1 /*- 2 * Copyright (c) 2005, 2006 The NetBSD Foundation, Inc. 3 * All rights reserved. 4 * 5 * This code is derived from software contributed to The NetBSD Foundation 6 * by Julio M. Merino Vidal, developed as part of Google's Summer of Code 7 * 2005 program. 8 * 9 * Redistribution and use in source and binary forms, with or without 10 * modification, are permitted provided that the following conditions 11 * are met: 12 * 1. Redistributions of source code must retain the above copyright 13 * notice, this list of conditions and the following disclaimer. 14 * 2. Redistributions in binary form must reproduce the above copyright 15 * notice, this list of conditions and the following disclaimer in the 16 * documentation and/or other materials provided with the distribution. 17 * 18 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS 19 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 20 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 21 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS 22 * BE 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 * $NetBSD: tmpfs_vnops.c,v 1.39 2007/07/23 15:41:01 jmmv Exp $ 31 */ 32 33 /* 34 * tmpfs vnode interface. 35 */ 36 37 #include <sys/kernel.h> 38 #include <sys/kern_syscall.h> 39 #include <sys/param.h> 40 #include <sys/uio.h> 41 #include <sys/fcntl.h> 42 #include <sys/lockf.h> 43 #include <sys/priv.h> 44 #include <sys/proc.h> 45 #include <sys/resourcevar.h> 46 #include <sys/sched.h> 47 #include <sys/stat.h> 48 #include <sys/systm.h> 49 #include <sys/sysctl.h> 50 #include <sys/unistd.h> 51 #include <sys/vfsops.h> 52 #include <sys/vnode.h> 53 #include <sys/mountctl.h> 54 55 #include <vm/vm.h> 56 #include <vm/vm_extern.h> 57 #include <vm/vm_object.h> 58 #include <vm/vm_page.h> 59 #include <vm/vm_pageout.h> 60 #include <vm/vm_pager.h> 61 #include <vm/swap_pager.h> 62 63 #include <sys/buf2.h> 64 #include <vm/vm_page2.h> 65 66 #include <vfs/fifofs/fifo.h> 67 #include <vfs/tmpfs/tmpfs_vnops.h> 68 #include "tmpfs.h" 69 70 static void tmpfs_strategy_done(struct bio *bio); 71 static void tmpfs_move_pages(vm_object_t src, vm_object_t dst, int movflags); 72 73 /* 74 * bufcache_mode: 75 * 0 Normal page queue operation on flush. Run through the buffer 76 * cache if free memory is under the minimum. 77 * 78 * 1 Try to keep in memory, but run through the buffer cache if 79 * the system is under memory pressure (though this might just 80 * require inactive cleaning). 81 * 82 * 2 Be a bit more aggressive when running writes through the 83 * buffer cache when the system is under memory pressure. 84 * 85 * 3 Always run tmpfs writes through the buffer cache, thus forcing 86 * them out to swap. 87 */ 88 __read_mostly static int tmpfs_cluster_rd_enable = 1; 89 __read_mostly static int tmpfs_cluster_wr_enable = 1; 90 __read_mostly int tmpfs_bufcache_mode = 0; 91 SYSCTL_NODE(_vfs, OID_AUTO, tmpfs, CTLFLAG_RW, 0, "TMPFS filesystem"); 92 SYSCTL_INT(_vfs_tmpfs, OID_AUTO, cluster_rd_enable, CTLFLAG_RW, 93 &tmpfs_cluster_rd_enable, 0, ""); 94 SYSCTL_INT(_vfs_tmpfs, OID_AUTO, cluster_wr_enable, CTLFLAG_RW, 95 &tmpfs_cluster_wr_enable, 0, ""); 96 SYSCTL_INT(_vfs_tmpfs, OID_AUTO, bufcache_mode, CTLFLAG_RW, 97 &tmpfs_bufcache_mode, 0, ""); 98 99 #define TMPFS_MOVF_FROMBACKING 0x0001 100 #define TMPFS_MOVF_DEACTIVATE 0x0002 101 102 103 static __inline 104 void 105 tmpfs_knote(struct vnode *vp, int flags) 106 { 107 if (flags) 108 KNOTE(&vp->v_pollinfo.vpi_kqinfo.ki_note, flags); 109 } 110 111 112 /* --------------------------------------------------------------------- */ 113 114 static int 115 tmpfs_nresolve(struct vop_nresolve_args *ap) 116 { 117 struct vnode *dvp = ap->a_dvp; 118 struct vnode *vp = NULL; 119 struct namecache *ncp = ap->a_nch->ncp; 120 struct tmpfs_node *tnode; 121 struct tmpfs_dirent *de; 122 struct tmpfs_node *dnode; 123 int error; 124 125 dnode = VP_TO_TMPFS_DIR(dvp); 126 127 TMPFS_NODE_LOCK_SH(dnode); 128 loop: 129 de = tmpfs_dir_lookup(dnode, NULL, ncp); 130 if (de == NULL) { 131 error = ENOENT; 132 } else { 133 /* 134 * Allocate a vnode for the node we found. Use 135 * tmpfs_alloc_vp()'s deadlock handling mode. 136 */ 137 tnode = de->td_node; 138 error = tmpfs_alloc_vp(dvp->v_mount, dnode, tnode, 139 LK_EXCLUSIVE | LK_RETRY, &vp); 140 if (error == EAGAIN) 141 goto loop; 142 if (error) 143 goto out; 144 KKASSERT(vp); 145 } 146 147 out: 148 TMPFS_NODE_UNLOCK(dnode); 149 150 if ((dnode->tn_status & TMPFS_NODE_ACCESSED) == 0) { 151 TMPFS_NODE_LOCK(dnode); 152 dnode->tn_status |= TMPFS_NODE_ACCESSED; 153 TMPFS_NODE_UNLOCK(dnode); 154 } 155 156 /* 157 * Store the result of this lookup in the cache. Avoid this if the 158 * request was for creation, as it does not improve timings on 159 * emprical tests. 160 */ 161 if (vp) { 162 vn_unlock(vp); 163 cache_setvp(ap->a_nch, vp); 164 vrele(vp); 165 } else if (error == ENOENT) { 166 cache_setvp(ap->a_nch, NULL); 167 } 168 return (error); 169 } 170 171 static int 172 tmpfs_nlookupdotdot(struct vop_nlookupdotdot_args *ap) 173 { 174 struct vnode *dvp = ap->a_dvp; 175 struct vnode **vpp = ap->a_vpp; 176 struct tmpfs_node *dnode = VP_TO_TMPFS_NODE(dvp); 177 struct ucred *cred = ap->a_cred; 178 int error; 179 180 *vpp = NULL; 181 182 /* Check accessibility of requested node as a first step. */ 183 error = VOP_ACCESS(dvp, VEXEC, cred); 184 if (error != 0) 185 return error; 186 187 if (dnode->tn_dir.tn_parent != NULL) { 188 /* Allocate a new vnode on the matching entry. */ 189 error = tmpfs_alloc_vp(dvp->v_mount, 190 NULL, dnode->tn_dir.tn_parent, 191 LK_EXCLUSIVE | LK_RETRY, vpp); 192 193 if (*vpp) 194 vn_unlock(*vpp); 195 } 196 return (*vpp == NULL) ? ENOENT : 0; 197 } 198 199 /* --------------------------------------------------------------------- */ 200 201 static int 202 tmpfs_ncreate(struct vop_ncreate_args *ap) 203 { 204 struct vnode *dvp = ap->a_dvp; 205 struct vnode **vpp = ap->a_vpp; 206 struct namecache *ncp = ap->a_nch->ncp; 207 struct vattr *vap = ap->a_vap; 208 struct ucred *cred = ap->a_cred; 209 int error; 210 211 KKASSERT(vap->va_type == VREG || vap->va_type == VSOCK); 212 213 error = tmpfs_alloc_file(dvp, vpp, vap, ncp, cred, NULL); 214 if (error == 0) { 215 cache_setunresolved(ap->a_nch); 216 cache_setvp(ap->a_nch, *vpp); 217 tmpfs_knote(dvp, NOTE_WRITE); 218 } 219 return (error); 220 } 221 /* --------------------------------------------------------------------- */ 222 223 static int 224 tmpfs_nmknod(struct vop_nmknod_args *ap) 225 { 226 struct vnode *dvp = ap->a_dvp; 227 struct vnode **vpp = ap->a_vpp; 228 struct namecache *ncp = ap->a_nch->ncp; 229 struct vattr *vap = ap->a_vap; 230 struct ucred *cred = ap->a_cred; 231 int error; 232 233 if (vap->va_type != VBLK && vap->va_type != VCHR && 234 vap->va_type != VFIFO) { 235 return (EINVAL); 236 } 237 238 error = tmpfs_alloc_file(dvp, vpp, vap, ncp, cred, NULL); 239 if (error == 0) { 240 cache_setunresolved(ap->a_nch); 241 cache_setvp(ap->a_nch, *vpp); 242 tmpfs_knote(dvp, NOTE_WRITE); 243 } 244 return error; 245 } 246 247 /* --------------------------------------------------------------------- */ 248 249 static int 250 tmpfs_open(struct vop_open_args *ap) 251 { 252 struct vnode *vp = ap->a_vp; 253 int mode = ap->a_mode; 254 struct tmpfs_node *node; 255 int error; 256 257 node = VP_TO_TMPFS_NODE(vp); 258 259 #if 0 260 /* The file is still active but all its names have been removed 261 * (e.g. by a "rmdir $(pwd)"). It cannot be opened any more as 262 * it is about to die. */ 263 if (node->tn_links < 1) 264 return (ENOENT); 265 #endif 266 267 /* If the file is marked append-only, deny write requests. */ 268 if ((node->tn_flags & APPEND) && 269 (mode & (FWRITE | O_APPEND)) == FWRITE) { 270 error = EPERM; 271 } else { 272 if (node->tn_reg.tn_pages_in_aobj) { 273 TMPFS_NODE_LOCK(node); 274 if (node->tn_reg.tn_pages_in_aobj) { 275 tmpfs_move_pages(node->tn_reg.tn_aobj, 276 vp->v_object, 277 TMPFS_MOVF_FROMBACKING); 278 node->tn_reg.tn_pages_in_aobj = 0; 279 } 280 TMPFS_NODE_UNLOCK(node); 281 } 282 error = vop_stdopen(ap); 283 } 284 285 return (error); 286 } 287 288 /* --------------------------------------------------------------------- */ 289 290 static int 291 tmpfs_close(struct vop_close_args *ap) 292 { 293 struct vnode *vp = ap->a_vp; 294 struct tmpfs_node *node; 295 int error; 296 297 node = VP_TO_TMPFS_NODE(vp); 298 299 if (node->tn_links > 0) { 300 /* 301 * Update node times. No need to do it if the node has 302 * been deleted, because it will vanish after we return. 