET(bp->b_cflags, BC_AGE)) { 1094 /* valid data */ 1095 bufq = &bufqueues[BQ_LRU]; 1096 } else { 1097 /* stale but valid data */ 1098 bufq = &bufqueues[BQ_AGE]; 1099 } 1100 binstailfree(bp, bufq); 1101 } 1102 already_queued: 1103 /* Unlock the buffer. */ 1104 CLR(bp->b_cflags, BC_AGE|BC_BUSY|BC_NOCACHE); 1105 CLR(bp->b_flags, B_ASYNC); 1106 cv_broadcast(&bp->b_busy); 1107 1108 if (bp->b_bufsize <= 0) 1109 brele(bp); 1110 } 1111 1112 void 1113 brelse(buf_t *bp, int set) 1114 { 1115 1116 mutex_enter(&bufcache_lock); 1117 brelsel(bp, set); 1118 mutex_exit(&bufcache_lock); 1119 } 1120 1121 /* 1122 * Determine if a block is in the cache. 1123 * Just look on what would be its hash chain. If it's there, return 1124 * a pointer to it, unless it's marked invalid. If it's marked invalid, 1125 * we normally don't return the buffer, unless the caller explicitly 1126 * wants us to. 1127 */ 1128 buf_t * 1129 incore(struct vnode *vp, daddr_t blkno) 1130 { 1131 buf_t *bp; 1132 1133 KASSERT(mutex_owned(&bufcache_lock)); 1134 1135 /* Search hash chain */ 1136 LIST_FOREACH(bp, BUFHASH(vp, blkno), b_hash) { 1137 if (bp->b_lblkno == blkno && bp->b_vp == vp && 1138 !ISSET(bp->b_cflags, BC_INVAL)) { 1139 KASSERT(bp->b_objlock == vp->v_interlock); 1140 return (bp); 1141 } 1142 } 1143 1144 return (NULL); 1145 } 1146 1147 /* 1148 * Get a block of requested size that is associated with 1149 * a given vnode and block offset. If it is found in the 1150 * block cache, mark it as having been found, make it busy 1151 * and return it. Otherwise, return an empty block of the 1152 * correct size. It is up to the caller to insure that the 1153 * cached blocks be of the correct size. 1154 */ 1155 buf_t * 1156 getblk(struct vnode *vp, daddr_t blkno, int size, int slpflag, int slptimeo) 1157 { 1158 int err, preserve; 1159 buf_t *bp; 1160 1161 mutex_enter(&bufcache_lock); 1162 loop: 1163 bp = incore(vp, blkno); 1164 if (bp != NULL) { 1165 err = bbusy(bp, ((slpflag & PCATCH) != 0), slptimeo, NULL); 1166 if (err != 0) { 1167 if (err == EPASSTHROUGH) 1168 goto loop; 1169 mutex_exit(&bufcache_lock); 1170 return (NULL); 1171 } 1172 KASSERT(!cv_has_waiters(&bp->b_done)); 1173 #ifdef DIAGNOSTIC 1174 if (ISSET(bp->b_oflags, BO_DONE|BO_DELWRI) && 1175 bp->b_bcount < size && vp->v_type != VBLK) 1176 panic("getblk: block size invariant failed"); 1177 #endif 1178 bremfree(bp); 1179 preserve = 1; 1180 } else { 1181 if ((bp = getnewbuf(slpflag, slptimeo, 0)) == NULL) 1182 goto loop; 1183 1184 if (incore(vp, blkno) != NULL) { 1185 /* The block has come into memory in the meantime. */ 1186 brelsel(bp, 0); 1187 goto loop; 1188 } 1189 1190 LIST_INSERT_HEAD(BUFHASH(vp, blkno), bp, b_hash); 1191 bp->b_blkno = bp->b_lblkno = bp->b_rawblkno = blkno; 1192 mutex_enter(vp->v_interlock); 1193 bgetvp(vp, bp); 1194 mutex_exit(vp->v_interlock); 1195 preserve = 0; 1196 } 1197 mutex_exit(&bufcache_lock); 1198 1199 /* 1200 * LFS can't track total size of B_LOCKED buffer (locked_queue_bytes) 1201 * if we re-size buffers here. 1202 */ 1203 if (ISSET(bp->b_flags, B_LOCKED)) { 1204 KASSERT(bp->b_bufsize >= size); 1205 } else { 1206 if (allocbuf(bp, size, preserve)) { 1207 mutex_enter(&bufcache_lock); 1208 LIST_REMOVE(bp, b_hash); 1209 mutex_exit(&bufcache_lock); 1210 brelse(bp, BC_INVAL); 