fs/ext4/ioctl.c

   1 // SPDX-License-Identifier: GPL-2.0
   2 /*
   3  * linux/fs/ext4/ioctl.c
   4  *
   5  * Copyright (C) 1993, 1994, 1995
   6  * Remy Card ([email protected])
   7  * Laboratoire MASI - Institut Blaise Pascal
   8  * Universite Pierre et Marie Curie (Paris VI)
   9  */
  10
  11 #include <linux/fs.h>
  12 #include <linux/capability.h>
  13 #include <linux/time.h>
  14 #include <linux/compat.h>
  15 #include <linux/mount.h>
  16 #include <linux/file.h>
  17 #include <linux/quotaops.h>
  18 #include <linux/random.h>
  19 #include <linux/uuid.h>
  20 #include <linux/uaccess.h>
  21 #include <linux/delay.h>
  22 #include <linux/iversion.h>
  23 #include "ext4_jbd2.h"
  24 #include "ext4.h"
  25 #include <linux/fsmap.h>
  26 #include "fsmap.h"
  27 #include <trace/events/ext4.h>
  28
  29 /**
  30  * Swap memory between @a and @b for @len bytes.
  31  *
  32  * @a:          pointer to first memory area
  33  * @b:          pointer to second memory area
  34  * @len:        number of bytes to swap
  35  *
  36  */
  37 static void memswap(void *a, void *b, size_t len)
  38 {
  39         unsigned char *ap, *bp;
  40
  41         ap = (unsigned char *)a;
  42         bp = (unsigned char *)b;
  43         while (len-- > 0) {
  44                 swap(*ap, *bp);
  45                 ap++;
  46                 bp++;
  47         }
  48 }
  49
  50 /**
  51  * Swap i_data and associated attributes between @inode1 and @inode2.
  52  * This function is used for the primary swap between inode1 and inode2
  53  * and also to revert this primary swap in case of errors.
  54  *
  55  * Therefore you have to make sure, that calling this method twice
  56  * will revert all changes.
  57  *
  58  * @inode1:     pointer to first inode
  59  * @inode2:     pointer to second inode
  60  */
  61 static void swap_inode_data(struct inode *inode1, struct inode *inode2)
  62 {
  63         loff_t isize;
  64         struct ext4_inode_info *ei1;
  65         struct ext4_inode_info *ei2;
  66         unsigned long tmp;
  67
  68         ei1 = EXT4_I(inode1);
  69         ei2 = EXT4_I(inode2);
  70
  71         swap(inode1->i_version, inode2->i_version);
  72         swap(inode1->i_atime, inode2->i_atime);
  73         swap(inode1->i_mtime, inode2->i_mtime);
  74
  75         memswap(ei1->i_data, ei2->i_data, sizeof(ei1->i_data));
  76         tmp = ei1->i_flags & EXT4_FL_SHOULD_SWAP;
  77         ei1->i_flags = (ei2->i_flags & EXT4_FL_SHOULD_SWAP) |
  78                 (ei1->i_flags & ~EXT4_FL_SHOULD_SWAP);
  79         ei2->i_flags = tmp | (ei2->i_flags & ~EXT4_FL_SHOULD_SWAP);
  80         swap(ei1->i_disksize, ei2->i_disksize);
  81         ext4_es_remove_extent(inode1, 0, EXT_MAX_BLOCKS);
  82         ext4_es_remove_extent(inode2, 0, EXT_MAX_BLOCKS);
  83
  84         isize = i_size_read(inode1);
  85         i_size_write(inode1, i_size_read(inode2));
  86         i_size_write(inode2, isize);
  87 }
  88
  89 static void reset_inode_seed(struct inode *inode)
  90 {
  91         struct ext4_inode_info *ei = EXT4_I(inode);
  92         struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
  93         __le32 inum = cpu_to_le32(inode->i_ino);
  94         __le32 gen = cpu_to_le32(inode->i_generation);
  95         __u32 csum;
  96
  97         if (!ext4_has_metadata_csum(inode->i_sb))
  98                 return;
  99
 100         csum = ext4_chksum(sbi, sbi->s_csum_seed, (__u8 *)&inum, sizeof(inum));
 101         ei->i_csum_seed = ext4_chksum(sbi, csum, (__u8 *)&gen, sizeof(gen));
 102 }
 103
 104 /**
 105  * Swap the information from the given @inode and the inode
 106  * EXT4_BOOT_LOADER_INO. It will basically swap i_data and all other
 107  * important fields of the inodes.