303 */ 304 tmpfs_update(vp); 305 } 306 307 error = vop_stdclose(ap); 308 309 return (error); 310 } 311 312 /* --------------------------------------------------------------------- */ 313 314 int 315 tmpfs_access(struct vop_access_args *ap) 316 { 317 struct vnode *vp = ap->a_vp; 318 int error; 319 struct tmpfs_node *node; 320 321 node = VP_TO_TMPFS_NODE(vp); 322 323 switch (vp->v_type) { 324 case VDIR: 325 /* FALLTHROUGH */ 326 case VLNK: 327 /* FALLTHROUGH */ 328 case VREG: 329 if ((ap->a_mode & VWRITE) && 330 (vp->v_mount->mnt_flag & MNT_RDONLY)) { 331 error = EROFS; 332 goto out; 333 } 334 break; 335 336 case VBLK: 337 /* FALLTHROUGH */ 338 case VCHR: 339 /* FALLTHROUGH */ 340 case VSOCK: 341 /* FALLTHROUGH */ 342 case VFIFO: 343 break; 344 345 default: 346 error = EINVAL; 347 goto out; 348 } 349 350 if ((ap->a_mode & VWRITE) && (node->tn_flags & IMMUTABLE)) { 351 error = EPERM; 352 goto out; 353 } 354 355 error = vop_helper_access(ap, node->tn_uid, node->tn_gid, 356 node->tn_mode, 0); 357 out: 358 return error; 359 } 360 361 /* --------------------------------------------------------------------- */ 362 363 int 364 tmpfs_getattr(struct vop_getattr_args *ap) 365 { 366 struct vnode *vp = ap->a_vp; 367 struct vattr *vap = ap->a_vap; 368 struct tmpfs_node *node; 369 370 node = VP_TO_TMPFS_NODE(vp); 371 372 tmpfs_update(vp); 373 374 vap->va_type = vp->v_type; 375 vap->va_mode = node->tn_mode; 376 vap->va_nlink = node->tn_links; 377 vap->va_uid = node->tn_uid; 378 vap->va_gid = node->tn_gid; 379 vap->va_fsid = vp->v_mount->mnt_stat.f_fsid.val[0]; 380 vap->va_fileid = node->tn_id; 381 vap->va_size = node->tn_size; 382 vap->va_blocksize = PAGE_SIZE; 383 vap->va_atime.tv_sec = node->tn_atime; 384 vap->va_atime.tv_nsec = node->tn_atimensec; 385 vap->va_mtime.tv_sec = node->tn_mtime; 386 vap->va_mtime.tv_nsec = node->tn_mtimensec; 387 vap->va_ctime.tv_sec = node->tn_ctime; 388 vap->va_ctime.tv_nsec = node->tn_ctimensec; 389 vap->va_gen = node->tn_gen; 390 vap->va_flags = node->tn_flags; 391 if (vp->v_type == VBLK || vp->v_type == VCHR) { 392 vap->va_rmajor = umajor(node->tn_rdev); 393 vap->va_rminor = uminor(node->tn_rdev); 394 } 395 vap->va_bytes = round_page(node->tn_size); 396 vap->va_filerev = 0; 397 398 return 0; 399 } 400 401 /* --------------------------------------------------------------------- */ 402 403 int 404 tmpfs_getattr_lite(struct vop_getattr_lite_args *ap) 405 { 406 struct vnode *vp = ap->a_vp; 407 struct vattr_lite *lvap = ap->a_lvap; 408 struct tmpfs_node *node; 409 410 node = VP_TO_TMPFS_NODE(vp); 411 412 tmpfs_update(vp); 413 414 lvap->va_type = vp->v_type; 415 lvap->va_mode = node->tn_mode; 416 lvap->va_nlink = node->tn_links; 417 lvap->va_uid = node->tn_uid; 418 lvap->va_gid = node->tn_gid; 419 #if 0 420 vap->va_fsid = vp->v_mount->mnt_stat.f_fsid.val[0]; 421 vap->va_fileid = node->tn_id; 422 #endif 423 lvap->va_size = node->tn_size; 424 #if 0 425 vap->va_blocksize = PAGE_SIZE; 426 vap->va_gen = node->tn_gen; 427 #endif 428 lvap->va_flags = node->tn_flags; 429 #if 0 430 if (vp->v_type == VBLK || vp->v_type == VCHR) { 431 vap->va_rmajor = umajor(node->tn_rdev); 432 vap->va_rminor = uminor(node->tn_rdev); 433 } 434 vap->va_bytes = -1; 435 vap->va_filerev = 0; 436 #endif 437 438 return 0; 439 } 440 441 442 /* --------------------------------------------------------------------- */ 443 444 int 445 tmpfs_setattr(struct vop_setattr_args *ap) 446 { 447 struct vnode *vp = ap->a_vp; 448 struct vattr *vap = ap->a_vap; 449 struct ucred *cred = ap->a_cred; 450 struct tmpfs_node *node = VP_TO_TMPFS_NODE(vp); 451 int error = 0; 452 int kflags = 0; 453 454 TMPFS_NODE_LOCK(node); 455 if (error == 0 && (vap->va_flags != VNOVAL)) { 456 error = tmpfs_chflags(vp, vap->va_flags, cred); 457 kflags |= NOTE_ATTRIB; 458 } 459 460 if (error == 0 && (vap->va_size != VNOVAL)) { 461 /* restore any saved pages before proceeding */ 462 if (node->tn_reg.tn_pages_in_aobj) { 463 tmpfs_move_pages(node->tn_reg.tn_aobj, vp->v_object, 464 TMPFS_MOVF_FROMBACKING | 465 TMPFS_MOVF_DEACTIVATE); 466 node->tn_reg.tn_pages_in_aobj = 0; 467 } 468 if (vap->va_size > node->tn_size) 469 kflags |= NOTE_WRITE | NOTE_EXTEND; 470 else 471 kflags |= NOTE_WRITE; 472 error = tmpfs_chsize(vp, vap->va_size, cred); 473 } 474 475 if (error == 0 && (vap->va_uid != (uid_t)VNOVAL || 476 vap->va_gid != (gid_t)VNOVAL)) { 477 error = tmpfs_chown(vp, vap->va_uid, vap->va_gid, cred); 478 kflags |= NOTE_ATTRIB; 479 } 480 481 if (error == 0 && (vap->va_mode != (mode_t)VNOVAL)) { 482 error = tmpfs_chmod(vp, vap->va_mode, cred); 483 kflags |= NOTE_ATTRIB; 484 } 485 486 if (error == 0 && ((vap->va_atime.tv_sec != VNOVAL && 487 vap->va_atime.tv_nsec != VNOVAL) || 488 (vap->va_mtime.tv_sec != VNOVAL && 489 vap->va_mtime.tv_nsec != VNOVAL) )) { 490 error = tmpfs_chtimes(vp, &vap->va_atime, &vap->va_mtime, 491 vap->va_vaflags, cred); 492 kflags |= NOTE_ATTRIB; 493 } 494 495 /* 496 * Update the node times. We give preference to the error codes 497 * generated by this function rather than the ones that may arise 498 * from tmpfs_update. 499 */ 500 tmpfs_update(vp); 501 TMPFS_NODE_UNLOCK(node); 502 tmpfs_knote(vp, kflags); 503 504 return (error); 505 } 506 507 /* --------------------------------------------------------------------- */ 508 509 /* 510 * fsync is usually a NOP, but we must take action when unmounting or 511 * when recycling. 512 */ 513 static int 514 tmpfs_fsync(struct vop_fsync_args *ap) 515 { 516 struct tmpfs_node *node; 517 struct vnode *vp = ap->a_vp; 518 519 node = VP_TO_TMPFS_NODE(vp); 520 521 /* 522 * tmpfs vnodes typically remain dirty, avoid long syncer scans 523 * by forcing removal from the syncer list. 524 */ 525 vn_syncer_remove(vp, 1); 526 527 tmpfs_update(vp); 528 if (vp->v_type == VREG) { 529 if (vp->v_flag & VRECLAIMED) { 530 if (node->tn_links == 0) 531 tmpfs_truncate(vp, 0); 532 else 533 vfsync(ap->a_vp, ap->a_waitfor, 1, NULL, NULL); 534 } 535 } 536 537 return 0; 538 } 539 540 /* --------------------------------------------------------------------- */ 541 542 static int 543 tmpfs_read(struct vop_read_args *ap) 544 { 545 struct buf *bp; 546 struct vnode *vp = ap->a_vp; 547 struct uio *uio = ap->a_uio; 548 struct tmpfs_node *node; 549 off_t base_offset; 550 size_t offset; 551 size_t len; 552 size_t resid; 553 int error; 554 int seqcount; 555 556 /* 557 * Check the basics 558 */ 559 if (uio->uio_offset < 0) 560 return (EINVAL); 561 if (vp->v_type != VREG) 562 return (EINVAL); 563 564 /* 565 * Extract node, try to shortcut the operation through 566 * the VM page cache, allowing us to avoid buffer cache 567 * overheads. 