1#include <linux/ceph/ceph_debug.h> 2 3#include <linux/module.h> 4#include <linux/sched.h> 5#include <linux/slab.h> 6#include <linux/file.h> 7#include <linux/mount.h> 8#include <linux/namei.h> 9#include <linux/writeback.h> 10#include <linux/falloc.h> 11 12#include "super.h" 13#include "mds_client.h" 14#include "cache.h" 15 16/* 17 * Ceph file operations 18 * 19 * Implement basic open/close functionality, and implement 20 * read/write. 21 * 22 * We implement three modes of file I/O: 23 * - buffered uses the generic_file_aio_{read,write} helpers 24 * 25 * - synchronous is used when there is multi-client read/write 26 * sharing, avoids the page cache, and synchronously waits for an 27 * ack from the OSD. 28 * 29 * - direct io takes the variant of the sync path that references 30 * user pages directly. 31 * 32 * fsync() flushes and waits on dirty pages, but just queues metadata 33 * for writeback: since the MDS can recover size and mtime there is no 34 * need to wait for MDS acknowledgement. 35 */ 36 37 38/* 39 * Prepare an open request. Preallocate ceph_cap to avoid an 40 * inopportune ENOMEM later. 41 */ 42static struct ceph_mds_request * 43prepare_open_request(struct super_block *sb, int flags, int create_mode) 44{ 45 struct ceph_fs_client *fsc = ceph_sb_to_client(sb); 46 struct ceph_mds_client *mdsc = fsc->mdsc; 47 struct ceph_mds_request *req; 48 int want_auth = USE_ANY_MDS; 49 int op = (flags & O_CREAT) ? CEPH_MDS_OP_CREATE : CEPH_MDS_OP_OPEN; 50 51 if (flags & (O_WRONLY|O_RDWR|O_CREAT|O_TRUNC)) 52 want_auth = USE_AUTH_MDS; 53 54 req = ceph_mdsc_create_request(mdsc, op, want_auth); 55 if (IS_ERR(req)) 56 goto out; 57 req->r_fmode = ceph_flags_to_mode(flags); 58 req->r_args.open.flags = cpu_to_le32(flags); 59 req->r_args.open.mode = cpu_to_le32(create_mode); 60out: 61 return req; 62} 63 64/* 65 * initialize private struct file data. 66 * if we fail, clean up by dropping fmode reference on the ceph_inode 67 */ 68static int ceph_init_file(struct inode *inode, struct file *file, int fmode) 69{ 70 struct ceph_file_info *cf; 71 int ret = 0; 72 struct ceph_inode_info *ci = ceph_inode(inode); 73 struct ceph_fs_client *fsc = ceph_sb_to_client(inode->i_sb); 74 struct ceph_mds_client *mdsc = fsc->mdsc; 75 76 switch (inode->i_mode & S_IFMT) { 77 case S_IFREG: 78 /* First file open request creates the cookie, we want to keep 79 * this cookie around for the filetime of the inode as not to 80 * have to worry about fscache register / revoke / operation 81 * races. 82 * 83 * Also, if we know the operation is going to invalidate data 84 * (non readonly) just nuke the cache right away. 85 */ 86 ceph_fscache_register_inode_cookie(mdsc->fsc, ci); 87 if ((fmode & CEPH_FILE_MODE_WR)) 88 ceph_fscache_invalidate(inode); 89 case S_IFDIR: 90 dout("init_file %p %p 0%o (regular)\n", inode, file, 91 inode->i_mode); 92 cf = kmem_cache_alloc(ceph_file_cachep, GFP_NOFS | __GFP_ZERO); 93 if (cf == NULL) { 94 ceph_put_fmode(ceph_inode(inode), fmode); /* clean up */ 95 return -ENOMEM; 96 } 97 cf->fmode = fmode; 98 cf->next_offset = 2; 99 file->private_data = cf; 100 BUG_ON(inode->i_fop->release != ceph_release); 101 break; 102 103 case S_IFLNK: 104 dout("init_file %p %p 0%o (symlink)\n", inode, file, 105 inode->i_mode); 106 ceph_put_fmode(ceph_inode(inode), fmode); /* clean up */ 107 break; 108 109 default: 110 dout("init_file %p %p 0%o (special)\n", inode, file, 111 inode->i_mode); 112 /* 113 * we need to drop the open ref now, since we don't 114 * have .release set to ceph_release. 115 */ 116 ceph_put_fmode(ceph_inode(inode), fmode); /* clean up */ 117 BUG_ON(inode->i_fop->release == ceph_release); 118 119 /* call the proper open fop */ 120 ret = inode->i_fop->open(inode, file); 121 } 122 return ret; 123} 124 125/* 126 * If we already have the requisite capabilities, we can satisfy 127 * the open request locally (no need to request new caps from the 128 * MDS). We do, however, need to inform the MDS (asynchronously) 129 * if our wanted caps set expands. 130 */ 131int ceph_open(struct inode *inode, struct file *file) 132{ 133 struct ceph_inode_info *ci = ceph_inode(inode); 134 struct ceph_fs_client *fsc = ceph_sb_to_client(inode->i_sb); 135 struct ceph_mds_client *mdsc = fsc->mdsc; 136 struct ceph_mds_request *req; 137 struct ceph_file_info *cf = file->private_data; 138 struct inode *parent_inode = NULL; 139 int err; 140 int flags, fmode, wanted; 141 142 if (cf) { 143 dout("open file %p is already opened\n", file); 144 return 0; 145 } 146 147 /* filter out O_CREAT|O_EXCL; vfs did that already. yuck. */ 148 flags = file->f_flags & ~(O_CREAT|O_EXCL); 149 if (S_ISDIR(inode->i_mode)) 150 flags = O_DIRECTORY; /* mds likes to know */ 151 152 dout("open inode %p ino %llx.