 108  *
 109  * @sb:         the super block of the filesystem
 110  * @inode:      the inode to swap with EXT4_BOOT_LOADER_INO
 111  *
 112  */
 113 static long swap_inode_boot_loader(struct super_block *sb,
 114                                 struct inode *inode)
 115 {
 116         handle_t *handle;
 117         int err;
 118         struct inode *inode_bl;
 119         struct ext4_inode_info *ei_bl;
 120         qsize_t size, size_bl, diff;
 121         blkcnt_t blocks;
 122         unsigned short bytes;
 123
 124         inode_bl = ext4_iget(sb, EXT4_BOOT_LOADER_INO, EXT4_IGET_SPECIAL);
 125         if (IS_ERR(inode_bl))
 126                 return PTR_ERR(inode_bl);
 127         ei_bl = EXT4_I(inode_bl);
 128
 129         /* Protect orig inodes against a truncate and make sure,
 130          * that only 1 swap_inode_boot_loader is running. */
 131         lock_two_nondirectories(inode, inode_bl);
 132
 133         if (inode->i_nlink != 1 || !S_ISREG(inode->i_mode) ||
 134             IS_SWAPFILE(inode) || IS_ENCRYPTED(inode) ||
 135             (EXT4_I(inode)->i_flags & EXT4_JOURNAL_DATA_FL) ||
 136             ext4_has_inline_data(inode)) {
 137                 err = -EINVAL;
 138                 goto journal_err_out;
 139         }
 140
 141         if (IS_RDONLY(inode) || IS_APPEND(inode) || IS_IMMUTABLE(inode) ||
 142             !inode_owner_or_capable(inode) || !capable(CAP_SYS_ADMIN)) {
 143                 err = -EPERM;
 144                 goto journal_err_out;
 145         }
 146
 147         down_write(&EXT4_I(inode)->i_mmap_sem);
 148         err = filemap_write_and_wait(inode->i_mapping);
 149         if (err)
 150                 goto err_out;
 151
 152         err = filemap_write_and_wait(inode_bl->i_mapping);
 153         if (err)
 154                 goto err_out;
 155
 156         /* Wait for all existing dio workers */
 157         inode_dio_wait(inode);
 158         inode_dio_wait(inode_bl);
 159
 160         truncate_inode_pages(&inode->i_data, 0);
 161         truncate_inode_pages(&inode_bl->i_data, 0);
 162
 163         handle = ext4_journal_start(inode_bl, EXT4_HT_MOVE_EXTENTS, 2);
 164         if (IS_ERR(handle)) {
 165                 err = -EINVAL;
 166                 goto err_out;
 167         }
 168
 169         /* Protect extent tree against block allocations via delalloc */
 170         ext4_double_down_write_data_sem(inode, inode_bl);
 171
 172         if (inode_bl->i_nlink == 0) {
 173                 /* this inode has never been used as a BOOT_LOADER */
 174                 set_nlink(inode_bl, 1);
 175                 i_uid_write(inode_bl, 0);
 176                 i_gid_write(inode_bl, 0);
 177                 inode_bl->i_flags = 0;
 178                 ei_bl->i_flags = 0;
 179                 inode_set_iversion(inode_bl, 1);
 180                 i_size_write(inode_bl, 0);
 181                 inode_bl->i_mode = S_IFREG;
 182                 if (ext4_has_feature_extents(sb)) {
 183                         ext4_set_inode_flag(inode_bl, EXT4_INODE_EXTENTS);
 184                         ext4_ext_tree_init(handle, inode_bl);
 185                 } else
 186                         memset(ei_bl->i_data, 0, sizeof(ei_bl->i_data));
 187         }
 188
 189         err = dquot_initialize(inode);
 190         if (err)
 191                 goto err_out1;
 192
 193         size = (qsize_t)(inode->i_blocks) * (1 << 9) + inode->i_bytes;
 194         size_bl = (qsize_t)(inode_bl->i_blocks) * (1 << 9) + inode_bl->i_bytes;
 195         diff = size - size_bl;
 196         swap_inode_data(inode, inode_bl);
 197
 198         inode->i_ctime = inode_bl->i_ctime = current_time(inode);
 199
 200         inode->i_generation = prandom_u32();
 201         inode_bl->i_generation = prandom_u32();