568 */ 569 node = VP_TO_TMPFS_NODE(vp); 570 resid = uio->uio_resid; 571 seqcount = ap->a_ioflag >> IO_SEQSHIFT; 572 error = vop_helper_read_shortcut(ap); 573 if (error) 574 return error; 575 if (uio->uio_resid == 0) { 576 if (resid) 577 goto finished; 578 return error; 579 } 580 581 /* 582 * restore any saved pages before proceeding 583 */ 584 if (node->tn_reg.tn_pages_in_aobj) { 585 TMPFS_NODE_LOCK(node); 586 if (node->tn_reg.tn_pages_in_aobj) { 587 tmpfs_move_pages(node->tn_reg.tn_aobj, vp->v_object, 588 TMPFS_MOVF_FROMBACKING); 589 node->tn_reg.tn_pages_in_aobj = 0; 590 } 591 TMPFS_NODE_UNLOCK(node); 592 } 593 594 /* 595 * Fall-through to our normal read code. 596 */ 597 while (uio->uio_resid > 0 && uio->uio_offset < node->tn_size) { 598 /* 599 * Use buffer cache I/O (via tmpfs_strategy) 600 */ 601 offset = (size_t)uio->uio_offset & TMPFS_BLKMASK64; 602 base_offset = (off_t)uio->uio_offset - offset; 603 bp = getcacheblk(vp, base_offset, 604 node->tn_blksize, GETBLK_KVABIO); 605 if (bp == NULL) { 606 if (tmpfs_cluster_rd_enable) { 607 error = cluster_readx(vp, node->tn_size, 608 base_offset, 609 node->tn_blksize, 610 B_NOTMETA | B_KVABIO, 611 uio->uio_resid, 612 seqcount * MAXBSIZE, 613 &bp); 614 } else { 615 error = bread_kvabio(vp, base_offset, 616 node->tn_blksize, &bp); 617 } 618 if (error) { 619 brelse(bp); 620 kprintf("tmpfs_read bread error %d\n", error); 621 break; 622 } 623 624 /* 625 * tmpfs pretty much fiddles directly with the VM 626 * system, don't let it exhaust it or we won't play 627 * nice with other processes. 628 * 629 * Only do this if the VOP is coming from a normal 630 * read/write. The VM system handles the case for 631 * UIO_NOCOPY. 632 */ 633 if (uio->uio_segflg != UIO_NOCOPY) 634 vm_wait_nominal(); 635 } 636 bp->b_flags |= B_CLUSTEROK; 637 bkvasync(bp); 638 639 /* 640 * Figure out how many bytes we can actually copy this loop. 641 */ 642 len = node->tn_blksize - offset; 643 if (len > uio->uio_resid) 644 len = uio->uio_resid; 645 if (len > node->tn_size - uio->uio_offset) 646 len = (size_t)(node->tn_size - uio->uio_offset); 647 648 error = uiomovebp(bp, (char *)bp->b_data + offset, len, uio); 649 bqrelse(bp); 650 if (error) { 651 kprintf("tmpfs_read uiomove error %d\n", error); 652 break; 653 } 654 } 655 656 finished: 657 if ((node->tn_status & TMPFS_NODE_ACCESSED) == 0) { 658 TMPFS_NODE_LOCK(node); 659 node->tn_status |= TMPFS_NODE_ACCESSED; 660 TMPFS_NODE_UNLOCK(node); 661 } 662 return (error); 663 } 664 665 static int 666 tmpfs_write(struct vop_write_args *ap) 667 { 668 struct buf *bp; 669 struct vnode *vp = ap->a_vp; 670 struct uio *uio = ap->a_uio; 671 struct thread *td = uio->uio_td; 672 struct tmpfs_node *node; 673 boolean_t extended; 674 off_t oldsize; 675 int error; 676 off_t base_offset; 677 size_t offset; 678 size_t len; 679 struct rlimit limit; 680 int trivial = 0; 681 int kflags = 0; 682 int seqcount; 683 684 error = 0; 685 if (uio->uio_resid == 0) { 686 return error; 687 } 688 689 node = VP_TO_TMPFS_NODE(vp); 690 691 if (vp->v_type != VREG) 692 return (EINVAL); 693 seqcount = ap->a_ioflag >> IO_SEQSHIFT; 694 695 TMPFS_NODE_LOCK(node); 696 697 /* 698 * restore any saved pages before proceeding 699 */ 700 if (node->tn_reg.tn_pages_in_aobj) { 701 tmpfs_move_pages(node->tn_reg.tn_aobj, vp->v_object, 702 TMPFS_MOVF_FROMBACKING); 703 node->tn_reg.tn_pages_in_aobj = 0; 704 } 705 706 oldsize = node->tn_size; 707 if (ap->a_ioflag & IO_APPEND) 708 uio->uio_offset = node->tn_size; 709 710 /* 711 * Check for illegal write offsets. 712 */ 713 if (uio->uio_offset + uio->uio_resid > 714 VFS_TO_TMPFS(vp->v_mount)->tm_maxfilesize) { 715 error = EFBIG; 716 goto done; 717 } 718 719 /* 720 * NOTE: Ignore if UIO does not come from a user thread (e.g. VN). 721 */ 722 if (vp->v_type == VREG && td != NULL && td->td_lwp != NULL) { 723 error = kern_getrlimit(RLIMIT_FSIZE, &limit); 724 if (error) 725 goto done; 726 if (uio->uio_offset + uio->uio_resid > limit.rlim_cur) { 727 ksignal(td->td_proc, SIGXFSZ); 728 error = EFBIG; 729 goto done; 730 } 731 } 732 733 /* 734 * Extend the file's size if necessary 735 */ 736 extended = ((uio->uio_offset + uio->uio_resid) > node->tn_size); 737 738 while (uio->uio_resid > 0) { 739 /* 740 * Don't completely blow out running buffer I/O 741 * when being hit from the pageout daemon. 742 */ 743 if (uio->uio_segflg == UIO_NOCOPY && 744 (ap->a_ioflag & IO_RECURSE) == 0) { 745 bwillwrite(node->tn_blksize); 746 } 747 748 /* 749 * Use buffer cache I/O (via tmpfs_strategy) 750 * 751 * Calculate the maximum bytes we can write to the buffer at 752 * this offset (after resizing). 753 */ 754 offset = (size_t)uio->uio_offset & TMPFS_BLKMASK64; 755 base_offset = (off_t)uio->uio_offset - offset; 756 len = uio->uio_resid; 757 if (len > TMPFS_BLKSIZE - offset) 758 len = TMPFS_BLKSIZE - offset; 759 760 if ((uio->uio_offset + len) > node->tn_size) { 761 trivial = (uio->uio_offset <= node->tn_size); 762 error = tmpfs_reg_resize(vp, uio->uio_offset + len, 763 trivial); 764 if (error) 765 break; 766 } 767 768 /* 769 * Read to fill in any gaps. Theoretically we could 770 * optimize this if the write covers the entire buffer 771 * and is not a UIO_NOCOPY write, however this can lead 772 * to a security violation exposing random kernel memory 773 * (whatever junk was in the backing VM pages before). 774 * 775 * So just use bread() to do the right thing. 776 */ 777 error = bread_kvabio(vp, base_offset, node->tn_blksize, &bp); 778 bkvasync(bp); 779 error = uiomovebp(bp, (char *)bp->b_data + offset, len, uio); 780 if (error) { 781 kprintf("tmpfs_write uiomove error %d\n", error); 782 brelse(bp); 783 break; 784 } 785 786 if (uio->uio_offset > node->tn_size) { 787 node->tn_size = uio->uio_offset; 788 kflags |= NOTE_EXTEND; 789 } 790 kflags |= NOTE_WRITE; 791 792 /* 793 * UIO_NOCOPY is a sensitive state due to potentially being 794 * issued from the pageout daemon while in a low-memory 795 * situation. However, in order to cluster the I/O nicely 796 * (e.g. 64KB+ writes instead of 16KB writes), we still try 797 * to follow the same semantics that any other filesystem 798 * might use. 799 * 800 * For the normal case we buwrite(), dirtying the underlying 801 * VM pages instead of dirtying the buffer and releasing the 802 * buffer as a clean buffer. This allows tmpfs to use 803 * essentially all available memory to cache file data. 804 * If we used bdwrite() the buffer cache would wind up 805 * flushing the data to swap too quickly. 806 * 807 * But because tmpfs can seriously load the VM system we 808 * fall-back to using bdwrite() when free memory starts 809 * to get low. This shifts the load away from the VM system 810 * and makes tmpfs act more like a normal filesystem with 811 * regards to disk activity. 812 * 813 * tmpfs pretty much fiddles directly with the VM 814 * system, don't let it exhaust it or we won't play 815 * nice with other processes. Only do this if the 816 * VOP is coming from a normal read/write. The VM system 817 * handles the case for UIO_NOCOPY. 818 */ 819 bp->b_flags |= B_CLUSTEROK; 820 if (uio->uio_segflg == UIO_NOCOPY) { 821 /* 822 * Flush from the pageout daemon, deal with potentially 823 * very heavy tmpfs write activity causing long stalls 824 * in the pageout daemon before pages get to free/cache. 825 * 826 * We have to be careful not to bypass the page queues 827 * entirely or we can cause write-read thrashing and 828 * delay the paging of data that is more pageable then 829 * our current data. 830 * 831 * (a) Under severe pressure setting B_DIRECT will 832 * cause a buffer release to try to free the 833 * underlying pages. 834 * 835 * (b) Under modest memory pressure the B_AGE flag 836 * we retire the buffer and its underlying pages 837 * more quickly than normal. 838 * 839 * We could also force this by setting B_NOTMETA 840 * but that might have other unintended side- 841 * effects (e.g. setting PG_NOTMETA on the VM page). 842 * 843 * (c) For the pageout->putpages->generic_putpages-> 844 * UIO_NOCOPY-write (here), issuing an immediate 845 * write prevents any real clustering from 846 * happening because the buffers probably aren't 847 * (yet) marked dirty, or lost due to prior use 848 * of buwrite(). Try to use the normal 849 * cluster_write() mechanism for performance. 