1211 return NULL; 1212 } 1213 } 1214 BIO_SETPRIO(bp, BPRIO_DEFAULT); 1215 return (bp); 1216 } 1217 1218 /* 1219 * Get an empty, disassociated buffer of given size. 1220 */ 1221 buf_t * 1222 geteblk(int size) 1223 { 1224 buf_t *bp; 1225 int error __diagused; 1226 1227 mutex_enter(&bufcache_lock); 1228 while ((bp = getnewbuf(0, 0, 0)) == NULL) 1229 ; 1230 1231 SET(bp->b_cflags, BC_INVAL); 1232 LIST_INSERT_HEAD(&invalhash, bp, b_hash); 1233 mutex_exit(&bufcache_lock); 1234 BIO_SETPRIO(bp, BPRIO_DEFAULT); 1235 error = allocbuf(bp, size, 0); 1236 KASSERT(error == 0); 1237 return (bp); 1238 } 1239 1240 /* 1241 * Expand or contract the actual memory allocated to a buffer. 1242 * 1243 * If the buffer shrinks, data is lost, so it's up to the 1244 * caller to have written it out *first*; this routine will not 1245 * start a write. If the buffer grows, it's the callers 1246 * responsibility to fill out the buffer's additional contents. 1247 */ 1248 int 1249 allocbuf(buf_t *bp, int size, int preserve) 1250 { 1251 void *addr; 1252 vsize_t oldsize, desired_size; 1253 int oldcount; 1254 int delta; 1255 1256 desired_size = buf_roundsize(size); 1257 if (desired_size > MAXBSIZE) 1258 printf("allocbuf: buffer larger than MAXBSIZE requested"); 1259 1260 oldcount = bp->b_bcount; 1261 1262 bp->b_bcount = size; 1263 1264 oldsize = bp->b_bufsize; 1265 if (oldsize == desired_size) { 1266 /* 1267 * Do not short cut the WAPBL resize, as the buffer length 1268 * could still have changed and this would corrupt the 1269 * tracking of the transaction length. 1270 */ 1271 goto out; 1272 } 1273 1274 /* 1275 * If we want a buffer of a different size, re-allocate the 1276 * buffer's memory; copy old content only if needed. 1277 */ 1278 addr = buf_alloc(desired_size); 1279 if (addr == NULL) 1280 return ENOMEM; 1281 if (preserve) 1282 memcpy(addr, bp->b_data, MIN(oldsize,desired_size)); 1283 if (bp->b_data != NULL) 1284 buf_mrelease(bp->b_data, oldsize); 1285 bp->b_data = addr; 1286 bp->b_bufsize = desired_size; 1287 1288 /* 1289 * Update overall buffer memory counter (protected by bufcache_lock) 1290 */ 1291 delta = (long)desired_size - (long)oldsize; 1292 1293 mutex_enter(&bufcache_lock); 1294 if ((bufmem += delta) > bufmem_hiwater) { 1295 /* 1296 * Need to trim overall memory usage. 1297 */ 1298 while (buf_canrelease()) { 1299 if (curcpu()->ci_schedstate.spc_flags & 1300 SPCF_SHOULDYIELD) { 1301 mutex_exit(&bufcache_lock); 1302 preempt(); 1303 mutex_enter(&bufcache_lock); 1304 } 1305 if (buf_trim() == 0) 1306 break; 1307 } 1308 } 1309 mutex_exit(&bufcache_lock); 1310 1311 out: 1312 if (wapbl_vphaswapbl(bp->b_vp)) 1313 WAPBL_RESIZE_BUF(wapbl_vptomp(bp->b_vp), bp, oldsize, oldcount); 1314 1315 return 0; 1316 } 1317 1318 /* 1319 * Find a buffer which is available for use. 1320 * Select something from a free list. 1321 * Preference is to AGE list, then LRU list. 1322 * 1323 * Called with the buffer queues locked. 1324 * Return buffer locked. 