%llx file %p flags %d (%d)\n", inode, 153 ceph_vinop(inode), file, flags, file->f_flags); 154 fmode = ceph_flags_to_mode(flags); 155 wanted = ceph_caps_for_mode(fmode); 156 157 /* snapped files are read-only */ 158 if (ceph_snap(inode) != CEPH_NOSNAP && (file->f_mode & FMODE_WRITE)) 159 return -EROFS; 160 161 /* trivially open snapdir */ 162 if (ceph_snap(inode) == CEPH_SNAPDIR) { 163 spin_lock(&ci->i_ceph_lock); 164 __ceph_get_fmode(ci, fmode); 165 spin_unlock(&ci->i_ceph_lock); 166 return ceph_init_file(inode, file, fmode); 167 } 168 169 /* 170 * No need to block if we have caps on the auth MDS (for 171 * write) or any MDS (for read). Update wanted set 172 * asynchronously. 173 */ 174 spin_lock(&ci->i_ceph_lock); 175 if (__ceph_is_any_real_caps(ci) && 176 (((fmode & CEPH_FILE_MODE_WR) == 0) || ci->i_auth_cap)) { 177 int mds_wanted = __ceph_caps_mds_wanted(ci); 178 int issued = __ceph_caps_issued(ci, NULL); 179 180 dout("open %p fmode %d want %s issued %s using existing\n", 181 inode, fmode, ceph_cap_string(wanted), 182 ceph_cap_string(issued)); 183 __ceph_get_fmode(ci, fmode); 184 spin_unlock(&ci->i_ceph_lock); 185 186 /* adjust wanted? */ 187 if ((issued & wanted) != wanted && 188 (mds_wanted & wanted) != wanted && 189 ceph_snap(inode) != CEPH_SNAPDIR) 190 ceph_check_caps(ci, 0, NULL); 191 192 return ceph_init_file(inode, file, fmode); 193 } else if (ceph_snap(inode) != CEPH_NOSNAP && 194 (ci->i_snap_caps & wanted) == wanted) { 195 __ceph_get_fmode(ci, fmode); 196 spin_unlock(&ci->i_ceph_lock); 197 return ceph_init_file(inode, file, fmode); 198 } 199 200 spin_unlock(&ci->i_ceph_lock); 201 202 dout("open fmode %d wants %s\n", fmode, ceph_cap_string(wanted)); 203 req = prepare_open_request(inode->i_sb, flags, 0); 204 if (IS_ERR(req)) { 205 err = PTR_ERR(req); 206 goto out; 207 } 208 req->r_inode = inode; 209 ihold(inode); 210 211 req->r_num_caps = 1; 212 if (flags & O_CREAT) 213 parent_inode = ceph_get_dentry_parent_inode(file->f_path.dentry); 214 err = ceph_mdsc_do_request(mdsc, parent_inode, req); 215 iput(parent_inode); 216 if (!err) 217 err = ceph_init_file(inode, file, req->r_fmode); 218 ceph_mdsc_put_request(req); 219 dout("open result=%d on %llx.%llx\n", err, ceph_vinop(inode)); 220out: 221 return err; 222} 223 224 225/* 226 * Do a lookup + open with a single request. If we get a non-existent 227 * file or symlink, return 1 so the VFS can retry. 228 */ 229int ceph_atomic_open(struct inode *dir, struct dentry *dentry, 230 struct file *file, unsigned flags, umode_t mode, 231 int *opened) 232{ 233 struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb); 234 struct ceph_mds_client *mdsc = fsc->mdsc; 235 struct ceph_mds_request *req; 236 struct dentry *dn; 237 struct ceph_acls_info acls = {}; 238 int err; 239 240 dout("atomic_open %p dentry %p '%pd' %s flags %d mode 0%o\n", 241 dir, dentry, dentry, 242 d_unhashed(dentry) ? "unhashed" : "hashed", flags, mode); 243 244 if (dentry->d_name.len > NAME_MAX) 245 return -ENAMETOOLONG; 246 247 err = ceph_init_dentry(dentry); 248 if (err < 0) 249 return err; 250 251 if (flags & O_CREAT) { 252 err = ceph_pre_init_acls(dir, &mode, &acls); 253 if (err < 0) 254 return err; 255 } 256 257 /* do the open */ 258 req = prepare_open_request(dir->i_sb, flags, mode); 259 if (IS_ERR(req)) { 260 err = PTR_ERR(req); 261 goto out_acl; 262 } 263 req->r_dentry = dget(dentry); 264 req->r_num_caps = 2; 265 if (flags & O_CREAT) { 266 req->r_dentry_drop = CEPH_CAP_FILE_SHARED; 267 req->r_dentry_unless = CEPH_CAP_FILE_EXCL; 268 if (acls.pagelist) { 269 req->r_pagelist = acls.pagelist; 270 acls.pagelist = NULL; 271 } 272 } 273 req->r_locked_dir = dir; /* caller holds dir->i_mutex */ 274 err = ceph_mdsc_do_request(mdsc, 275 (flags & (O_CREAT|O_TRUNC)) ? dir : NULL, 276 req); 277 err = ceph_handle_snapdir(req, dentry, err); 278 if (err) 279 goto out_req; 280 281 if (err == 0 && (flags & O_CREAT) && !req->r_reply_info.head->is_dentry) 282 err = ceph_handle_notrace_create(dir, dentry); 283 284 if (d_unhashed(dentry)) { 285 dn = ceph_finish_lookup(req, dentry, err); 286 if (IS_ERR(dn)) 287 err = PTR_ERR(dn); 288 } else { 289 /* we were given a hashed negative dentry */ 290 dn = NULL; 291 } 292 if (err) 293 goto out_req; 294 if (dn || d_really_is_negative(dentry) || d_is_symlink(dentry)) { 295 /* make vfs retry on splice, ENOENT, or symlink */ 296 dout("atomic_open finish_no_open on dn %p\n", dn); 297 err = finish_no_open(file, dn); 298 } else { 299 dout("atomic_open finish_open on dn %p\n", dn); 300 if (req->r_op == CEPH_MDS_OP_CREATE && req->r_reply_info.has_create_ino) { 301 ceph_init_inode_acls(d_inode(dentry), &acls); 302 *opened |= FILE_CREATED; 303 } 304 err = finish_open(file, dentry, ceph_open, opened); 305 } 306out_req: 307 if (!req->r_err && req->r_target_inode) 308 ceph_put_fmode(ceph_inode(req->r_target_inode), req->r_fmode); 309 ceph_mdsc_put_request(req); 