850 * 851 * Hopefully this will unblock the VM system more 852 * quickly under extreme tmpfs write load. 853 */ 854 if (tmpfs_bufcache_mode >= 2) { 855 if (vm_paging_min_dnc(vm_page_free_hysteresis)) 856 bp->b_flags |= B_DIRECT | B_TTC; 857 if (vm_pages_needed || vm_paging_start(0)) 858 bp->b_flags |= B_AGE; 859 } 860 bp->b_flags |= B_RELBUF; 861 bp->b_act_count = 0; /* buffer->deactivate pgs */ 862 if (tmpfs_cluster_wr_enable && 863 (ap->a_ioflag & (IO_SYNC | IO_DIRECT)) == 0) { 864 cluster_write(bp, node->tn_size, 865 node->tn_blksize, seqcount); 866 } else { 867 cluster_awrite(bp); 868 } 869 } else if (vm_paging_min() || 870 ((vm_pages_needed || vm_paging_start(0)) && 871 tmpfs_bufcache_mode >= 1)) { 872 /* 873 * If the pageout daemon is running we cycle the 874 * write through the buffer cache normally to 875 * pipeline the flush, thus avoiding adding any 876 * more memory pressure to the pageout daemon. 877 */ 878 bp->b_act_count = 0; /* buffer->deactivate pgs */ 879 if (tmpfs_cluster_wr_enable) { 880 cluster_write(bp, node->tn_size, 881 node->tn_blksize, seqcount); 882 } else { 883 bdwrite(bp); 884 } 885 } else { 886 /* 887 * Otherwise run the buffer directly through to the 888 * backing VM store, leaving the buffer clean so 889 * buffer limits do not force early flushes to swap. 890 */ 891 buwrite(bp); 892 /*vm_wait_nominal();*/ 893 } 894 895 if (bp->b_error) { 896 kprintf("tmpfs_write bwrite error %d\n", bp->b_error); 897 break; 898 } 899 } 900 901 if (error) { 902 if (extended) { 903 (void)tmpfs_reg_resize(vp, oldsize, trivial); 904 kflags &= ~NOTE_EXTEND; 905 } 906 goto done; 907 } 908 909 /* 910 * Currently we don't set the mtime on files modified via mmap() 911 * because we can't tell the difference between those modifications 912 * and an attempt by the pageout daemon to flush tmpfs pages to 913 * swap. 914 * 915 * This is because in order to defer flushes as long as possible 916 * buwrite() works by marking the underlying VM pages dirty in 917 * order to be able to dispose of the buffer cache buffer without 918 * flushing it. 919 */ 920 if (uio->uio_segflg == UIO_NOCOPY) { 921 if (vp->v_flag & VLASTWRITETS) { 922 node->tn_mtime = vp->v_lastwrite_ts.tv_sec; 923 node->tn_mtimensec = vp->v_lastwrite_ts.tv_nsec; 924 } 925 } else { 926 node->tn_status |= TMPFS_NODE_MODIFIED; 927 vclrflags(vp, VLASTWRITETS); 928 } 929 930 if (extended) 931 node->tn_status |= TMPFS_NODE_CHANGED; 932 933 if (node->tn_mode & (S_ISUID | S_ISGID)) { 934 if (priv_check_cred(ap->a_cred, PRIV_VFS_RETAINSUGID, 0)) 935 node->tn_mode &= ~(S_ISUID | S_ISGID); 936 } 937 done: 938 TMPFS_NODE_UNLOCK(node); 939 if (kflags) 940 tmpfs_knote(vp, kflags); 941 942 return(error); 943 } 944 945 static int 946 tmpfs_advlock(struct vop_advlock_args *ap) 947 { 948 struct tmpfs_node *node; 949 struct vnode *vp = ap->a_vp; 950 int error; 951 952 node = VP_TO_TMPFS_NODE(vp); 953 error = (lf_advlock(ap, &node->tn_advlock, node->tn_size)); 954 955 return (error); 956 } 957 958 /* 959 * The strategy function is typically only called when memory pressure 960 * forces the system to attempt to pageout pages. It can also be called 961 * by [n]vtruncbuf() when a truncation cuts a page in half. Normal write 962 * operations 963 * 964 * We set VKVABIO for VREG files so bp->b_data may not be synchronized to 965 * our cpu. swap_pager_strategy() is all we really use, and it directly 966 * supports this. 967 */ 968 static int 969 tmpfs_strategy(struct vop_strategy_args *ap) 970 { 971 struct bio *bio = ap->a_bio; 972 struct bio *nbio; 973 struct buf *bp = bio->bio_buf; 974 struct vnode *vp = ap->a_vp; 975 struct tmpfs_node *node; 976 vm_object_t uobj; 977 vm_page_t m; 978 int i; 979 980 if (vp->v_type != VREG) { 981 bp->b_resid = bp->b_bcount; 982 bp->b_flags |= B_ERROR | B_INVAL; 983 bp->b_error = EINVAL; 984 biodone(bio); 985 return(0); 986 } 987 988 node = VP_TO_TMPFS_NODE(vp); 989 990 uobj = node->tn_reg.tn_aobj; 991 992 /* 993 * Don't bother flushing to swap if there is no swap, just 994 * ensure that the pages are marked as needing a commit (still). 995 */ 996 if (bp->b_cmd == BUF_CMD_WRITE && vm_swap_size == 0) { 997 for (i = 0; i < bp->b_xio.xio_npages; ++i) { 998 m = bp->b_xio.xio_pages[i]; 999 vm_page_need_commit(m); 1000 } 1001 bp->b_resid = 0; 1002 bp->b_error = 0; 1003 biodone(bio); 1004 } else { 1005 #if 0 1006 /* 1007 * XXX removed, this does not work well because under heavy 1008 * filesystem loads it often 1009 * forces the data to be read right back in again after 1010 * being written due to bypassing normal LRU operation. 1011 * 1012 * Tell the buffer cache to try to recycle the pages 1013 * to PQ_CACHE on release. 1014 */ 1015 if (tmpfs_bufcache_mode >= 2 || 1016 (tmpfs_bufcache_mode == 1 && vm_paging_needed(0))) { 1017 bp->b_flags |= B_TTC; 1018 } 1019 #endif 1020 nbio = push_bio(bio); 1021 nbio->bio_done = tmpfs_strategy_done; 1022 nbio->bio_offset = bio->bio_offset; 1023 swap_pager_strategy(uobj, nbio); 1024 } 1025 return 0; 1026 } 1027 1028 /* 1029 * If we were unable to commit the pages to swap make sure they are marked 1030 * as needing a commit (again). If we were, clear the flag to allow the 1031 * pages to be freed. 1032 * 1033 * Do not error-out the buffer. In particular, vinvalbuf() needs to 1034 * always work. 1035 */ 1036 static void 1037 tmpfs_strategy_done(struct bio *bio) 1038 { 1039 struct buf *bp; 1040 vm_page_t m; 1041 int i; 1042 1043 bp = bio->bio_buf; 1044 1045 if (bp->b_flags & B_ERROR) { 1046 bp->b_flags &= ~B_ERROR; 1047 bp->b_error = 0; 1048 bp->b_resid = 0; 1049 for (i = 0; i < bp->b_xio.xio_npages; ++i) { 1050 m = bp->b_xio.xio_pages[i]; 1051 vm_page_need_commit(m); 1052 } 1053 } else { 1054 for (i = 0; i < bp->b_xio.xio_npages; ++i) { 1055 m = bp->b_xio.xio_pages[i]; 1056 vm_page_clear_commit(m); 1057 } 1058 } 1059 bio = pop_bio(bio); 1060 biodone(bio); 1061 } 1062 1063 /* 1064 * To make write clustering work well make the backing store look 1065 * contiguous to the cluster_*() code. The swap_strategy() function 1066 * will take it from there. 1067 * 1068 * Use MAXBSIZE-sized chunks as a micro-optimization to make random 1069 * flushes leave full-sized gaps. 1070 */ 1071 static int 1072 tmpfs_bmap(struct vop_bmap_args *ap) 1073 { 1074 if (ap->a_doffsetp != NULL) 1075 *ap->a_doffsetp = ap->a_loffset; 1076 if (ap->a_runp != NULL) 1077 *ap->a_runp = MAXBSIZE - (ap->a_loffset & (MAXBSIZE - 1)); 1078 if (ap->a_runb != NULL) 1079 *ap->a_runb = ap->a_loffset & (MAXBSIZE - 1); 1080 1081 return 0; 1082 } 1083 1084 /* --------------------------------------------------------------------- */ 1085 1086 static int 1087 tmpfs_nremove(struct vop_nremove_args *ap) 1088 { 1089 struct vnode *dvp = ap->a_dvp; 1090 struct namecache *ncp = ap->a_nch->ncp; 1091 struct vnode *vp; 1092 int error; 1093 struct tmpfs_dirent *de; 1094 struct tmpfs_mount *tmp; 1095 struct tmpfs_node *dnode; 1096 struct tmpfs_node *node; 1097 1098 /* 1099 * We have to acquire the vp from ap->a_nch because we will likely 1100 * unresolve the namecache entry, and a vrele/vput is needed to 1101 * trigger the tmpfs_inactive/tmpfs_reclaim sequence. 