 202         reset_inode_seed(inode);
 203         reset_inode_seed(inode_bl);
 204
 205         ext4_discard_preallocations(inode);
 206
 207         err = ext4_mark_inode_dirty(handle, inode);
 208         if (err < 0) {
 209                 /* No need to update quota information. */
 210                 ext4_warning(inode->i_sb,
 211                         "couldn't mark inode #%lu dirty (err %d)",
 212                         inode->i_ino, err);
 213                 /* Revert all changes: */
 214                 swap_inode_data(inode, inode_bl);
 215                 ext4_mark_inode_dirty(handle, inode);
 216                 goto err_out1;
 217         }
 218
 219         blocks = inode_bl->i_blocks;
 220         bytes = inode_bl->i_bytes;
 221         inode_bl->i_blocks = inode->i_blocks;
 222         inode_bl->i_bytes = inode->i_bytes;
 223         err = ext4_mark_inode_dirty(handle, inode_bl);
 224         if (err < 0) {
 225                 /* No need to update quota information. */
 226                 ext4_warning(inode_bl->i_sb,
 227                         "couldn't mark inode #%lu dirty (err %d)",
 228                         inode_bl->i_ino, err);
 229                 goto revert;
 230         }
 231
 232         /* Bootloader inode should not be counted into quota information. */
 233         if (diff > 0)
 234                 dquot_free_space(inode, diff);
 235         else
 236                 err = dquot_alloc_space(inode, -1 * diff);
 237
 238         if (err < 0) {
 239 revert:
 240                 /* Revert all changes: */
 241                 inode_bl->i_blocks = blocks;
 242                 inode_bl->i_bytes = bytes;
 243                 swap_inode_data(inode, inode_bl);
 244                 ext4_mark_inode_dirty(handle, inode);
 245                 ext4_mark_inode_dirty(handle, inode_bl);
 246         }
 247
 248 err_out1:
 249         ext4_journal_stop(handle);
 250         ext4_double_up_write_data_sem(inode, inode_bl);
 251
 252 err_out:
 253         up_write(&EXT4_I(inode)->i_mmap_sem);
 254 journal_err_out:
 255         unlock_two_nondirectories(inode, inode_bl);
 256         iput(inode_bl);
 257         return err;
 258 }
 259
 260 #ifdef CONFIG_FS_ENCRYPTION
 261 static int uuid_is_zero(__u8 u[16])
 262 {
 263         int     i;
 264
 265         for (i = 0; i < 16; i++)
 266                 if (u[i])
 267                         return 0;
 268         return 1;
 269 }
 270 #endif
 271
 272 static int ext4_ioctl_setflags(struct inode *inode,
 273                                unsigned int flags)
 274 {
 275         struct ext4_inode_info *ei = EXT4_I(inode);
 276         handle_t *handle = NULL;
 277         int err = -EPERM, migrate = 0;
 278         struct ext4_iloc iloc;
 279         unsigned int oldflags, mask, i;
 280         unsigned int jflag;
 281         struct super_block *sb = inode->i_sb;
 282
 283         /* Is it quota file? Do not allow user to mess with it */
 284         if (ext4_is_quota_file(inode))
 285                 goto flags_out;
 286
 287         oldflags = ei->i_flags;
 288
 289         /* The JOURNAL_DATA flag is modifiable only by root */
 290         jflag = flags & EXT4_JOURNAL_DATA_FL;
 291
 292         /*
 293          * The IMMUTABLE and APPEND_ONLY flags can only be changed by
 294          * the relevant capability.
 295          *
 296          * This test looks nicer. Thanks to Pauline Middelink
 297          */
 298         if ((flags ^ oldflags) & (EXT4_APPEND_FL | EXT4_IMMUTABLE_FL)) {
 299                 if (!capable(CAP_LINUX_IMMUTABLE))
 300                         goto flags_out;
 301         }
 302
 303         /*
 304          * The JOURNAL_DATA flag can only be changed by
 305          * the relevant capability.