310out_acl: 311 ceph_release_acls_info(&acls); 312 dout("atomic_open result=%d\n", err); 313 return err; 314} 315 316int ceph_release(struct inode *inode, struct file *file) 317{ 318 struct ceph_inode_info *ci = ceph_inode(inode); 319 struct ceph_file_info *cf = file->private_data; 320 321 dout("release inode %p file %p\n", inode, file); 322 ceph_put_fmode(ci, cf->fmode); 323 if (cf->last_readdir) 324 ceph_mdsc_put_request(cf->last_readdir); 325 kfree(cf->last_name); 326 kfree(cf->dir_info); 327 dput(cf->dentry); 328 kmem_cache_free(ceph_file_cachep, cf); 329 330 /* wake up anyone waiting for caps on this inode */ 331 wake_up_all(&ci->i_cap_wq); 332 return 0; 333} 334 335enum { 336 CHECK_EOF = 1, 337 READ_INLINE = 2, 338}; 339 340/* 341 * Read a range of bytes striped over one or more objects. Iterate over 342 * objects we stripe over. (That's not atomic, but good enough for now.) 343 * 344 * If we get a short result from the OSD, check against i_size; we need to 345 * only return a short read to the caller if we hit EOF. 346 */ 347static int striped_read(struct inode *inode, 348 u64 off, u64 len, 349 struct page **pages, int num_pages, 350 int *checkeof, bool o_direct, 351 unsigned long buf_align) 352{ 353 struct ceph_fs_client *fsc = ceph_inode_to_client(inode); 354 struct ceph_inode_info *ci = ceph_inode(inode); 355 u64 pos, this_len, left; 356 int io_align, page_align; 357 int pages_left; 358 int read; 359 struct page **page_pos; 360 int ret; 361 bool hit_stripe, was_short; 362 363 /* 364 * we may need to do multiple reads. not atomic, unfortunately. 365 */ 366 pos = off; 367 left = len; 368 page_pos = pages; 369 pages_left = num_pages; 370 read = 0; 371 io_align = off & ~PAGE_MASK; 372 373more: 374 if (o_direct) 375 page_align = (pos - io_align + buf_align) & ~PAGE_MASK; 376 else 377 page_align = pos & ~PAGE_MASK; 378 this_len = left; 379 ret = ceph_osdc_readpages(&fsc->client->osdc, ceph_vino(inode), 380 &ci->i_layout, pos, &this_len, 381 ci->i_truncate_seq, 382 ci->i_truncate_size, 383 page_pos, pages_left, page_align); 384 if (ret == -ENOENT) 385 ret = 0; 386 hit_stripe = this_len < left; 387 was_short = ret >= 0 && ret < this_len; 388 dout("striped_read %llu~%llu (read %u) got %d%s%s\n", pos, left, read, 389 ret, hit_stripe ? " HITSTRIPE" : "", was_short ? " SHORT" : ""); 390 391 if (ret >= 0) { 392 int didpages; 393 if (was_short && (pos + ret < inode->i_size)) { 394 int zlen = min(this_len - ret, 395 inode->i_size - pos - ret); 396 int zoff = (o_direct ? buf_align : io_align) + 397 read + ret; 398 dout(" zero gap %llu to %llu\n", 399 pos + ret, pos + ret + zlen); 400 ceph_zero_page_vector_range(zoff, zlen, pages); 401 ret += zlen; 402 } 403 404 didpages = (page_align + ret) >> PAGE_CACHE_SHIFT; 405 pos += ret; 406 read = pos - off; 407 left -= ret; 408 page_pos += didpages; 409 pages_left -= didpages; 410 411 /* hit stripe and need continue*/ 412 if (left && hit_stripe && pos < inode->i_size) 413 goto more; 414 } 415 416 if (read > 0) { 417 ret = read; 418 /* did we bounce off eof? */ 419 if (pos + left > inode->i_size) 420 *checkeof = CHECK_EOF; 421 } 422 423 dout("striped_read returns %d\n", ret); 424 return ret; 425} 426 427/* 428 * Completely synchronous read and write methods. Direct from __user 429 * buffer to osd, or directly to user pages (if O_DIRECT). 430 * 431 * If the read spans object boundary, just do multiple reads. 432 */ 433static ssize_t ceph_sync_read(struct kiocb *iocb, struct iov_iter *i, 434 int *checkeof) 435{ 436 struct file *file = iocb->ki_filp; 437 struct inode *inode = file_inode(file); 438 struct page **pages; 439 u64 off = iocb->ki_pos; 440 int num_pages, ret; 441 size_t len = iov_iter_count(i); 442 443 dout("sync_read on file %p %llu~%u %s\n", file, off, 444 (unsigned)len, 445 (file->f_flags & O_DIRECT) ? "O_DIRECT" : ""); 446 447 if (!len) 448 return 0; 449 /* 450 * flush any page cache pages in this range. this 451 * will make concurrent normal and sync io slow, 452 * but it will at least behave sensibly when they are 453 * in sequence. 454 */ 455 ret = filemap_write_and_wait_range(inode->i_mapping, off, 456 off + len); 457 if (ret < 0) 458 return ret; 459 460 if (iocb->ki_flags & IOCB_DIRECT) { 461 while (iov_iter_count(i)) { 462 size_t start; 463 ssize_t n; 464 465 n = iov_iter_get_pages_alloc(i, &pages, INT_MAX, &start); 466 if (n < 0) 467 return n; 468 469 num_pages = (n + start + PAGE_SIZE - 1) / PAGE_SIZE; 470 471 ret = striped_read(inode, off, n, 472 pages, num_pages, checkeof, 473 1, start); 474 475 ceph_put_page_vector(pages, num_pages, true); 476 477 if (ret <= 0) 478 break; 479 off += ret; 480 iov_iter_advance(i, ret); 481 if (ret < n) 482 break; 483 } 484 } else { 485 num_pages = calc_pages_for(off, len); 486 pages = ceph_alloc_page_vector(num_pages, GFP_NOFS); 