1102 * 1103 * We have to use vget to clear any inactive state on the vnode, 1104 * otherwise the vnode may remain inactive and thus tmpfs_inactive 1105 * will not get called when we release it. 1106 */ 1107 error = cache_vget(ap->a_nch, ap->a_cred, LK_SHARED, &vp); 1108 KKASSERT(vp->v_mount == dvp->v_mount); 1109 KKASSERT(error == 0); 1110 vn_unlock(vp); 1111 1112 if (vp->v_type == VDIR) { 1113 error = EISDIR; 1114 goto out2; 1115 } 1116 1117 dnode = VP_TO_TMPFS_DIR(dvp); 1118 node = VP_TO_TMPFS_NODE(vp); 1119 tmp = VFS_TO_TMPFS(vp->v_mount); 1120 1121 TMPFS_NODE_LOCK(dnode); 1122 TMPFS_NODE_LOCK(node); 1123 de = tmpfs_dir_lookup(dnode, node, ncp); 1124 if (de == NULL) { 1125 error = ENOENT; 1126 TMPFS_NODE_UNLOCK(node); 1127 TMPFS_NODE_UNLOCK(dnode); 1128 goto out; 1129 } 1130 1131 /* Files marked as immutable or append-only cannot be deleted. */ 1132 if ((node->tn_flags & (IMMUTABLE | APPEND | NOUNLINK)) || 1133 (dnode->tn_flags & APPEND)) { 1134 error = EPERM; 1135 TMPFS_NODE_UNLOCK(node); 1136 TMPFS_NODE_UNLOCK(dnode); 1137 goto out; 1138 } 1139 1140 /* Remove the entry from the directory; as it is a file, we do not 1141 * have to change the number of hard links of the directory. */ 1142 tmpfs_dir_detach_locked(dnode, de); 1143 TMPFS_NODE_UNLOCK(dnode); 1144 1145 /* Free the directory entry we just deleted. Note that the node 1146 * referred by it will not be removed until the vnode is really 1147 * reclaimed. */ 1148 tmpfs_free_dirent(tmp, de); 1149 1150 if (node->tn_links > 0) 1151 node->tn_status |= TMPFS_NODE_CHANGED; 1152 TMPFS_NODE_UNLOCK(node); 1153 1154 cache_unlink(ap->a_nch); 1155 tmpfs_knote(vp, NOTE_DELETE); 1156 error = 0; 1157 1158 out: 1159 if (error == 0) 1160 tmpfs_knote(dvp, NOTE_WRITE); 1161 out2: 1162 vrele(vp); 1163 1164 return error; 1165 } 1166 1167 /* --------------------------------------------------------------------- */ 1168 1169 static int 1170 tmpfs_nlink(struct vop_nlink_args *ap) 1171 { 1172 struct vnode *dvp = ap->a_dvp; 1173 struct vnode *vp = ap->a_vp; 1174 struct tmpfs_mount *tmp = VFS_TO_TMPFS(vp->v_mount); 1175 struct namecache *ncp = ap->a_nch->ncp; 1176 struct tmpfs_dirent *de; 1177 struct tmpfs_node *node; 1178 struct tmpfs_node *dnode; 1179 int error; 1180 1181 KKASSERT(dvp != vp); /* XXX When can this be false? */ 1182 1183 node = VP_TO_TMPFS_NODE(vp); 1184 dnode = VP_TO_TMPFS_NODE(dvp); 1185 TMPFS_NODE_LOCK(dnode); 1186 1187 /* XXX: Why aren't the following two tests done by the caller? */ 1188 1189 /* Hard links of directories are forbidden. */ 1190 if (vp->v_type == VDIR) { 1191 error = EPERM; 1192 goto out; 1193 } 1194 1195 /* Cannot create cross-device links. */ 1196 if (dvp->v_mount != vp->v_mount) { 1197 error = EXDEV; 1198 goto out; 1199 } 1200 1201 /* Cannot hard-link into a deleted directory */ 1202 if (dnode != tmp->tm_root && dnode->tn_dir.tn_parent == NULL) { 1203 error = ENOENT; 1204 goto out; 1205 } 1206 1207 /* Ensure that we do not overflow the maximum number of links imposed 1208 * by the system. */ 1209 KKASSERT(node->tn_links <= LINK_MAX); 1210 if (node->tn_links >= LINK_MAX) { 1211 error = EMLINK; 1212 goto out; 1213 } 1214 1215 /* We cannot create links of files marked immutable or append-only. */ 1216 if (node->tn_flags & (IMMUTABLE | APPEND)) { 1217 error = EPERM; 1218 goto out; 1219 } 1220 1221 /* Allocate a new directory entry to represent the node. */ 1222 error = tmpfs_alloc_dirent(VFS_TO_TMPFS(vp->v_mount), node, 1223 ncp->nc_name, ncp->nc_nlen, &de); 1224 if (error != 0) 1225 goto out; 1226 1227 /* Insert the new directory entry into the appropriate directory. */ 1228 tmpfs_dir_attach_locked(dnode, de); 1229 1230 /* vp link count has changed, so update node times. */ 1231 1232 TMPFS_NODE_LOCK(node); 1233 node->tn_status |= TMPFS_NODE_CHANGED; 1234 TMPFS_NODE_UNLOCK(node); 1235 tmpfs_update(vp); 1236 1237 tmpfs_knote(vp, NOTE_LINK); 1238 cache_setunresolved(ap->a_nch); 1239 cache_setvp(ap->a_nch, vp); 1240 error = 0; 1241 1242 out: 1243 TMPFS_NODE_UNLOCK(dnode); 1244 if (error == 0) 1245 tmpfs_knote(dvp, NOTE_WRITE); 1246 return error; 1247 } 1248 1249 /* --------------------------------------------------------------------- */ 1250 1251 static int 1252 tmpfs_nrename(struct vop_nrename_args *ap) 1253 { 1254 struct vnode *fdvp = ap->a_fdvp; 1255 struct namecache *fncp = ap->a_fnch->ncp; 1256 struct vnode *fvp = fncp->nc_vp; 1257 struct vnode *tdvp = ap->a_tdvp; 1258 struct namecache *tncp = ap->a_tnch->ncp; 1259 struct vnode *tvp; 1260 struct tmpfs_dirent *de, *tde, *de2; 1261 struct tmpfs_mount *tmp; 1262 struct tmpfs_node *fdnode; 1263 struct tmpfs_node *tdnode; 1264 struct tmpfs_node *fnode; 1265 struct tmpfs_node *tnode; 1266 char *newname; 1267 char *oldname; 1268 int error; 1269 1270 KKASSERT(fdvp->v_mount == fvp->v_mount); 1271 1272 /* 1273 * Because tvp can get overwritten we have to vget it instead of 1274 * just vref or use it, otherwise it's VINACTIVE flag may not get 1275 * cleared and the node won't get destroyed. 1276 */ 1277 error = cache_vget(ap->a_tnch, ap->a_cred, LK_SHARED, &tvp); 1278 if (error == 0) { 1279 tnode = VP_TO_TMPFS_NODE(tvp); 1280 vn_unlock(tvp); 1281 } else { 1282 tnode = NULL; 1283 } 1284 1285 /* Disallow cross-device renames. 1286 * XXX Why isn't this done by the caller? */ 1287 if (fvp->v_mount != tdvp->v_mount || 1288 (tvp != NULL && fvp->v_mount != tvp->v_mount)) { 1289 error = EXDEV; 1290 goto out; 1291 } 1292 1293 tmp = VFS_TO_TMPFS(tdvp->v_mount); 1294 tdnode = VP_TO_TMPFS_DIR(tdvp); 1295 1296 /* If source and target are the same file, there is nothing to do. */ 1297 if (fvp == tvp) { 1298 error = 0; 1299 goto out; 1300 } 1301 1302 fdnode = VP_TO_TMPFS_DIR(fdvp); 1303 fnode = VP_TO_TMPFS_NODE(fvp); 1304 1305 tmpfs_lock4(fdnode, tdnode, fnode, tnode); 1306 1307 /* 1308 * Cannot rename into a deleted directory 1309 */ 1310 if (tdnode != tmp->tm_root && tdnode->tn_dir.tn_parent == NULL) { 1311 error = ENOENT; 1312 goto out_locked; 1313 } 1314 1315 /* Avoid manipulating '.' and '..' entries. */ 1316 de = tmpfs_dir_lookup(fdnode, fnode, fncp); 1317 if (de == NULL) { 1318 error = ENOENT; 1319 goto out_locked; 1320 } 1321 KKASSERT(de->td_node == fnode); 1322 1323 /* 1324 * If replacing an entry in the target directory and that entry 1325 * is a directory, it must be empty. 1326 * 1327 * Kern_rename gurantees the destination to be a directory 1328 * if the source is one (it does?). 1329 */ 1330 if (tvp != NULL) { 1331 KKASSERT(tnode != NULL); 1332 1333 if ((tnode->tn_flags & (NOUNLINK | IMMUTABLE | APPEND)) || 1334 (tdnode->tn_flags & (APPEND | IMMUTABLE))) { 1335 error = EPERM; 1336 goto out_locked; 1337 } 1338 1339 if (fnode->tn_type == VDIR && tnode->tn_type == VDIR) { 1340 if (tnode->tn_size > 0) { 1341 error = ENOTEMPTY; 1342 goto out_locked; 1343 } 1344 } else if (fnode->tn_type == VDIR && tnode->tn_type != VDIR) { 1345 error = ENOTDIR; 1346 goto out_locked; 1347 } else if (fnode->tn_type != VDIR && tnode->tn_type == VDIR) { 1348 error = EISDIR; 1349 goto out_locked; 1350 } else { 1351 KKASSERT(fnode->tn_type != VDIR && 1352 tnode->tn_type != VDIR); 1353 } 1354 } 1355 1356 if ((fnode->tn_flags & (NOUNLINK | IMMUTABLE | APPEND)) || 1357 (fdnode->tn_flags & (APPEND | IMMUTABLE))) { 1358 error = EPERM; 1359 goto out_locked; 1360 } 1361 1362 /* 1363 * Ensure that we have enough memory to hold the new name, if it 1364 * has to be changed. 1365 */ 1366 if (fncp->nc_nlen != tncp->nc_nlen || 1367 bcmp(fncp->nc_name, tncp->nc_name, fncp->nc_nlen) != 0) { 1368 newname = kmalloc(tncp->nc_nlen + 1, tmp->tm_name_zone, 1369 M_WAITOK | M_NULLOK); 1370 if (newname == NULL) { 1371 error = ENOSPC; 1372 goto out_locked; 1373 } 1374 bcopy(tncp->nc_name, newname, tncp->nc_nlen); 1375 newname[tncp->nc_nlen] = '\0'; 1376 } else { 1377 newname = NULL; 1378 } 1379 1380 /* 1381 * Unlink entry from source directory. Note that the kernel has 1382 * already checked for illegal recursion cases (renaming a directory 1383 * into a subdirectory of itself). 1384 */ 1385 if (fdnode != tdnode) { 1386 tmpfs_dir_detach_locked(fdnode, de); 1387 } else { 1388 /* XXX depend on namecache lock */ 1389 KKASSERT(de == tmpfs_dir_lookup(fdnode, fnode, fncp)); 1390 RB_REMOVE(tmpfs_dirtree, &fdnode->tn_dir.tn_dirtree, de); 1391 RB_REMOVE(tmpfs_dirtree_cookie, 1392 &fdnode->tn_dir.tn_cookietree, de); 1393 } 1394 1395 /* 1396 * Handle any name change. Swap with newname, we will 1397 * deallocate it at the end. 1398 */ 1399 if (newname != NULL) { 1400 oldname = de->td_name; 1401 de->td_name = newname; 1402 de->td_namelen = (uint16_t)tncp->nc_nlen; 1403 newname = oldname; 1404 } 1405 1406 /* 1407 * If we are overwriting an entry, we have to remove the old one 1408 * from the target directory. 