1325 */ 1326 buf_t * 1327 getnewbuf(int slpflag, int slptimeo, int from_bufq) 1328 { 1329 buf_t *bp; 1330 struct vnode *vp; 1331 1332 start: 1333 KASSERT(mutex_owned(&bufcache_lock)); 1334 1335 /* 1336 * Get a new buffer from the pool. 1337 */ 1338 if (!from_bufq && buf_lotsfree()) { 1339 mutex_exit(&bufcache_lock); 1340 bp = pool_cache_get(buf_cache, PR_NOWAIT); 1341 if (bp != NULL) { 1342 memset((char *)bp, 0, sizeof(*bp)); 1343 buf_init(bp); 1344 SET(bp->b_cflags, BC_BUSY); /* mark buffer busy */ 1345 mutex_enter(&bufcache_lock); 1346 #if defined(DIAGNOSTIC) 1347 bp->b_freelistindex = -1; 1348 #endif /* defined(DIAGNOSTIC) */ 1349 return (bp); 1350 } 1351 mutex_enter(&bufcache_lock); 1352 } 1353 1354 KASSERT(mutex_owned(&bufcache_lock)); 1355 if ((bp = TAILQ_FIRST(&bufqueues[BQ_AGE].bq_queue)) != NULL || 1356 (bp = TAILQ_FIRST(&bufqueues[BQ_LRU].bq_queue)) != NULL) { 1357 KASSERT(!ISSET(bp->b_cflags, BC_BUSY) || ISSET(bp->b_cflags, BC_VFLUSH)); 1358 bremfree(bp); 1359 1360 /* Buffer is no longer on free lists. */ 1361 SET(bp->b_cflags, BC_BUSY); 1362 } else { 1363 /* 1364 * XXX: !from_bufq should be removed. 1365 */ 1366 if (!from_bufq || curlwp != uvm.pagedaemon_lwp) { 1367 /* wait for a free buffer of any kind */ 1368 if ((slpflag & PCATCH) != 0) 1369 (void)cv_timedwait_sig(&needbuffer_cv, 1370 &bufcache_lock, slptimeo); 1371 else 1372 (void)cv_timedwait(&needbuffer_cv, 1373 &bufcache_lock, slptimeo); 1374 } 1375 return (NULL); 1376 } 1377 1378 #ifdef DIAGNOSTIC 1379 if (bp->b_bufsize <= 0) 1380 panic("buffer %p: on queue but empty", bp); 1381 #endif 1382 1383 if (ISSET(bp->b_cflags, BC_VFLUSH)) { 1384 /* 1385 * This is a delayed write buffer being flushed to disk. Make 1386 * sure it gets aged out of the queue when it's finished, and 1387 * leave it off the LRU queue. 1388 */ 1389 CLR(bp->b_cflags, BC_VFLUSH); 1390 SET(bp->b_cflags, BC_AGE); 1391 goto start; 1392 } 1393 1394 KASSERT(ISSET(bp->b_cflags, BC_BUSY)); 1395 KASSERT(bp->b_refcnt > 0); 1396 KASSERT(!cv_has_waiters(&bp->b_done)); 1397 1398 /* 1399 * If buffer was a delayed write, start it and return NULL 1400 * (since we might sleep while starting the write). 1401 */ 1402 if (ISSET(bp->b_oflags, BO_DELWRI)) { 1403 /* 1404 * This buffer has gone through the LRU, so make sure it gets 1405 * reused ASAP. 1406 */ 1407 SET(bp->b_cflags, BC_AGE); 1408 mutex_exit(&bufcache_lock); 1409 bawrite(bp); 1410 mutex_enter(&bufcache_lock); 1411 return (NULL); 1412 } 1413 1414 vp = bp->b_vp; 1415 1416 /* clear out various other fields */ 1417 bp->b_cflags = BC_BUSY; 1418 bp->b_oflags = 0; 1419 bp->b_flags = 0; 1420 bp->b_dev = NODEV; 1421 bp->b_blkno = 0; 1422 bp->b_lblkno = 0; 1423 bp->b_rawblkno = 0; 1424 bp->b_iodone = 0; 1425 bp->b_error = 0; 1426 bp->b_resid = 0; 1427 bp->b_bcount = 0; 1428 1429 LIST_REMOVE(bp, b_hash); 1430 1431 /* Disassociate us from our vnode, if we had one... */ 1432 if (vp != NULL) { 1433 mutex_enter(vp->v_interlock); 1434 brelvp(bp); 1435 mutex_exit(vp->v_interlock); 1436 } 1437 1438 return (bp); 1439 } 1440 1441 /* 1442 * Attempt to free an aged buffer off the queues. 1443 * Called with queue lock held. 1444 * Returns the amount of buffer memory freed. 