487 if (IS_ERR(pages)) 488 return PTR_ERR(pages); 489 ret = striped_read(inode, off, len, pages, 490 num_pages, checkeof, 0, 0); 491 if (ret > 0) { 492 int l, k = 0; 493 size_t left = ret; 494 495 while (left) { 496 size_t page_off = off & ~PAGE_MASK; 497 size_t copy = min_t(size_t, 498 PAGE_SIZE - page_off, left); 499 l = copy_page_to_iter(pages[k++], page_off, 500 copy, i); 501 off += l; 502 left -= l; 503 if (l < copy) 504 break; 505 } 506 } 507 ceph_release_page_vector(pages, num_pages); 508 } 509 510 if (off > iocb->ki_pos) { 511 ret = off - iocb->ki_pos; 512 iocb->ki_pos = off; 513 } 514 515 dout("sync_read result %d\n", ret); 516 return ret; 517} 518 519/* 520 * Write commit request unsafe callback, called to tell us when a 521 * request is unsafe (that is, in flight--has been handed to the 522 * messenger to send to its target osd). It is called again when 523 * we've received a response message indicating the request is 524 * "safe" (its CEPH_OSD_FLAG_ONDISK flag is set), or when a request 525 * is completed early (and unsuccessfully) due to a timeout or 526 * interrupt. 527 * 528 * This is used if we requested both an ACK and ONDISK commit reply 529 * from the OSD. 530 */ 531static void ceph_sync_write_unsafe(struct ceph_osd_request *req, bool unsafe) 532{ 533 struct ceph_inode_info *ci = ceph_inode(req->r_inode); 534 535 dout("%s %p tid %llu %ssafe\n", __func__, req, req->r_tid, 536 unsafe ? "un" : ""); 537 if (unsafe) { 538 ceph_get_cap_refs(ci, CEPH_CAP_FILE_WR); 539 spin_lock(&ci->i_unsafe_lock); 540 list_add_tail(&req->r_unsafe_item, 541 &ci->i_unsafe_writes); 542 spin_unlock(&ci->i_unsafe_lock); 543 } else { 544 spin_lock(&ci->i_unsafe_lock); 545 list_del_init(&req->r_unsafe_item); 546 spin_unlock(&ci->i_unsafe_lock); 547 ceph_put_cap_refs(ci, CEPH_CAP_FILE_WR); 548 } 549} 550 551 552/* 553 * Synchronous write, straight from __user pointer or user pages. 554 * 555 * If write spans object boundary, just do multiple writes. (For a 556 * correct atomic write, we should e.g. take write locks on all 557 * objects, rollback on failure, etc.) 558 */ 559static ssize_t 560ceph_sync_direct_write(struct kiocb *iocb, struct iov_iter *from, loff_t pos) 561{ 562 struct file *file = iocb->ki_filp; 563 struct inode *inode = file_inode(file); 564 struct ceph_inode_info *ci = ceph_inode(inode); 565 struct ceph_fs_client *fsc = ceph_inode_to_client(inode); 566 struct ceph_snap_context *snapc; 567 struct ceph_vino vino; 568 struct ceph_osd_request *req; 569 struct page **pages; 570 int num_pages; 571 int written = 0; 572 int flags; 573 int check_caps = 0; 574 int ret; 575 struct timespec mtime = CURRENT_TIME; 576 size_t count = iov_iter_count(from); 577 578 if (ceph_snap(file_inode(file)) != CEPH_NOSNAP) 579 return -EROFS; 580 581 dout("sync_direct_write on file %p %lld~%u\n", file, pos, 582 (unsigned)count); 583 584 ret = filemap_write_and_wait_range(inode->i_mapping, pos, pos + count); 585 if (ret < 0) 586 return ret; 587 588 ret = invalidate_inode_pages2_range(inode->i_mapping, 589 pos >> PAGE_CACHE_SHIFT, 590 (pos + count) >> PAGE_CACHE_SHIFT); 591 if (ret < 0) 592 dout("invalidate_inode_pages2_range returned %d\n", ret); 593 594 flags = CEPH_OSD_FLAG_ORDERSNAP | 595 CEPH_OSD_FLAG_ONDISK | 596 CEPH_OSD_FLAG_WRITE; 597 598 while (iov_iter_count(from) > 0) { 599 u64 len = iov_iter_single_seg_count(from); 600 size_t start; 601 ssize_t n; 602 603 snapc = ci->i_snap_realm->cached_context; 604 vino = ceph_vino(inode); 605 req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout, 606 vino, pos, &len, 0, 607 2,/*include a 'startsync' command*/ 608 CEPH_OSD_OP_WRITE, flags, snapc, 609 ci->i_truncate_seq, 610 ci->i_truncate_size, 611 false); 612 if (IS_ERR(req)) { 613 ret = PTR_ERR(req); 614 break; 615 } 616 617 osd_req_op_init(req, 1, CEPH_OSD_OP_STARTSYNC); 618 619 n = iov_iter_get_pages_alloc(from, &pages, len, &start); 620 if (unlikely(n < 0)) { 621 ret = n; 622 ceph_osdc_put_request(req); 623 break; 624 } 625 626 num_pages = (n + start + PAGE_SIZE - 1) / PAGE_SIZE; 627 /* 628 * throw out any page cache pages in this range. this 629 * may block. 630 */ 631 truncate_inode_pages_range(inode->i_mapping, pos, 632 (pos+n) | (PAGE_CACHE_SIZE-1)); 633 osd_req_op_extent_osd_data_pages(req, 0, pages, n, start, 634 false, false); 635 636 /* BUG_ON(vino.snap != CEPH_NOSNAP); */ 637 ceph_osdc_build_request(req, pos, snapc, vino.snap, &mtime); 638 639 ret = ceph_osdc_start_request(&fsc->client->osdc, req, false); 640 if (!ret) 641 ret = ceph_osdc_wait_request(&fsc->client->osdc, req); 642 643 ceph_put_page_vector(pages, num_pages, false); 644 645 ceph_osdc_put_request(req); 646 if (ret) 647 break; 648 pos += n; 649 written += n; 650 iov_iter_advance(from, n); 651 652 if (pos > i_size_read(inode)) { 653 check_caps = ceph_inode_set_size(inode, pos); 654 if (check_caps) 655 ceph_check_caps(ceph_inode(inode), 656 CHECK_CAPS_AUTHONLY, 657 NULL); 658 } 659 } 660 661 if (ret != -EOLDSNAPC && written > 0) { 662 iocb->ki_pos = pos; 663 ret = written; 664 } 665 return ret; 666} 667 668 669/* 670 * Synchronous write, straight from __user pointer or user pages. 