1409 */ 1410 if (tvp != NULL) { 1411 /* Remove the old entry from the target directory. */ 1412 tde = tmpfs_dir_lookup(tdnode, tnode, tncp); 1413 tmpfs_dir_detach_locked(tdnode, tde); 1414 tmpfs_knote(tdnode->tn_vnode, NOTE_DELETE); 1415 1416 /* 1417 * Free the directory entry we just deleted. Note that the 1418 * node referred by it will not be removed until the vnode is 1419 * really reclaimed. 1420 */ 1421 tmpfs_free_dirent(VFS_TO_TMPFS(tvp->v_mount), tde); 1422 /*cache_inval_vp(tvp, CINV_DESTROY);*/ 1423 } 1424 1425 /* 1426 * Link entry to target directory. If the entry 1427 * represents a directory move the parent linkage 1428 * as well. 1429 */ 1430 if (fdnode != tdnode) { 1431 if (de->td_node->tn_type == VDIR) { 1432 TMPFS_VALIDATE_DIR(fnode); 1433 } 1434 tmpfs_dir_attach_locked(tdnode, de); 1435 } else { 1436 tdnode->tn_status |= TMPFS_NODE_MODIFIED; 1437 de2 = RB_INSERT(tmpfs_dirtree, &tdnode->tn_dir.tn_dirtree, de); 1438 KASSERT(de2 == NULL, 1439 ("tmpfs_nrenameA: duplicate insertion of %p, has %p\n", 1440 de, de2)); 1441 de2 = RB_INSERT(tmpfs_dirtree_cookie, 1442 &tdnode->tn_dir.tn_cookietree, de); 1443 KASSERT(de2 == NULL, 1444 ("tmpfs_nrenameB: duplicate insertion of %p, has %p\n", 1445 de, de2)); 1446 } 1447 tmpfs_unlock4(fdnode, tdnode, fnode, tnode); 1448 1449 /* 1450 * Finish up 1451 */ 1452 if (newname) { 1453 kfree(newname, tmp->tm_name_zone); 1454 newname = NULL; 1455 } 1456 cache_rename(ap->a_fnch, ap->a_tnch); 1457 tmpfs_knote(ap->a_fdvp, NOTE_WRITE); 1458 tmpfs_knote(ap->a_tdvp, NOTE_WRITE); 1459 if (fnode->tn_vnode) 1460 tmpfs_knote(fnode->tn_vnode, NOTE_RENAME); 1461 if (tvp) 1462 vrele(tvp); 1463 return 0; 1464 1465 out_locked: 1466 tmpfs_unlock4(fdnode, tdnode, fnode, tnode); 1467 out: 1468 if (tvp) 1469 vrele(tvp); 1470 return error; 1471 } 1472 1473 /* --------------------------------------------------------------------- */ 1474 1475 static int 1476 tmpfs_nmkdir(struct vop_nmkdir_args *ap) 1477 { 1478 struct vnode *dvp = ap->a_dvp; 1479 struct vnode **vpp = ap->a_vpp; 1480 struct namecache *ncp = ap->a_nch->ncp; 1481 struct vattr *vap = ap->a_vap; 1482 struct ucred *cred = ap->a_cred; 1483 int error; 1484 1485 KKASSERT(vap->va_type == VDIR); 1486 1487 error = tmpfs_alloc_file(dvp, vpp, vap, ncp, cred, NULL); 1488 if (error == 0) { 1489 cache_setunresolved(ap->a_nch); 1490 cache_setvp(ap->a_nch, *vpp); 1491 tmpfs_knote(dvp, NOTE_WRITE | NOTE_LINK); 1492 } 1493 return error; 1494 } 1495 1496 /* --------------------------------------------------------------------- */ 1497 1498 static int 1499 tmpfs_nrmdir(struct vop_nrmdir_args *ap) 1500 { 1501 struct vnode *dvp = ap->a_dvp; 1502 struct namecache *ncp = ap->a_nch->ncp; 1503 struct vnode *vp; 1504 struct tmpfs_dirent *de; 1505 struct tmpfs_mount *tmp; 1506 struct tmpfs_node *dnode; 1507 struct tmpfs_node *node; 1508 int error; 1509 1510 /* 1511 * We have to acquire the vp from ap->a_nch because we will likely 1512 * unresolve the namecache entry, and a vrele/vput is needed to 1513 * trigger the tmpfs_inactive/tmpfs_reclaim sequence. 1514 * 1515 * We have to use vget to clear any inactive state on the vnode, 1516 * otherwise the vnode may remain inactive and thus tmpfs_inactive 1517 * will not get called when we release it. 1518 */ 1519 error = cache_vget(ap->a_nch, ap->a_cred, LK_SHARED, &vp); 1520 KKASSERT(error == 0); 1521 vn_unlock(vp); 1522 1523 /* 1524 * Prevalidate so we don't hit an assertion later 1525 */ 1526 if (vp->v_type != VDIR) { 1527 error = ENOTDIR; 1528 goto out; 1529 } 1530 1531 tmp = VFS_TO_TMPFS(dvp->v_mount); 1532 dnode = VP_TO_TMPFS_DIR(dvp); 1533 node = VP_TO_TMPFS_DIR(vp); 1534 1535 /* 1536 * 1537 */ 1538 TMPFS_NODE_LOCK(dnode); 1539 TMPFS_NODE_LOCK(node); 1540 1541 /* 1542 * Only empty directories can be removed. 1543 */ 1544 if (node->tn_size > 0) { 1545 error = ENOTEMPTY; 1546 goto out_locked; 1547 } 1548 1549 if ((dnode->tn_flags & APPEND) 1550 || (node->tn_flags & (NOUNLINK | IMMUTABLE | APPEND))) { 1551 error = EPERM; 1552 goto out_locked; 1553 } 1554 1555 /* 1556 * This invariant holds only if we are not trying to 1557 * remove "..". We checked for that above so this is safe now. 1558 */ 1559 KKASSERT(node->tn_dir.tn_parent == dnode); 1560 1561 /* 1562 * Get the directory entry associated with node (vp) 1563 */ 1564 de = tmpfs_dir_lookup(dnode, node, ncp); 1565 KKASSERT(TMPFS_DIRENT_MATCHES(de, ncp->nc_name, ncp->nc_nlen)); 1566 1567 /* Check flags to see if we are allowed to remove the directory. */ 1568 if ((dnode->tn_flags & APPEND) || 1569 node->tn_flags & (NOUNLINK | IMMUTABLE | APPEND)) { 1570 error = EPERM; 1571 goto out_locked; 1572 } 1573 1574 /* Detach the directory entry from the directory (dnode). */ 1575 tmpfs_dir_detach_locked(dnode, de); 1576 1577 /* 1578 * Must set parent linkage to NULL (tested by ncreate to disallow 1579 * the creation of new files/dirs in a deleted directory) 1580 */ 1581 node->tn_status |= TMPFS_NODE_CHANGED; 1582 1583 dnode->tn_status |= TMPFS_NODE_ACCESSED | TMPFS_NODE_CHANGED | 1584 TMPFS_NODE_MODIFIED; 1585 1586 /* Free the directory entry we just deleted. Note that the node 1587 * referred by it will not be removed until the vnode is really 1588 * reclaimed. */ 1589 tmpfs_free_dirent(tmp, de); 1590 1591 /* Release the deleted vnode (will destroy the node, notify 1592 * interested parties and clean it from the cache). */ 1593 1594 dnode->tn_status |= TMPFS_NODE_CHANGED; 1595 1596 TMPFS_NODE_UNLOCK(node); 1597 TMPFS_NODE_UNLOCK(dnode); 1598 1599 tmpfs_update(dvp); 1600 cache_unlink(ap->a_nch); 1601 tmpfs_knote(dvp, NOTE_WRITE | NOTE_LINK); 1602 vrele(vp); 1603 return 0; 1604 1605 out_locked: 1606 TMPFS_NODE_UNLOCK(node); 1607 TMPFS_NODE_UNLOCK(dnode); 1608 1609 out: 1610 vrele(vp); 1611 1612 return error; 1613 } 1614 1615 /* --------------------------------------------------------------------- */ 1616 1617 static int 1618 tmpfs_nsymlink(struct vop_nsymlink_args *ap) 1619 { 1620 struct vnode *dvp = ap->a_dvp; 1621 struct vnode **vpp = ap->a_vpp; 1622 struct namecache *ncp = ap->a_nch->ncp; 1623 struct vattr *vap = ap->a_vap; 1624 struct ucred *cred = ap->a_cred; 1625 char *target = ap->a_target; 1626 int error; 1627 1628 vap->va_type = VLNK; 1629 error = tmpfs_alloc_file(dvp, vpp, vap, ncp, cred, target); 1630 if (error == 0) { 1631 tmpfs_knote(*vpp, NOTE_WRITE); 1632 cache_setunresolved(ap->a_nch); 1633 cache_setvp(ap->a_nch, *vpp); 1634 } 1635 return error; 1636 } 1637 1638 /* --------------------------------------------------------------------- */ 1639 1640 static int 1641 tmpfs_readdir(struct vop_readdir_args *ap) 1642 { 1643 struct vnode *vp = ap->a_vp; 1644 struct uio *uio = ap->a_uio; 1645 int *eofflag = ap->a_eofflag; 1646 off_t **cookies = ap->a_cookies; 1647 int *ncookies = ap->a_ncookies; 1648 struct tmpfs_mount *tmp; 1649 int error; 1650 off_t startoff; 1651 off_t cnt = 0; 1652 struct tmpfs_node *node; 1653 1654 /* This operation only makes sense on directory nodes. */ 1655 if (vp->v_type != VDIR) { 1656 return ENOTDIR; 1657 } 1658 1659 tmp = VFS_TO_TMPFS(vp->v_mount); 1660 node = VP_TO_TMPFS_DIR(vp); 1661 startoff = uio->uio_offset; 1662 1663 if (uio->uio_offset == TMPFS_DIRCOOKIE_DOT) { 1664 error = tmpfs_dir_getdotdent(node, uio); 1665 if (error && error != EINVAL) { 1666 TMPFS_NODE_LOCK_SH(node); 1667 goto outok; 1668 } 1669 cnt++; 1670 } 1671 1672 if (uio->uio_offset == TMPFS_DIRCOOKIE_DOTDOT) { 1673 /* may lock parent, cannot