1445 */ 1446 static int 1447 buf_trim(void) 1448 { 1449 buf_t *bp; 1450 long size; 1451 1452 KASSERT(mutex_owned(&bufcache_lock)); 1453 1454 /* Instruct getnewbuf() to get buffers off the queues */ 1455 if ((bp = getnewbuf(PCATCH, 1, 1)) == NULL) 1456 return 0; 1457 1458 KASSERT((bp->b_cflags & BC_WANTED) == 0); 1459 size = bp->b_bufsize; 1460 bufmem -= size; 1461 if (size > 0) { 1462 buf_mrelease(bp->b_data, size); 1463 bp->b_bcount = bp->b_bufsize = 0; 1464 } 1465 /* brelse() will return the buffer to the global buffer pool */ 1466 brelsel(bp, 0); 1467 return size; 1468 } 1469 1470 int 1471 buf_drain(int n) 1472 { 1473 int size = 0, sz; 1474 1475 KASSERT(mutex_owned(&bufcache_lock)); 1476 1477 while (size < n && bufmem > bufmem_lowater) { 1478 sz = buf_trim(); 1479 if (sz <= 0) 1480 break; 1481 size += sz; 1482 } 1483 1484 return size; 1485 } 1486 1487 SDT_PROVIDER_DEFINE(io); 1488 1489 SDT_PROBE_DEFINE1(io, kernel, , wait__start, "struct buf *"/*bp*/); 1490 SDT_PROBE_DEFINE1(io, kernel, , wait__done, "struct buf *"/*bp*/); 1491 1492 /* 1493 * Wait for operations on the buffer to complete. 1494 * When they do, extract and return the I/O's error value. 1495 */ 1496 int 1497 biowait(buf_t *bp) 1498 { 1499 1500 KASSERT(ISSET(bp->b_cflags, BC_BUSY)); 1501 KASSERT(bp->b_refcnt > 0); 1502 1503 SDT_PROBE1(io, kernel, , wait__start, bp); 1504 1505 mutex_enter(bp->b_objlock); 1506 while (!ISSET(bp->b_oflags, BO_DONE | BO_DELWRI)) 1507 cv_wait(&bp->b_done, bp->b_objlock); 1508 mutex_exit(bp->b_objlock); 1509 1510 SDT_PROBE1(io, kernel, , wait__done, bp); 1511 1512 return bp->b_error; 1513 } 1514 1515 /* 1516 * Mark I/O complete on a buffer. 1517 * 1518 * If a callback has been requested, e.g. the pageout 1519 * daemon, do so. Otherwise, awaken waiting processes. 1520 * 1521 * [ Leffler, et al., says on p.247: 1522 * "This routine wakes up the blocked process, frees the buffer 1523 * for an asynchronous write, or, for a request by the pagedaemon 1524 * process, invokes a procedure specified in the buffer structure" ] 1525 * 1526 * In real life, the pagedaemon (or other system processes) wants 1527 * to do async stuff to, and doesn't want the buffer brelse()'d. 1528 * (for swap pager, that puts swap buffers on the free lists (!!!), 1529 * for the vn device, that puts allocated buffers on the free lists!) 1530 */ 1531 void 1532 biodone(buf_t *bp) 1533 { 1534 int s; 1535 1536 KASSERT(!ISSET(bp->b_oflags, BO_DONE)); 1537 1538 if (cpu_intr_p()) { 1539 /* From interrupt mode: defer to a soft interrupt. */ 1540 s = splvm(); 1541 TAILQ_INSERT_TAIL(&curcpu()->ci_data.cpu_biodone, bp, b_actq); 1542 softint_schedule(biodone_sih); 1543 splx(s); 1544 } else { 1545 /* Process now - the buffer may be freed soon. */ 1546 biodone2(bp); 1547 } 1548 } 1549 1550 SDT_PROBE_DEFINE1(io, kernel, , done, "struct buf *"/*bp*/); 1551 1552 static void 1553 biodone2(buf_t *bp) 1554 { 1555 void (*callout)(buf_t *); 1556 1557 SDT_PROBE1(io, kernel, ,done, bp); 1558 1559 mutex_enter(bp->b_objlock); 1560 /* Note that the transfer is done. */ 1561 if (ISSET(bp->b_oflags, BO_DONE)) 1562 panic("biodone2 already"); 1563 CLR(bp->b_flags, B_COWDONE); 1564 SET(bp->b_oflags, BO_DONE); 1565 BIO_SETPRIO(bp, BPRIO_DEFAULT); 1566 1567 /* Wake up waiting writers. */ 1568 if (!ISSET(bp->b_flags, B_READ)) 1569 vwakeup(bp); 1570 1571 if ((callout = bp->b_iodone) != NULL) { 1572 /* Note callout done, then call out. */ 1573 KASSERT(!cv_has_waiters(&bp->b_done)); 1574 KERNEL_LOCK(1, NULL); /* XXXSMP */ 1575 bp->b_iodone = NULL; 1576 mutex_exit(bp->b_objlock); 1577 (*callout)(bp); 1578 KERNEL_UNLOCK_ONE(NULL); /* XXXSMP */ 1579 } else if (ISSET(bp->b_flags, B_ASYNC)) { 