 306          */
 307         if ((jflag ^ oldflags) & (EXT4_JOURNAL_DATA_FL)) {
 308                 if (!capable(CAP_SYS_RESOURCE))
 309                         goto flags_out;
 310         }
 311         if ((flags ^ oldflags) & EXT4_EXTENTS_FL)
 312                 migrate = 1;
 313
 314         if (flags & EXT4_EOFBLOCKS_FL) {
 315                 /* we don't support adding EOFBLOCKS flag */
 316                 if (!(oldflags & EXT4_EOFBLOCKS_FL)) {
 317                         err = -EOPNOTSUPP;
 318                         goto flags_out;
 319                 }
 320         } else if (oldflags & EXT4_EOFBLOCKS_FL) {
 321                 err = ext4_truncate(inode);
 322                 if (err)
 323                         goto flags_out;
 324         }
 325
 326         if ((flags ^ oldflags) & EXT4_CASEFOLD_FL) {
 327                 if (!ext4_has_feature_casefold(sb)) {
 328                         err = -EOPNOTSUPP;
 329                         goto flags_out;
 330                 }
 331
 332                 if (!S_ISDIR(inode->i_mode)) {
 333                         err = -ENOTDIR;
 334                         goto flags_out;
 335                 }
 336
 337                 if (!ext4_empty_dir(inode)) {
 338                         err = -ENOTEMPTY;
 339                         goto flags_out;
 340                 }
 341         }
 342
 343         handle = ext4_journal_start(inode, EXT4_HT_INODE, 1);
 344         if (IS_ERR(handle)) {
 345                 err = PTR_ERR(handle);
 346                 goto flags_out;
 347         }
 348         if (IS_SYNC(inode))
 349                 ext4_handle_sync(handle);
 350         err = ext4_reserve_inode_write(handle, inode, &iloc);
 351         if (err)
 352                 goto flags_err;
 353
 354         for (i = 0, mask = 1; i < 32; i++, mask <<= 1) {
 355                 if (!(mask & EXT4_FL_USER_MODIFIABLE))
 356                         continue;
 357                 /* These flags get special treatment later */
 358                 if (mask == EXT4_JOURNAL_DATA_FL || mask == EXT4_EXTENTS_FL)
 359                         continue;
 360                 if (mask & flags)
 361                         ext4_set_inode_flag(inode, i);
 362                 else
 363                         ext4_clear_inode_flag(inode, i);
 364         }
 365
 366         ext4_set_inode_flags(inode);
 367         inode->i_ctime = current_time(inode);
 368
 369         err = ext4_mark_iloc_dirty(handle, inode, &iloc);
 370 flags_err:
 371         ext4_journal_stop(handle);
 372         if (err)
 373                 goto flags_out;
 374
 375         if ((jflag ^ oldflags) & (EXT4_JOURNAL_DATA_FL)) {
 376                 /*
 377                  * Changes to the journaling mode can cause unsafe changes to
 378                  * S_DAX if we are using the DAX mount option.
 379                  */
 380                 if (test_opt(inode->i_sb, DAX)) {
 381                         err = -EBUSY;
 382                         goto flags_out;
 383                 }
 384
 385                 err = ext4_change_inode_journal_flag(inode, jflag);
 386                 if (err)
 387                         goto flags_out;
 388         }
 389         if (migrate) {
 390                 if (flags & EXT4_EXTENTS_FL)
 391                         err = ext4_ext_migrate(inode);
 392                 else
 393                         err = ext4_ind_migrate(inode);
 394         }
 395
 396 flags_out:
 397         return err;
 398 }
 399
 400 #ifdef CONFIG_QUOTA
 401 static int ext4_ioctl_setproject(struct file *filp, __u32 projid)
 402 {
 403         struct inode *inode = file_inode(filp);
 404         struct super_block *sb = inode->i_sb;
 405         struct ext4_inode_info *ei = EXT4_I(inode);
 406         int err, rc;
 407         handle_t *handle;
 408         kprojid_t kprojid;
 409         struct ext4_iloc iloc;
 410         struct ext4_inode *raw_inode;
 411         struct dquot *transfer_to[MAXQUOTAS] = { };