hold node lock */ 1674 error = tmpfs_dir_getdotdotdent(tmp, node, uio); 1675 if (error && error != EINVAL) { 1676 TMPFS_NODE_LOCK_SH(node); 1677 goto outok; 1678 } 1679 cnt++; 1680 } 1681 1682 TMPFS_NODE_LOCK_SH(node); 1683 error = tmpfs_dir_getdents(node, uio, &cnt); 1684 1685 outok: 1686 KKASSERT(error >= -1); 1687 1688 if (error == -1) 1689 error = 0; 1690 1691 if (eofflag != NULL) 1692 *eofflag = 1693 (error == 0 && uio->uio_offset == TMPFS_DIRCOOKIE_EOF); 1694 1695 /* Update NFS-related variables. */ 1696 if (error == 0 && cookies != NULL && ncookies != NULL) { 1697 off_t i; 1698 off_t off = startoff; 1699 struct tmpfs_dirent *de = NULL; 1700 1701 *ncookies = cnt; 1702 *cookies = kmalloc(cnt * sizeof(off_t), M_TEMP, M_WAITOK); 1703 1704 for (i = 0; i < cnt; i++) { 1705 KKASSERT(off != TMPFS_DIRCOOKIE_EOF); 1706 if (off == TMPFS_DIRCOOKIE_DOT) { 1707 off = TMPFS_DIRCOOKIE_DOTDOT; 1708 } else { 1709 if (off == TMPFS_DIRCOOKIE_DOTDOT) { 1710 de = RB_MIN(tmpfs_dirtree_cookie, 1711 &node->tn_dir.tn_cookietree); 1712 } else if (de != NULL) { 1713 de = RB_NEXT(tmpfs_dirtree_cookie, 1714 &node->tn_dir.tn_cookietree, de); 1715 } else { 1716 de = tmpfs_dir_lookupbycookie(node, off, 1717 1); 1718 KKASSERT(de != NULL); 1719 de = RB_NEXT(tmpfs_dirtree_cookie, 1720 &node->tn_dir.tn_cookietree, de); 1721 } 1722 if (de == NULL) 1723 off = TMPFS_DIRCOOKIE_EOF; 1724 else 1725 off = tmpfs_dircookie(de); 1726 } 1727 (*cookies)[i] = off; 1728 } 1729 KKASSERT(uio->uio_offset == off); 1730 } 1731 TMPFS_NODE_UNLOCK(node); 1732 1733 if ((node->tn_status & TMPFS_NODE_ACCESSED) == 0) { 1734 TMPFS_NODE_LOCK(node); 1735 node->tn_status |= TMPFS_NODE_ACCESSED; 1736 TMPFS_NODE_UNLOCK(node); 1737 } 1738 return error; 1739 } 1740 1741 /* --------------------------------------------------------------------- */ 1742 1743 static int 1744 tmpfs_readlink(struct vop_readlink_args *ap) 1745 { 1746 struct vnode *vp = ap->a_vp; 1747 struct uio *uio = ap->a_uio; 1748 int error; 1749 struct tmpfs_node *node; 1750 1751 KKASSERT(uio->uio_offset == 0); 1752 KKASSERT(vp->v_type == VLNK); 1753 1754 node = VP_TO_TMPFS_NODE(vp); 1755 TMPFS_NODE_LOCK_SH(node); 1756 error = uiomove(node->tn_link, 1757 MIN(node->tn_size, uio->uio_resid), uio); 1758 TMPFS_NODE_UNLOCK(node); 1759 if ((node->tn_status & TMPFS_NODE_ACCESSED) == 0) { 1760 TMPFS_NODE_LOCK(node); 1761 node->tn_status |= TMPFS_NODE_ACCESSED; 1762 TMPFS_NODE_UNLOCK(node); 1763 } 1764 return error; 1765 } 1766 1767 /* --------------------------------------------------------------------- */ 1768 1769 static int 1770 tmpfs_inactive(struct vop_inactive_args *ap) 1771 { 1772 struct vnode *vp = ap->a_vp; 1773 struct tmpfs_node *node; 1774 struct mount *mp; 1775 1776 mp = vp->v_mount; 1777 lwkt_gettoken(&mp->mnt_token); 1778 node = VP_TO_TMPFS_NODE(vp); 1779 1780 /* 1781 * Degenerate case 1782 */ 1783 if (node == NULL) { 1784 vrecycle(vp); 1785 lwkt_reltoken(&mp->mnt_token); 1786 return(0); 1787 } 1788 1789 /* 1790 * Get rid of unreferenced deleted vnodes sooner rather than 1791 * later so the data memory can be recovered immediately. 1792 * 1793 * We must truncate the vnode to prevent the normal reclamation 1794 * path from flushing the data for the removed file to disk. 1795 */ 1796 TMPFS_NODE_LOCK(node); 1797 if (node->tn_links == 0) { 1798 node->tn_vpstate = TMPFS_VNODE_DOOMED; 1799 TMPFS_NODE_UNLOCK(node); 1800 if (node->tn_type == VREG) 1801 tmpfs_truncate(vp, 0); 1802 vrecycle(vp); 1803 } else { 1804 /* 1805 * We must retain any VM pages belonging to the vnode's 1806 * object as the vnode will destroy the object during a 1807 * later reclaim. We call vinvalbuf(V_SAVE) to clean 1808 * out the buffer cache. 1809 * 1810 * On DragonFlyBSD, vnodes are not immediately deactivated 1811 * on the 1->0 refs, so this is a relatively optimal 1812 * operation. We have to do this in tmpfs_inactive() 1813 * because the pages will have already been thrown away 1814 * at the time tmpfs_reclaim() is called. 1815 */ 1816 if (node->tn_type == VREG && 1817 node->tn_reg.tn_pages_in_aobj == 0) { 1818 vinvalbuf(vp, V_SAVE, 0, 0); 1819 KKASSERT(RB_EMPTY(&vp->v_rbdirty_tree)); 1820 KKASSERT(RB_EMPTY(&vp->v_rbclean_tree)); 1821 tmpfs_move_pages(vp->v_object, node->tn_reg.tn_aobj, 1822 TMPFS_MOVF_DEACTIVATE); 1823 node->tn_reg.tn_pages_in_aobj = 1; 1824 } 1825 1826 TMPFS_NODE_UNLOCK(node); 1827 } 1828 lwkt_reltoken(&mp->mnt_token); 1829 1830 return 0; 1831 } 1832 1833 /* --------------------------------------------------------------------- */ 1834 1835 int 1836 tmpfs_reclaim(struct vop_reclaim_args *ap) 1837 { 1838 struct vnode *vp = ap->a_vp; 1839 struct tmpfs_mount *tmp; 1840 struct tmpfs_node *node; 1841 struct mount *mp; 1842 1843 mp = vp->v_mount; 1844 lwkt_gettoken(&mp->mnt_token); 1845 1846 node = VP_TO_TMPFS_NODE(vp); 1847 tmp = VFS_TO_TMPFS(vp->v_mount); 1848 KKASSERT(mp == tmp->tm_mount); 1849 1850 TMPFS_NODE_LOCK(node); 1851 KKASSERT(node->tn_vnode == vp); 1852 node->tn_vnode = NULL; 1853 vp->v_data = NULL; 1854 1855 /* 1856 * If the node referenced by this vnode was deleted by the 1857 * user, we must free its associated data structures now that 1858 * the vnode is being reclaimed. 1859 * 1860 * Directories have an extra link ref. 1861 */ 1862 if (node->tn_links == 0) { 1863 node->tn_vpstate = TMPFS_VNODE_DOOMED; 1864 tmpfs_free_node(tmp, node); 1865 /* eats the lock */ 1866 } else { 1867 TMPFS_NODE_UNLOCK(node); 1868 } 1869 lwkt_reltoken(&mp->mnt_token); 1870 1871 KKASSERT(vp->v_data == NULL); 1872 return 0; 1873 } 1874 1875 /* --------------------------------------------------------------------- */ 1876 1877 static int 1878 tmpfs_mountctl(struct vop_mountctl_args *ap) 1879 { 1880 struct tmpfs_mount *tmp; 1881 struct mount *mp; 1882 int rc; 1883 1884 mp = ap->a_head.a_ops->head.vv_mount; 1885 lwkt_gettoken(&mp->mnt_token); 1886 1887 switch (ap->a_op) { 1888 case (MOUNTCTL_SET_EXPORT): 1889 tmp = (struct tmpfs_mount *) mp->mnt_data; 1890 1891 if (ap->a_ctllen != sizeof(struct export_args)) 1892 rc = (EINVAL); 1893 else 1894 rc = vfs_export(mp, &tmp->tm_export, 1895 (const struct export_args *) ap->a_ctl); 1896 break; 1897 default: 1898 rc = vop_stdmountctl(ap); 1899 break; 1900 } 1901 1902 lwkt_reltoken(&mp->mnt_token); 1903 return (rc); 1904 } 1905 1906 /* --------------------------------------------------------------------- */ 1907 1908 static int 1909 tmpfs_print(struct vop_print_args *ap) 1910 { 1911 struct vnode *vp = ap->a_vp; 1912 1913 struct tmpfs_node *node; 1914 1915 node = VP_TO_TMPFS_NODE(vp); 1916 1917 kprintf("tag VT_TMPFS, tmpfs_node %p, flags 0x%x, links %d\n", 1918 node, node->tn_flags, node->tn_links); 1919 kprintf("\tmode 0%o, owner %d, group %d, size %ju, status 0x%x\n", 1920 node->tn_mode, node->tn_uid, node->tn_gid, 1921 (uintmax_t)node->tn_size, node->tn_status); 1922 1923 if (vp->v_type == VFIFO) 1924 fifo_printinfo(vp); 1925 1926 kprintf("\n"); 1927 1928 return 0; 1929 } 1930 1931 /* --------------------------------------------------------------------- */ 1932 1933 static int 1934 tmpfs_pathconf(struct vop_pathconf_args *ap) 1935 { 1936 struct vnode *vp = ap->a_vp; 1937 int name = ap->a_name; 1938 register_t *retval = ap->a_retval; 1939 struct tmpfs_mount *tmp; 1940 int error; 1941 1942 error = 0; 1943 1944 switch (name) { 1945 case _PC_CHOWN_RESTRICTED: 1946 *retval = 1; 1947 break; 1948 1949 case _PC_FILESIZEBITS: 1950 tmp = VFS_TO_TMPFS(vp->v_mount); 1951 *retval = max(32, flsll(tmp->tm_pages_max * PAGE_SIZE) + 1); 1952 break; 1953 1954 case _PC_LINK_MAX: 1955 *retval = LINK_MAX; 1956 break; 1957 1958 case _PC_NAME_MAX: 1959 *retval = NAME_MAX; 1960 break; 1961 1962 case _PC_NO_TRUNC: 1963 *retval = 1; 1964 break; 1965 1966 case _PC_PATH_MAX: 1967 *retval = PATH_MAX; 1968 break; 1969 1970 case _PC_PIPE_BUF: 