1580 /* If async, release. */ 1581 KASSERT(!cv_has_waiters(&bp->b_done)); 1582 mutex_exit(bp->b_objlock); 1583 brelse(bp, 0); 1584 } else { 1585 /* Otherwise just wake up waiters in biowait(). */ 1586 cv_broadcast(&bp->b_done); 1587 mutex_exit(bp->b_objlock); 1588 } 1589 } 1590 1591 static void 1592 biointr(void *cookie) 1593 { 1594 struct cpu_info *ci; 1595 buf_t *bp; 1596 int s; 1597 1598 ci = curcpu(); 1599 1600 while (!TAILQ_EMPTY(&ci->ci_data.cpu_biodone)) { 1601 KASSERT(curcpu() == ci); 1602 1603 s = splvm(); 1604 bp = TAILQ_FIRST(&ci->ci_data.cpu_biodone); 1605 TAILQ_REMOVE(&ci->ci_data.cpu_biodone, bp, b_actq); 1606 splx(s); 1607 1608 biodone2(bp); 1609 } 1610 } 1611 1612 /* 1613 * Wait for all buffers to complete I/O 1614 * Return the number of "stuck" buffers. 1615 */ 1616 int 1617 buf_syncwait(void) 1618 { 1619 buf_t *bp; 1620 int iter, nbusy, nbusy_prev = 0, ihash; 1621 1622 for (iter = 0; iter < 20;) { 1623 mutex_enter(&bufcache_lock); 1624 nbusy = 0; 1625 for (ihash = 0; ihash < bufhash+1; ihash++) { 1626 LIST_FOREACH(bp, &bufhashtbl[ihash], b_hash) { 1627 if ((bp->b_cflags & (BC_BUSY|BC_INVAL)) == BC_BUSY) 1628 nbusy += ((bp->b_flags & B_READ) == 0); 1629 } 1630 } 1631 mutex_exit(&bufcache_lock); 1632 1633 if (nbusy == 0) 1634 break; 1635 if (nbusy_prev == 0) 1636 nbusy_prev = nbusy; 1637 printf("%d ", nbusy); 1638 kpause("bflush", false, MAX(1, hz / 25 * iter), NULL); 1639 if (nbusy >= nbusy_prev) /* we didn't flush anything */ 1640 iter++; 1641 else 1642 nbusy_prev = nbusy; 1643 } 1644 1645 if (nbusy) { 1646 #if defined(DEBUG) || defined(DEBUG_HALT_BUSY) 1647 printf("giving up\nPrinting vnodes for busy buffers\n"); 1648 for (ihash = 0; ihash < bufhash+1; ihash++) { 1649 LIST_FOREACH(bp, &bufhashtbl[ihash], b_hash) { 1650 if ((bp->b_cflags & (BC_BUSY|BC_INVAL)) == BC_BUSY && 1651 (bp->b_flags & B_READ) == 0) 1652 vprint(NULL, bp->b_vp); 1653 } 1654 } 1655 #endif 1656 } 1657 1658 return nbusy; 1659 } 1660 1661 static void 1662 sysctl_fillbuf(buf_t *i, struct buf_sysctl *o) 1663 { 1664 1665 o->b_flags = i->b_flags | i->b_cflags | i->b_oflags; 1666 o->b_error = i->b_error; 1667 o->b_prio = i->b_prio; 1668 o->b_dev = i->b_dev; 1669 o->b_bufsize = i->b_bufsize; 1670 o->b_bcount = i->b_bcount; 1671 o->b_resid = i->b_resid; 1672 o->b_addr = PTRTOUINT64(i->b_data); 1673 o->b_blkno = i->b_blkno; 1674 o->b_rawblkno = i->b_rawblkno; 1675 o->b_iodone = PTRTOUINT64(i->b_iodone); 1676 o->b_proc = PTRTOUINT64(i->b_proc); 1677 o->b_vp = PTRTOUINT64(i->b_vp); 1678 o->b_saveaddr = PTRTOUINT64(i->b_saveaddr); 1679 o->b_lblkno = i->b_lblkno; 1680 } 1681 1682 #define KERN_BUFSLOP 20 1683 static int 1684 sysctl_dobuf(SYSCTLFN_ARGS) 1685 { 1686 buf_t *bp; 1687 struct buf_sysctl bs; 1688 struct bqueue *bq; 1689 char *dp; 1690 u_int i, op, arg; 1691 size_t len, needed, elem_size, out_size; 1692 int error, elem_count, retries; 1693 1694 if (namelen == 1 && name[0] == CTL_QUERY) 1695 return (sysctl_query(SYSCTLFN_CALL(rnode))); 1696 1697 if (namelen != 4) 1698 return (EINVAL); 1699 1700 retries = 100; 1701 retry: 1702 dp = oldp; 1703 len = (oldp != NULL) ? *oldlenp : 0; 1704 op = name[0]; 1705 arg = name[1]; 1706 elem_size = name[2]; 1707 elem_count = name[3]; 1708 out_size = MIN(sizeof(bs), elem_size); 1709 1710 /* 1711 * at the moment, these are just "placeholders" to make the 1712 * API for retrieving kern.buf data more extensible in the 1713 * future. 