671 * 672 * If write spans object boundary, just do multiple writes. (For a 673 * correct atomic write, we should e.g. take write locks on all 674 * objects, rollback on failure, etc.) 675 */ 676static ssize_t 677ceph_sync_write(struct kiocb *iocb, struct iov_iter *from, loff_t pos) 678{ 679 struct file *file = iocb->ki_filp; 680 struct inode *inode = file_inode(file); 681 struct ceph_inode_info *ci = ceph_inode(inode); 682 struct ceph_fs_client *fsc = ceph_inode_to_client(inode); 683 struct ceph_snap_context *snapc; 684 struct ceph_vino vino; 685 struct ceph_osd_request *req; 686 struct page **pages; 687 u64 len; 688 int num_pages; 689 int written = 0; 690 int flags; 691 int check_caps = 0; 692 int ret; 693 struct timespec mtime = CURRENT_TIME; 694 size_t count = iov_iter_count(from); 695 696 if (ceph_snap(file_inode(file)) != CEPH_NOSNAP) 697 return -EROFS; 698 699 dout("sync_write on file %p %lld~%u\n", file, pos, (unsigned)count); 700 701 ret = filemap_write_and_wait_range(inode->i_mapping, pos, pos + count); 702 if (ret < 0) 703 return ret; 704 705 ret = invalidate_inode_pages2_range(inode->i_mapping, 706 pos >> PAGE_CACHE_SHIFT, 707 (pos + count) >> PAGE_CACHE_SHIFT); 708 if (ret < 0) 709 dout("invalidate_inode_pages2_range returned %d\n", ret); 710 711 flags = CEPH_OSD_FLAG_ORDERSNAP | 712 CEPH_OSD_FLAG_ONDISK | 713 CEPH_OSD_FLAG_WRITE | 714 CEPH_OSD_FLAG_ACK; 715 716 while ((len = iov_iter_count(from)) > 0) { 717 size_t left; 718 int n; 719 720 snapc = ci->i_snap_realm->cached_context; 721 vino = ceph_vino(inode); 722 req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout, 723 vino, pos, &len, 0, 1, 724 CEPH_OSD_OP_WRITE, flags, snapc, 725 ci->i_truncate_seq, 726 ci->i_truncate_size, 727 false); 728 if (IS_ERR(req)) { 729 ret = PTR_ERR(req); 730 break; 731 } 732 733 /* 734 * write from beginning of first page, 735 * regardless of io alignment 736 */ 737 num_pages = (len + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; 738 739 pages = ceph_alloc_page_vector(num_pages, GFP_NOFS); 740 if (IS_ERR(pages)) { 741 ret = PTR_ERR(pages); 742 goto out; 743 } 744 745 left = len; 746 for (n = 0; n < num_pages; n++) { 747 size_t plen = min_t(size_t, left, PAGE_SIZE); 748 ret = copy_page_from_iter(pages[n], 0, plen, from); 749 if (ret != plen) { 750 ret = -EFAULT; 751 break; 752 } 753 left -= ret; 754 } 755 756 if (ret < 0) { 757 ceph_release_page_vector(pages, num_pages); 758 goto out; 759 } 760 761 /* get a second commit callback */ 762 req->r_unsafe_callback = ceph_sync_write_unsafe; 763 req->r_inode = inode; 764 765 osd_req_op_extent_osd_data_pages(req, 0, pages, len, 0, 766 false, true); 767 768 /* BUG_ON(vino.snap != CEPH_NOSNAP); */ 769 ceph_osdc_build_request(req, pos, snapc, vino.snap, &mtime); 770 771 ret = ceph_osdc_start_request(&fsc->client->osdc, req, false); 772 if (!ret) 773 ret = ceph_osdc_wait_request(&fsc->client->osdc, req); 774 775out: 776 ceph_osdc_put_request(req); 777 if (ret == 0) { 778 pos += len; 779 written += len; 780 781 if (pos > i_size_read(inode)) { 782 check_caps = ceph_inode_set_size(inode, pos); 783 if (check_caps) 784 ceph_check_caps(ceph_inode(inode), 785 CHECK_CAPS_AUTHONLY, 786 NULL); 787 } 788 } else 789 break; 790 } 791 792 if (ret != -EOLDSNAPC && written > 0) { 793 ret = written; 794 iocb->ki_pos = pos; 795 } 796 return ret; 797} 798 799/* 800 * Wrap generic_file_aio_read with checks for cap bits on the inode. 801 * Atomically grab references, so that those bits are not released 802 * back to the MDS mid-read. 803 * 804 * Hmm, the sync read case isn't actually async... should it be? 805 */ 806static ssize_t ceph_read_iter(struct kiocb *iocb, struct iov_iter *to) 807{ 808 struct file *filp = iocb->ki_filp; 809 struct ceph_file_info *fi = filp->private_data; 810 size_t len = iov_iter_count(to); 811 struct inode *inode = file_inode(filp); 812 struct ceph_inode_info *ci = ceph_inode(inode); 813 struct page *pinned_page = NULL; 814 ssize_t ret; 815 int want, got = 0; 816 int retry_op = 0, read = 0; 817 818again: 819 dout("aio_read %p %llx.%llx %llu~%u trying to get caps on %p\n", 820 inode, ceph_vinop(inode), iocb->ki_pos, (unsigned)len, inode); 821 822 if (fi->fmode & CEPH_FILE_MODE_LAZY) 823 want = CEPH_CAP_FILE_CACHE | CEPH_CAP_FILE_LAZYIO; 824 else 825 want = CEPH_CAP_FILE_CACHE; 826 ret = ceph_get_caps(ci, CEPH_CAP_FILE_RD, want, -1, &got, &pinned_page); 827 if (ret < 0) 828 return ret; 829 830 if ((got & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) == 0 || 831 (iocb->ki_flags & IOCB_DIRECT) || 832 (fi->flags & CEPH_F_SYNC)) { 833 834 dout("aio_sync_read %p %llx.