1971 *retval = PIPE_BUF; 1972 break; 1973 1974 case _PC_SYNC_IO: 1975 *retval = 1; 1976 break; 1977 1978 case _PC_2_SYMLINKS: 1979 *retval = 1; 1980 break; 1981 1982 default: 1983 error = EINVAL; 1984 } 1985 1986 return error; 1987 } 1988 1989 /************************************************************************ 1990 * KQFILTER OPS * 1991 ************************************************************************/ 1992 1993 static void filt_tmpfsdetach(struct knote *kn); 1994 static int filt_tmpfsread(struct knote *kn, long hint); 1995 static int filt_tmpfswrite(struct knote *kn, long hint); 1996 static int filt_tmpfsvnode(struct knote *kn, long hint); 1997 1998 static struct filterops tmpfsread_filtops = 1999 { FILTEROP_ISFD | FILTEROP_MPSAFE, 2000 NULL, filt_tmpfsdetach, filt_tmpfsread }; 2001 static struct filterops tmpfswrite_filtops = 2002 { FILTEROP_ISFD | FILTEROP_MPSAFE, 2003 NULL, filt_tmpfsdetach, filt_tmpfswrite }; 2004 static struct filterops tmpfsvnode_filtops = 2005 { FILTEROP_ISFD | FILTEROP_MPSAFE, 2006 NULL, filt_tmpfsdetach, filt_tmpfsvnode }; 2007 2008 static int 2009 tmpfs_kqfilter (struct vop_kqfilter_args *ap) 2010 { 2011 struct vnode *vp = ap->a_vp; 2012 struct knote *kn = ap->a_kn; 2013 2014 switch (kn->kn_filter) { 2015 case EVFILT_READ: 2016 kn->kn_fop = &tmpfsread_filtops; 2017 break; 2018 case EVFILT_WRITE: 2019 kn->kn_fop = &tmpfswrite_filtops; 2020 break; 2021 case EVFILT_VNODE: 2022 kn->kn_fop = &tmpfsvnode_filtops; 2023 break; 2024 default: 2025 return (EOPNOTSUPP); 2026 } 2027 2028 kn->kn_hook = (caddr_t)vp; 2029 2030 knote_insert(&vp->v_pollinfo.vpi_kqinfo.ki_note, kn); 2031 2032 return(0); 2033 } 2034 2035 static void 2036 filt_tmpfsdetach(struct knote *kn) 2037 { 2038 struct vnode *vp = (void *)kn->kn_hook; 2039 2040 knote_remove(&vp->v_pollinfo.vpi_kqinfo.ki_note, kn); 2041 } 2042 2043 static int 2044 filt_tmpfsread(struct knote *kn, long hint) 2045 { 2046 struct vnode *vp = (void *)kn->kn_hook; 2047 struct tmpfs_node *node = VP_TO_TMPFS_NODE(vp); 2048 off_t off; 2049 2050 if (hint == NOTE_REVOKE) { 2051 kn->kn_flags |= (EV_EOF | EV_NODATA | EV_ONESHOT); 2052 return(1); 2053 } 2054 2055 /* 2056 * Interlock against MP races when performing this function. 2057 */ 2058 TMPFS_NODE_LOCK_SH(node); 2059 off = node->tn_size - kn->kn_fp->f_offset; 2060 kn->kn_data = (off < INTPTR_MAX) ? off : INTPTR_MAX; 2061 if (kn->kn_sfflags & NOTE_OLDAPI) { 2062 TMPFS_NODE_UNLOCK(node); 2063 return(1); 2064 } 2065 if (kn->kn_data == 0) { 2066 kn->kn_data = (off < INTPTR_MAX) ? off : INTPTR_MAX; 2067 } 2068 TMPFS_NODE_UNLOCK(node); 2069 return (kn->kn_data != 0); 2070 } 2071 2072 static int 2073 filt_tmpfswrite(struct knote *kn, long hint) 2074 { 2075 if (hint == NOTE_REVOKE) 2076 kn->kn_flags |= (EV_EOF | EV_NODATA | EV_ONESHOT); 2077 kn->kn_data = 0; 2078 return (1); 2079 } 2080 2081 static int 2082 filt_tmpfsvnode(struct knote *kn, long hint) 2083 { 2084 if (kn->kn_sfflags & hint) 2085 kn->kn_fflags |= hint; 2086 if (hint == NOTE_REVOKE) { 2087 kn->kn_flags |= (EV_EOF | EV_NODATA); 2088 return (1); 2089 } 2090 return (kn->kn_fflags != 0); 2091 } 2092 2093 /* 2094 * Helper to move VM pages between objects 2095 * 2096 * NOTE: The vm_page_rename() dirties the page, so we can clear the 2097 * PG_NEED_COMMIT flag. If the pages are being moved into tn_aobj, 2098 * the pageout daemon will be able to page them out. 2099 */ 2100 static int 2101 tmpfs_move_pages_callback(vm_page_t p, void *data) 2102 { 2103 struct rb_vm_page_scan_info *info = data; 2104 vm_pindex_t pindex; 2105 2106 /* 2107 * Take control of the page 2108 */ 2109 pindex = p->pindex; 2110 if (vm_page_busy_try(p, TRUE)) { 2111 vm_page_sleep_busy(p, TRUE, "tpgmov"); 2112 info->error = -1; 2113 return -1; 2114 } 2115 if (p->object != info->object || p->pindex != pindex) { 2116 vm_page_wakeup(p); 2117 info->error = -1; 2118 return -1; 2119 } 2120 2121 /* 2122 * Make sure the page is not mapped. These flags might also still be 2123 * set heuristically even if we know the page is not mapped and must 2124 * be properly cleaned up. 2125 */ 2126 if (__predict_false((p->flags & (PG_MAPPED|PG_WRITEABLE)) != 0)) 2127 vm_page_protect(p, VM_PROT_NONE); 2128 2129 /* 2130 * Free or rename the page as appropriate 2131 */ 2132 if ((info->pagerflags & TMPFS_MOVF_FROMBACKING) && 2133 (p->flags & PG_SWAPPED) && 2134 (p->flags & PG_NEED_COMMIT) == 0 && 2135 p->dirty == 0) { 2136 /* 2137 * If the page in the backing aobj was paged out to swap 2138 * it will be clean and it is better to free it rather 2139 * than re-dirty it. We will assume that the page was 2140 * paged out to swap for a reason! 2141 * 2142 * This helps avoid unnecessary swap thrashing on the page. 2143 */ 2144 vm_page_free(p); 2145 } else if ((info->pagerflags & TMPFS_MOVF_FROMBACKING) == 0 && 2146 (p->flags & PG_NEED_COMMIT) == 0 && 2147 p->dirty == 0) { 2148 /* 2149 * If the page associated with the vnode was cleaned via 2150 * a tmpfs_strategy() call, it exists as a swap block in 2151 * aobj and it is again better to free it rather than 2152 * re-dirty it. We will assume that the page was 2153 * paged out to swap for a reason! 2154 * 2155 * This helps avoid unnecessary swap thrashing on the page. 2156 */ 2157 vm_page_free(p); 2158 } else { 2159 /* 2160 * Rename the page, which will also ensure that it is flagged 2161 * as dirty and check whether a swap block association exists 2162 * in the target object or not, setting appropriate flags if 2163 * it does. 2164 */ 2165 vm_page_rename(p, info->dest_object, pindex); 2166 vm_page_clear_commit(p); 2167 if (info->pagerflags & TMPFS_MOVF_DEACTIVATE) 2168 vm_page_deactivate(p); 2169 vm_page_wakeup(p); 2170 /* page automaticaly made dirty */ 2171 } 2172 2173 return 0; 2174 } 2175 2176 static 2177 void 2178 tmpfs_move_pages(vm_object_t src, vm_object_t dst, int movflags) 2179 { 2180 struct rb_vm_page_scan_info info; 2181 2182 vm_object_hold(src); 2183 vm_object_hold(dst); 2184 info.object = src; 2185 info.dest_object = dst; 2186 info.pagerflags = movflags; 2187 do { 2188 if (src->paging_in_progress) 2189 vm_object_pip_wait(src, "objtfs"); 2190 info.error = 1; 2191 vm_page_rb_tree_RB_SCAN(&src->rb_memq, NULL, 2192 tmpfs_move_pages_callback, &info); 2193 } while (info.error < 0 || !RB_EMPTY(&src->rb_memq) || 2194 src->paging_in_progress); 2195 vm_object_drop(dst); 2196 vm_object_drop(src); 2197 } 2198 2199 /* --------------------------------------------------------------------- */ 2200 2201 /* 2202 * vnode operations vector used for files stored in a tmpfs file system. 2203 */ 2204 struct vop_ops tmpfs_vnode_vops = { 2205 .vop_default = vop_defaultop, 2206 .vop_getpages = vop_stdgetpages, 2207 .vop_putpages = vop_stdputpages, 2208 .vop_ncreate = tmpfs_ncreate, 2209 .vop_nresolve = tmpfs_nresolve, 2210 .vop_nlookupdotdot = tmpfs_nlookupdotdot, 2211 .vop_nmknod = tmpfs_nmknod, 2212 .vop_open = tmpfs_open, 2213 .vop_close = tmpfs_close, 2214 .vop_access = tmpfs_access, 2215 .vop_getattr = tmpfs_getattr, 2216 .vop_getattr_lite = tmpfs_getattr_lite, 2217 .vop_setattr = tmpfs_setattr, 2218 .vop_read = tmpfs_read, 2219 .vop_write = tmpfs_write, 2220 .vop_fsync = tmpfs_fsync, 2221 .vop_mountctl = tmpfs_mountctl, 2222 .vop_nremove = tmpfs_nremove, 2223 .vop_nlink = tmpfs_nlink, 2224 .vop_nrename = tmpfs_nrename, 2225 .vop_nmkdir = tmpfs_nmkdir, 2226 .vop_nrmdir = tmpfs_nrmdir, 2227 .vop_nsymlink = tmpfs_nsymlink, 2228 .vop_readdir = tmpfs_readdir, 2229 .vop_readlink = tmpfs_readlink, 2230 .vop_inactive = tmpfs_inactive, 2231 .vop_reclaim = tmpfs_reclaim, 2232 .vop_print = tmpfs_print, 2233 .vop_pathconf = tmpfs_pathconf, 2234 .vop_bmap = tmpfs_bmap, 2235 .vop_strategy = tmpfs_strategy, 2236 .vop_advlock = tmpfs_advlock, 2237 .vop_kqfilter = tmpfs_kqfilter 2238 }; 2239