1714 * 1715 * XXX kern.buf currently has "netbsd32" issues. hopefully 1716 * these will be resolved at a later point. 1717 */ 1718 if (op != KERN_BUF_ALL || arg != KERN_BUF_ALL || 1719 elem_size < 1 || elem_count < 0) 1720 return (EINVAL); 1721 1722 error = 0; 1723 needed = 0; 1724 sysctl_unlock(); 1725 mutex_enter(&bufcache_lock); 1726 for (i = 0; i < BQUEUES; i++) { 1727 bq = &bufqueues[i]; 1728 TAILQ_FOREACH(bp, &bq->bq_queue, b_freelist) { 1729 bq->bq_marker = bp; 1730 if (len >= elem_size && elem_count > 0) { 1731 sysctl_fillbuf(bp, &bs); 1732 mutex_exit(&bufcache_lock); 1733 error = copyout(&bs, dp, out_size); 1734 mutex_enter(&bufcache_lock); 1735 if (error) 1736 break; 1737 if (bq->bq_marker != bp) { 1738 /* 1739 * This sysctl node is only for 1740 * statistics. Retry; if the 1741 * queue keeps changing, then 1742 * bail out. 1743 */ 1744 if (retries-- == 0) { 1745 error = EAGAIN; 1746 break; 1747 } 1748 mutex_exit(&bufcache_lock); 1749 sysctl_relock(); 1750 goto retry; 1751 } 1752 dp += elem_size; 1753 len -= elem_size; 1754 } 1755 needed += elem_size; 1756 if (elem_count > 0 && elem_count != INT_MAX) 1757 elem_count--; 1758 } 1759 if (error != 0) 1760 break; 1761 } 1762 mutex_exit(&bufcache_lock); 1763 sysctl_relock(); 1764 1765 *oldlenp = needed; 1766 if (oldp == NULL) 1767 *oldlenp += KERN_BUFSLOP * sizeof(buf_t); 1768 1769 return (error); 1770 } 1771 1772 static int 1773 sysctl_bufvm_update(SYSCTLFN_ARGS) 1774 { 1775 int error, rv; 1776 struct sysctlnode node; 1777 unsigned int temp_bufcache; 1778 unsigned long temp_water; 1779 1780 /* Take a copy of the supplied node and its data */ 1781 node = *rnode; 1782 if (node.sysctl_data == &bufcache) { 1783 node.sysctl_data = &temp_bufcache; 1784 temp_bufcache = *(unsigned int *)rnode->sysctl_data; 1785 } else { 1786 node.sysctl_data = &temp_water; 1787 temp_water = *(unsigned long *)rnode->sysctl_data; 1788 } 1789 1790 /* Update the copy */ 1791 error = sysctl_lookup(SYSCTLFN_CALL(&node)); 1792 if (error || newp == NULL) 1793 return (error); 1794 1795 if (rnode->sysctl_data == &bufcache) { 1796 if (temp_bufcache > 100) 1797 return (EINVAL); 1798 bufcache = temp_bufcache; 1799 buf_setwm(); 1800 } else if (rnode->sysctl_data == &bufmem_lowater) { 1801 if (bufmem_hiwater - temp_water < 16) 1802 return (EINVAL); 1803 bufmem_lowater = temp_water; 1804 } else if (rnode->sysctl_data == &bufmem_hiwater) { 1805 if (temp_water - bufmem_lowater < 16) 1806 return (EINVAL); 1807 bufmem_hiwater = temp_water; 1808 } else 1809 return (EINVAL); 1810 1811 /* Drain until below new high water mark */ 1812 sysctl_unlock(); 1813 mutex_enter(&bufcache_lock); 1814 while (bufmem > bufmem_hiwater) { 1815 rv = buf_drain((bufmem - bufmem_hiwater) / (2 * 1024)); 1816 if (rv <= 0) 1817 break; 1818 } 1819 mutex_exit(&bufcache_lock); 1820 sysctl_relock(); 1821 1822 return 0; 1823 } 1824 1825 static struct sysctllog *vfsbio_sysctllog; 1826 1827 static void 1828 sysctl_kern_buf_setup(void) 1829 { 1830 1831 sysctl_createv(&vfsbio_sysctllog, 0, NULL, NULL, 1832 CTLFLAG_PERMANENT, 1833 CTLTYPE_NODE, "buf", 1834 SYSCTL_DESCR("Kernel buffer cache information"), 1835 