%llx %llu~%u got cap refs on %s\n", 835 inode, ceph_vinop(inode), iocb->ki_pos, (unsigned)len, 836 ceph_cap_string(got)); 837 838 if (ci->i_inline_version == CEPH_INLINE_NONE) { 839 /* hmm, this isn't really async... */ 840 ret = ceph_sync_read(iocb, to, &retry_op); 841 } else { 842 retry_op = READ_INLINE; 843 } 844 } else { 845 dout("aio_read %p %llx.%llx %llu~%u got cap refs on %s\n", 846 inode, ceph_vinop(inode), iocb->ki_pos, (unsigned)len, 847 ceph_cap_string(got)); 848 849 ret = generic_file_read_iter(iocb, to); 850 } 851 dout("aio_read %p %llx.%llx dropping cap refs on %s = %d\n", 852 inode, ceph_vinop(inode), ceph_cap_string(got), (int)ret); 853 if (pinned_page) { 854 page_cache_release(pinned_page); 855 pinned_page = NULL; 856 } 857 ceph_put_cap_refs(ci, got); 858 if (retry_op && ret >= 0) { 859 int statret; 860 struct page *page = NULL; 861 loff_t i_size; 862 if (retry_op == READ_INLINE) { 863 page = __page_cache_alloc(GFP_NOFS); 864 if (!page) 865 return -ENOMEM; 866 } 867 868 statret = __ceph_do_getattr(inode, page, 869 CEPH_STAT_CAP_INLINE_DATA, !!page); 870 if (statret < 0) { 871 __free_page(page); 872 if (statret == -ENODATA) { 873 BUG_ON(retry_op != READ_INLINE); 874 goto again; 875 } 876 return statret; 877 } 878 879 i_size = i_size_read(inode); 880 if (retry_op == READ_INLINE) { 881 BUG_ON(ret > 0 || read > 0); 882 if (iocb->ki_pos < i_size && 883 iocb->ki_pos < PAGE_CACHE_SIZE) { 884 loff_t end = min_t(loff_t, i_size, 885 iocb->ki_pos + len); 886 end = min_t(loff_t, end, PAGE_CACHE_SIZE); 887 if (statret < end) 888 zero_user_segment(page, statret, end); 889 ret = copy_page_to_iter(page, 890 iocb->ki_pos & ~PAGE_MASK, 891 end - iocb->ki_pos, to); 892 iocb->ki_pos += ret; 893 read += ret; 894 } 895 if (iocb->ki_pos < i_size && read < len) { 896 size_t zlen = min_t(size_t, len - read, 897 i_size - iocb->ki_pos); 898 ret = iov_iter_zero(zlen, to); 899 iocb->ki_pos += ret; 900 read += ret; 901 } 902 __free_pages(page, 0); 903 return read; 904 } 905 906 /* hit EOF or hole? */ 907 if (retry_op == CHECK_EOF && iocb->ki_pos < i_size && 908 ret < len) { 909 dout("sync_read hit hole, ppos %lld < size %lld" 910 ", reading more\n", iocb->ki_pos, 911 inode->i_size); 912 913 read += ret; 914 len -= ret; 915 retry_op = 0; 916 goto again; 917 } 918 } 919 920 if (ret >= 0) 921 ret += read; 922 923 return ret; 924} 925 926/* 927 * Take cap references to avoid releasing caps to MDS mid-write. 928 * 929 * If we are synchronous, and write with an old snap context, the OSD 930 * may return EOLDSNAPC. In that case, retry the write.. _after_ 931 * dropping our cap refs and allowing the pending snap to logically 932 * complete _before_ this write occurs. 933 * 934 * If we are near ENOSPC, write synchronously. 935 */ 936static ssize_t ceph_write_iter(struct kiocb *iocb, struct iov_iter *from) 937{ 938 struct file *file = iocb->ki_filp; 939 struct ceph_file_info *fi = file->private_data; 940 struct inode *inode = file_inode(file); 941 struct ceph_inode_info *ci = ceph_inode(inode); 942 struct ceph_osd_client *osdc = 943 &ceph_sb_to_client(inode->i_sb)->client->osdc; 944 ssize_t count, written = 0; 945 int err, want, got; 946 loff_t pos; 947 948 if (ceph_snap(inode) != CEPH_NOSNAP) 949 return -EROFS; 950 951 mutex_lock(&inode->i_mutex); 952 953 /* We can write back this queue in page reclaim */ 954 current->backing_dev_info = inode_to_bdi(inode); 955 956 err = generic_write_checks(iocb, from); 957 if (err <= 0) 958 goto out; 959 960 pos = iocb->ki_pos; 961 count = iov_iter_count(from); 962 err = file_remove_suid(file); 963 if (err) 964 goto out; 965 966 err = file_update_time(file); 967 if (err) 968 goto out; 969 970 if (ci->i_inline_version != CEPH_INLINE_NONE) { 971 err = ceph_uninline_data(file, NULL); 972 if (err < 0) 973 goto out; 974 } 975 976retry_snap: 977 if (ceph_osdmap_flag(osdc->osdmap, CEPH_OSDMAP_FULL)) { 978 err = -ENOSPC; 979 goto out; 980 } 981 982 dout("aio_write %p %llx.%llx %llu~%zd getting caps. i_size %llu\n", 983 inode, ceph_vinop(inode), pos, count, inode->i_size); 984 if (fi->fmode & CEPH_FILE_MODE_LAZY) 985 want = CEPH_CAP_FILE_BUFFER | CEPH_CAP_FILE_LAZYIO; 986 else 987 want = CEPH_CAP_FILE_BUFFER; 988 got = 0; 989 err = ceph_get_caps(ci, CEPH_CAP_FILE_WR, want, pos + count, 990 &got, NULL); 991 if (err < 0) 992 goto out; 993 994 dout("aio_write %p %llx.