sysctl_dobuf, 0, NULL, 0, 1836 CTL_KERN, KERN_BUF, CTL_EOL); 1837 } 1838 1839 static void 1840 sysctl_vm_buf_setup(void) 1841 { 1842 1843 sysctl_createv(&vfsbio_sysctllog, 0, NULL, NULL, 1844 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 1845 CTLTYPE_INT, "bufcache", 1846 SYSCTL_DESCR("Percentage of physical memory to use for " 1847 "buffer cache"), 1848 sysctl_bufvm_update, 0, &bufcache, 0, 1849 CTL_VM, CTL_CREATE, CTL_EOL); 1850 sysctl_createv(&vfsbio_sysctllog, 0, NULL, NULL, 1851 CTLFLAG_PERMANENT|CTLFLAG_READONLY, 1852 CTLTYPE_LONG, "bufmem", 1853 SYSCTL_DESCR("Amount of kernel memory used by buffer " 1854 "cache"), 1855 NULL, 0, &bufmem, 0, 1856 CTL_VM, CTL_CREATE, CTL_EOL); 1857 sysctl_createv(&vfsbio_sysctllog, 0, NULL, NULL, 1858 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 1859 CTLTYPE_LONG, "bufmem_lowater", 1860 SYSCTL_DESCR("Minimum amount of kernel memory to " 1861 "reserve for buffer cache"), 1862 sysctl_bufvm_update, 0, &bufmem_lowater, 0, 1863 CTL_VM, CTL_CREATE, CTL_EOL); 1864 sysctl_createv(&vfsbio_sysctllog, 0, NULL, NULL, 1865 CTLFLAG_PERMANENT|CTLFLAG_READWRITE, 1866 CTLTYPE_LONG, "bufmem_hiwater", 1867 SYSCTL_DESCR("Maximum amount of kernel memory to use " 1868 "for buffer cache"), 1869 sysctl_bufvm_update, 0, &bufmem_hiwater, 0, 1870 CTL_VM, CTL_CREATE, CTL_EOL); 1871 } 1872 1873 #ifdef DEBUG 1874 /* 1875 * Print out statistics on the current allocation of the buffer pool. 1876 * Can be enabled to print out on every ``sync'' by setting "syncprt" 1877 * in vfs_syscalls.c using sysctl. 1878 */ 1879 void 1880 vfs_bufstats(void) 1881 { 1882 int i, j, count; 1883 buf_t *bp; 1884 struct bqueue *dp; 1885 int counts[(MAXBSIZE / PAGE_SIZE) + 1]; 1886 static const char *bname[BQUEUES] = { "LOCKED", "LRU", "AGE" }; 1887 1888 for (dp = bufqueues, i = 0; dp < &bufqueues[BQUEUES]; dp++, i++) { 1889 count = 0; 1890 for (j = 0; j <= MAXBSIZE/PAGE_SIZE; j++) 1891 counts[j] = 0; 1892 TAILQ_FOREACH(bp, &dp->bq_queue, b_freelist) { 1893 counts[bp->b_bufsize/PAGE_SIZE]++; 1894 count++; 1895 } 1896 printf("%s: total-%d", bname[i], count); 1897 for (j = 0; j <= MAXBSIZE/PAGE_SIZE; j++) 1898 if (counts[j] != 0) 1899 printf(", %d-%d", j * PAGE_SIZE, counts[j]); 1900 printf("\n"); 1901 } 1902 } 1903 #endif /* DEBUG */ 1904 1905 /* ------------------------------ */ 1906 1907 buf_t * 1908 getiobuf(struct vnode *vp, bool waitok) 1909 { 1910 buf_t *bp; 1911 1912 bp = pool_cache_get(bufio_cache, (waitok ? PR_WAITOK : PR_NOWAIT)); 1913 if (bp == NULL) 1914 return bp; 1915 1916 buf_init(bp); 1917 1918 if ((bp->b_vp = vp) == NULL) 1919 bp->b_objlock = &buffer_lock; 1920 else 1921 bp->b_objlock = vp->v_interlock; 1922 1923 return bp; 1924 } 1925 1926 void 1927 putiobuf(buf_t *bp) 1928 { 1929 1930 buf_destroy(bp); 1931 pool_cache_put(bufio_cache, bp); 1932 } 1933 1934 /* 1935 * nestiobuf_iodone: b_iodone callback for nested buffers. 1936 */ 1937 1938 void 1939 nestiobuf_iodone(buf_t *bp) 1940 { 1941 buf_t *mbp = bp->b_private; 1942 int error; 1943 int donebytes; 1944 1945 KASSERT(bp->b_bcount <= bp->b_bufsize); 1946 KASSERT(mbp != bp); 1947 1948 error = bp->b_error; 1949 if (bp->b_error == 0 && 1950 (bp->b_bcount < bp->b_bufsize || bp->b_resid > 0)) { 1951 /* 1952 * Not all got transfered, raise an error. We have no way to 1953 * propagate these conditions to mbp. 1954 */ 1955 error = EIO; 1956 } 1957 1958 donebytes = bp->b_bufsize; 1959 1960 putiobuf(bp); 1961 nestiobuf_done(mbp, donebytes, error); 1962 } 1963 1964 /* 1965 * nestiobuf_setup: setup a "nested" buffer. 1966 * 1967 * => 'mbp' is a "master" buffer which is being divided into sub pieces. 1968 * => 'bp' should be a buffer allocated by getiobuf. 1969 * => 'offset' is a byte offset in the master buffer. 1970 * => 'size' is a size in bytes of this nested buffer. 1971 */ 1972 1973 void 1974 nestiobuf_setup(buf_t *mbp, buf_t *bp, int offset, size_t size) 1975 { 1976 const int b_read = mbp->b_flags & B_READ; 1977 struct vnode *vp = mbp->b_vp; 1978 1979 KASSERT(mbp->b_bcount >= offset + size); 1980 bp->b_vp = vp; 1981 bp->b_dev = mbp->b_dev; 1982 bp->b_objlock = mbp->b_objlock; 1983 bp->b_cflags = BC_BUSY; 1984 bp->b_flags = B_ASYNC | b_read; 1985 bp->b_iodone = nestiobuf_iodone; 1986 bp->b_data = (char *)mbp->b_data + offset; 1987 bp->b_resid = bp->b_bcount = size; 1988 bp->b_bufsize = bp->b_bcount; 1989 bp->b_private = mbp; 1990 BIO_COPYPRIO(bp, mbp); 1991 if (!b_read && vp != NULL) { 1992 mutex_enter(vp->v_interlock); 1993 vp->v_numoutput++; 1994 mutex_exit(vp->v_interlock); 1995 } 1996 } 1997 1998 /* 1999 * nestiobuf_done: propagate completion to the master buffer. 2000 * 2001 * => 'donebytes' specifies how many bytes in the 'mbp' is completed. 2002 * => 'error' is an errno(2) that 'donebytes' has been completed with. 2003 */ 2004 2005 void 2006 nestiobuf_done(buf_t *mbp, int donebytes, int error) 2007 { 2008 2009 if (donebytes == 0) { 2010 return; 2011 } 2012 mutex_enter(mbp->b_objlock); 2013 KASSERT(mbp->b_resid >= donebytes); 2014 mbp->b_resid -= donebytes; 2015 if (error) 2016 mbp->b_error = error; 2017 if (mbp->b_resid == 0) { 2018 if (mbp->b_error) 2019 mbp->b_resid = mbp->b_bcount; 2020 mutex_exit(mbp->b_objlock); 2021 biodone(mbp); 2022 } else 2023 mutex_exit(mbp->b_objlock); 2024 } 2025 2026 void 2027 buf_init(buf_t *bp) 2028 { 2029 2030 cv_init(&bp->b_busy, "biolock"); 2031 cv_init(&bp->b_done, "biowait"); 2032 bp->b_dev = NODEV; 2033 bp->b_error = 0; 2034 bp->b_flags = 0; 2035 bp->b_cflags = 0; 2036 bp->b_oflags = 0; 2037 bp->b_objlock = &buffer_lock; 2038 bp->b_iodone = NULL; 2039 bp->b_refcnt = 1; 2040 bp->b_dev = NODEV; 2041 bp->b_vnbufs.le_next = NOLIST; 2042 BIO_SETPRIO(bp, BPRIO_DEFAULT); 2043 } 2044 2045 void 2046 buf_destroy(buf_t *bp) 2047 { 2048 2049 cv_destroy(&bp->b_done); 2050 cv_destroy(&bp->b_busy); 2051 } 2052 2053 int 2054 bbusy(buf_t *bp, bool intr, int timo, kmutex_t *interlock) 2055 { 2056 int error; 2057 2058 KASSERT(mutex_owned(&bufcache_lock)); 2059 2060 if ((bp->b_cflags & BC_BUSY) != 0) { 2061 if (curlwp == uvm.pagedaemon_lwp) 2062 return EDEADLK; 2063 bp->b_cflags |= BC_WANTED; 2064 bref(bp); 2065 if (interlock != NULL) 2066 mutex_exit(interlock); 2067 if (intr) { 2068 error = cv_timedwait_sig(&bp->b_busy, &bufcache_lock, 2069 timo); 2070 } else { 2071 error = cv_timedwait(&bp->b_busy, &bufcache_lock, 2072 timo); 2073 } 2074 brele(bp); 2075 if (interlock != NULL) 2076 mutex_enter(interlock); 2077 if (error != 0) 2078 return error; 2079 return EPASSTHROUGH; 2080 } 2081 bp->b_cflags |= BC_BUSY; 2082 2083 return 0; 2084 } 2085