%llx %llu~%zd got cap refs on %s\n", 995 inode, ceph_vinop(inode), pos, count, ceph_cap_string(got)); 996 997 if ((got & (CEPH_CAP_FILE_BUFFER|CEPH_CAP_FILE_LAZYIO)) == 0 || 998 (iocb->ki_flags & IOCB_DIRECT) || (fi->flags & CEPH_F_SYNC)) { 999 struct iov_iter data; 1000 mutex_unlock(&inode->i_mutex); 1001 /* we might need to revert back to that point */ 1002 data = *from; 1003 if (iocb->ki_flags & IOCB_DIRECT) 1004 written = ceph_sync_direct_write(iocb, &data, pos); 1005 else 1006 written = ceph_sync_write(iocb, &data, pos); 1007 if (written == -EOLDSNAPC) { 1008 dout("aio_write %p %llx.%llx %llu~%u" 1009 "got EOLDSNAPC, retrying\n", 1010 inode, ceph_vinop(inode), 1011 pos, (unsigned)count); 1012 mutex_lock(&inode->i_mutex); 1013 goto retry_snap; 1014 } 1015 if (written > 0) 1016 iov_iter_advance(from, written); 1017 } else { 1018 loff_t old_size = inode->i_size; 1019 /* 1020 * No need to acquire the i_truncate_mutex. Because 1021 * the MDS revokes Fwb caps before sending truncate 1022 * message to us. We can't get Fwb cap while there 1023 * are pending vmtruncate. So write and vmtruncate 1024 * can not run at the same time 1025 */ 1026 written = generic_perform_write(file, from, pos); 1027 if (likely(written >= 0)) 1028 iocb->ki_pos = pos + written; 1029 if (inode->i_size > old_size) 1030 ceph_fscache_update_objectsize(inode); 1031 mutex_unlock(&inode->i_mutex); 1032 } 1033 1034 if (written >= 0) { 1035 int dirty; 1036 spin_lock(&ci->i_ceph_lock); 1037 ci->i_inline_version = CEPH_INLINE_NONE; 1038 dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR); 1039 spin_unlock(&ci->i_ceph_lock); 1040 if (dirty) 1041 __mark_inode_dirty(inode, dirty); 1042 } 1043 1044 dout("aio_write %p %llx.%llx %llu~%u dropping cap refs on %s\n", 1045 inode, ceph_vinop(inode), pos, (unsigned)count, 1046 ceph_cap_string(got)); 1047 ceph_put_cap_refs(ci, got); 1048 1049 if (written >= 0 && 1050 ((file->f_flags & O_SYNC) || IS_SYNC(file->f_mapping->host) || 1051 ceph_osdmap_flag(osdc->osdmap, CEPH_OSDMAP_NEARFULL))) { 1052 err = vfs_fsync_range(file, pos, pos + written - 1, 1); 1053 if (err < 0) 1054 written = err; 1055 } 1056 1057 goto out_unlocked; 1058 1059out: 1060 mutex_unlock(&inode->i_mutex); 1061out_unlocked: 1062 current->backing_dev_info = NULL; 1063 return written ? written : err; 1064} 1065 1066/* 1067 * llseek. be sure to verify file size on SEEK_END. 1068 */ 1069static loff_t ceph_llseek(struct file *file, loff_t offset, int whence) 1070{ 1071 struct inode *inode = file->f_mapping->host; 1072 int ret; 1073 1074 mutex_lock(&inode->i_mutex); 1075 1076 if (whence == SEEK_END || whence == SEEK_DATA || whence == SEEK_HOLE) { 1077 ret = ceph_do_getattr(inode, CEPH_STAT_CAP_SIZE, false); 1078 if (ret < 0) { 1079 offset = ret; 1080 goto out; 1081 } 1082 } 1083 1084 switch (whence) { 1085 case SEEK_END: 1086 offset += inode->i_size; 1087 break; 1088 case SEEK_CUR: 1089 /* 1090 * Here we special-case the lseek(fd, 0, SEEK_CUR) 1091 * position-querying operation. Avoid rewriting the "same" 1092 * f_pos value back to the file because a concurrent read(), 1093 * write() or lseek() might have altered it 1094 */ 1095 if (offset == 0) { 1096 offset = file->f_pos; 1097 goto out; 1098 } 1099 offset += file->f_pos; 1100 break; 1101 case SEEK_DATA: 1102 if (offset >= inode->i_size) { 1103 ret = -ENXIO; 1104 goto out; 1105 } 1106 break; 1107 case SEEK_HOLE: 1108 if (offset >= inode->i_size) { 1109 ret = -ENXIO; 1110 goto out; 1111 } 1112 offset = inode->i_size; 1113 break; 1114 } 1115 1116 offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes); 1117 1118out: 1119 mutex_unlock(&inode->i_mutex); 1120 return offset; 1121} 1122 1123static inline void ceph_zero_partial_page( 1124 struct inode *inode, loff_t offset, unsigned size) 1125{ 1126 struct page *page; 1127 pgoff_t index = offset >> PAGE_CACHE_SHIFT; 1128 1129 page = find_lock_page(inode->i_mapping, index); 1130 if (page) { 1131 wait_on_page_writeback(page); 1132 zero_user(page, offset & (PAGE_CACHE_SIZE - 1), size); 1133 unlock_page(page); 1134 page_cache_release(page); 1135 } 1136} 1137 1138static void ceph_zero_pagecache_range(struct inode *inode, loff_t offset, 1139 loff_t length) 1140{ 1141 loff_t nearly = round_up(offset, PAGE_CACHE_SIZE); 1142 if (offset < nearly) { 1143 loff_t size = nearly - offset; 1144 if (length < size) 1145 size = length; 1146 ceph_zero_partial_page(inode, offset, size); 1147 offset += size; 1148 length -= size; 1149 } 1150 if (length >= PAGE_CACHE_SIZE) { 1151 loff_t size = round_down(length, PAGE_CACHE_SIZE); 1152 truncate_pagecache_range(inode, offset, offset + size - 1); 1153 offset += size; 1154 length -= size; 1155 } 1156 if (length) 1157 ceph_zero_partial_page(inode, offset, length); 1158} 1159 1160static int ceph_zero_partial_object(struct inode *inode, 1161 loff_t offset, loff_t *length) 1162{ 1163 struct ceph_inode_info *ci = ceph_inode(inode); 1164 struct ceph_fs_client *fsc = ceph_inode_to_client(inode); 1165 struct ceph_osd_request *req; 1166 int ret = 0; 1167 loff_t zero = 0; 1168 int op; 1169 1170 if (!length) { 1171 op = offset ? CEPH_OSD_OP_DELETE : CEPH_OSD_OP_TRUNCATE; 1172 length = &zero; 1173 } else { 1174 op = CEPH_OSD_OP_ZERO; 1175 } 1176 1177 req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout, 1178 ceph_vino(inode), 1179 offset, length, 1180 0, 1, op, 1181 CEPH_OSD_FLAG_WRITE | 1182 CEPH_OSD_FLAG_ONDISK, 1183 NULL, 0, 0, false); 1184 if (IS_ERR(req)) { 1185 ret = PTR_ERR(req); 1186 goto out; 1187 } 1188 1189 ceph_osdc_build_request(req, offset, NULL, ceph_vino(inode).snap, 1190 &inode->i_mtime); 1191 1192 ret = ceph_osdc_start_request(&fsc->client->osdc, req, false); 1193 if (!ret) { 1194 ret = ceph_osdc_wait_request(&fsc->client->osdc, req); 1195 if (ret == -ENOENT) 1196 ret = 0; 1197 } 1198 ceph_osdc_put_request(req); 1199 1200out: 1201 return ret; 1202} 1203 1204static int ceph_zero_objects(struct inode *inode, loff_t offset, loff_t length) 1205{ 1206 int ret = 0; 1207 struct ceph_inode_info *ci = ceph_inode(inode); 1208 s32 stripe_unit = ceph_file_layout_su(ci->i_layout); 1209 s32 stripe_count = ceph_file_layout_stripe_count(ci->i_layout); 1210 s32 object_size = ceph_file_layout_object_size(ci->i_layout); 1211 u64 object_set_size = object_size * stripe_count; 1212 u64 nearly, t; 1213 1214 /* round offset up to next period boundary */ 1215 nearly = offset + object_set_size - 1; 1216 t = nearly; 1217 nearly -= do_div(t, object_set_size); 1218 1219 while (length && offset < nearly) { 1220 loff_t size = length; 1221 ret = ceph_zero_partial_object(inode, offset, &size); 1222 if (ret < 0) 1223 return ret; 1224 offset += size; 1225 length -= size; 1226 } 1227 while (length >= object_set_size) { 1228 int i; 1229 loff_t pos = offset; 1230 for (i = 0; i < stripe_count; ++i) { 1231 ret = ceph_zero_partial_object(inode, pos, NULL); 1232 if (ret < 0) 1233 return ret; 1234 pos += stripe_unit; 1235 } 1236 offset += object_set_size; 1237 length -= object_set_size; 1238 } 1239 while (length) { 1240 loff_t size = length; 1241 ret = ceph_zero_partial_object(inode, offset, &size); 1242 if (ret < 0) 1243 return ret; 1244 offset += size; 1245 length -= size; 1246 } 1247 return ret; 1248} 1249 1250static long ceph_fallocate(struct file *file, int mode, 1251 loff_t offset, loff_t length) 1252{ 1253 struct ceph_file_info *fi = file->private_data; 1254 struct inode *inode = file_inode(file); 1255 struct ceph_inode_info *ci = ceph_inode(inode); 1256 struct ceph_osd_client *osdc = 1257 &ceph_inode_to_client(inode)->client->osdc; 1258 int want, got = 0; 1259 int dirty; 1260 int ret = 0; 1261 loff_t endoff = 0; 1262 loff_t size; 1263 1264 if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE)) 1265 return -EOPNOTSUPP; 1266 1267 if (!S_ISREG(inode->i_mode)) 1268 return -EOPNOTSUPP; 1269 1270 mutex_lock(&inode->i_mutex); 1271 1272 if (ceph_snap(inode) != CEPH_NOSNAP) { 1273 ret = -EROFS; 1274 goto unlock; 1275 } 1276 1277 if (ceph_osdmap_flag(osdc->osdmap, CEPH_OSDMAP_FULL) && 1278 !(mode & FALLOC_FL_PUNCH_HOLE)) { 1279 ret = -ENOSPC; 1280 goto unlock; 1281 } 1282 1283 if (ci->i_inline_version != CEPH_INLINE_NONE) { 1284 ret = ceph_uninline_data(file, NULL); 1285 if (ret < 0) 1286 goto unlock; 1287 } 1288 1289 size = i_size_read(inode); 1290 if (!(mode & FALLOC_FL_KEEP_SIZE)) 1291 endoff = offset + length; 1292 1293 if (fi->fmode & CEPH_FILE_MODE_LAZY) 1294 want = CEPH_CAP_FILE_BUFFER | CEPH_CAP_FILE_LAZYIO; 1295 else 1296 want = CEPH_CAP_FILE_BUFFER; 1297 1298 ret = ceph_get_caps(ci, CEPH_CAP_FILE_WR, want, endoff, &got, NULL); 1299 if (ret < 0) 1300 goto unlock; 1301 1302 if (mode & FALLOC_FL_PUNCH_HOLE) { 1303 if (offset < size) 1304 ceph_zero_pagecache_range(inode, offset, length); 1305 ret = ceph_zero_objects(inode, offset, length); 1306 } else if (endoff > size) { 1307 truncate_pagecache_range(inode, size, -1); 1308 if (ceph_inode_set_size(inode, endoff)) 1309 ceph_check_caps(ceph_inode(inode), 1310 CHECK_CAPS_AUTHONLY, NULL); 1311 } 1312 1313 if (!ret) { 1314 spin_lock(&ci->i_ceph_lock); 1315 ci->i_inline_version = CEPH_INLINE_NONE; 1316 dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR); 1317 spin_unlock(&ci->i_ceph_lock); 1318 if (dirty) 1319 __mark_inode_dirty(inode, dirty); 1320 } 1321 1322 ceph_put_cap_refs(ci, got); 1323unlock: 1324 mutex_unlock(&inode->i_mutex); 1325 return ret; 1326} 1327 1328const struct file_operations ceph_file_fops = { 1329 .open = ceph_open, 1330 .release = ceph_release, 1331 .llseek = ceph_llseek, 1332 .read_iter = ceph_read_iter, 1333 .write_iter = ceph_write_iter, 1334 .mmap = ceph_mmap, 1335 .fsync = ceph_fsync, 1336 .lock = ceph_lock, 1337 .flock = ceph_flock, 1338 .splice_read = generic_file_splice_read, 1339 .splice_write = iter_file_splice_write, 1340 .unlocked_ioctl = ceph_ioctl, 1341 .compat_ioctl = ceph_ioctl, 1342 .fallocate = ceph_fallocate, 1343}; 1344 1345