 412
 413         if (!ext4_has_feature_project(sb)) {
 414                 if (projid != EXT4_DEF_PROJID)
 415                         return -EOPNOTSUPP;
 416                 else
 417                         return 0;
 418         }
 419
 420         if (EXT4_INODE_SIZE(sb) <= EXT4_GOOD_OLD_INODE_SIZE)
 421                 return -EOPNOTSUPP;
 422
 423         kprojid = make_kprojid(&init_user_ns, (projid_t)projid);
 424
 425         if (projid_eq(kprojid, EXT4_I(inode)->i_projid))
 426                 return 0;
 427
 428         err = -EPERM;
 429         /* Is it quota file? Do not allow user to mess with it */
 430         if (ext4_is_quota_file(inode))
 431                 return err;
 432
 433         err = ext4_get_inode_loc(inode, &iloc);
 434         if (err)
 435                 return err;
 436
 437         raw_inode = ext4_raw_inode(&iloc);
 438         if (!EXT4_FITS_IN_INODE(raw_inode, ei, i_projid)) {
 439                 err = ext4_expand_extra_isize(inode,
 440                                               EXT4_SB(sb)->s_want_extra_isize,
 441                                               &iloc);
 442                 if (err)
 443                         return err;
 444         } else {
 445                 brelse(iloc.bh);
 446         }
 447
 448         err = dquot_initialize(inode);
 449         if (err)
 450                 return err;
 451
 452         handle = ext4_journal_start(inode, EXT4_HT_QUOTA,
 453                 EXT4_QUOTA_INIT_BLOCKS(sb) +
 454                 EXT4_QUOTA_DEL_BLOCKS(sb) + 3);
 455         if (IS_ERR(handle))
 456                 return PTR_ERR(handle);
 457
 458         err = ext4_reserve_inode_write(handle, inode, &iloc);
 459         if (err)
 460                 goto out_stop;
 461
 462         transfer_to[PRJQUOTA] = dqget(sb, make_kqid_projid(kprojid));
 463         if (!IS_ERR(transfer_to[PRJQUOTA])) {
 464
 465                 /* __dquot_transfer() calls back ext4_get_inode_usage() which
 466                  * counts xattr inode references.
 467                  */
 468                 down_read(&EXT4_I(inode)->xattr_sem);
 469                 err = __dquot_transfer(inode, transfer_to);
 470                 up_read(&EXT4_I(inode)->xattr_sem);
 471                 dqput(transfer_to[PRJQUOTA]);
 472                 if (err)
 473                         goto out_dirty;
 474         }
 475
 476         EXT4_I(inode)->i_projid = kprojid;
 477         inode->i_ctime = current_time(inode);
 478 out_dirty:
 479         rc = ext4_mark_iloc_dirty(handle, inode, &iloc);
 480         if (!err)
 481                 err = rc;
 482 out_stop:
 483         ext4_journal_stop(handle);
 484         return err;
 485 }
 486 #else
 487 static int ext4_ioctl_setproject(struct file *filp, __u32 projid)
 488 {
 489         if (projid != EXT4_DEF_PROJID)
 490                 return -EOPNOTSUPP;
 491         return 0;
 492 }
 493 #endif
 494
 495 /* Transfer internal flags to xflags */
 496 static inline __u32 ext4_iflags_to_xflags(unsigned long iflags)
 497 {
 498         __u32 xflags = 0;
 499
 500         if (iflags & EXT4_SYNC_FL)
 501                 xflags |= FS_XFLAG_SYNC;
 502         if (iflags & EXT4_IMMUTABLE_FL)
 503                 xflags |= FS_XFLAG_IMMUTABLE;
 504         if (iflags & EXT4_APPEND_FL)
 505                 xflags |= FS_XFLAG_APPEND;
 506         if (iflags & EXT4_NODUMP_FL)
 507                 xflags |= FS_XFLAG_NODUMP;
 508         if (iflags & EXT4_NOATIME_FL)
 509                 xflags |= FS_XFLAG_NOATIME;
 510         if (iflags & EXT4_PROJINHERIT_FL)
 511                 xflags |= FS_XFLAG_PROJINHERIT;
 512         return xflags;
 513 }
 514
 515 #define EXT4_SUPPORTED_FS_XFLAGS (FS_XFLAG_SYNC | FS_XFLAG_IMMUTABLE | \
 516                                   FS_XFLAG_APPEND | FS_XFLAG_NODUMP | \
 517                                   FS_XFLAG_NOATIME | FS_XFLAG_PROJINHERIT)
 518
 519 /* Transfer xflags flags to internal */
 520 static inline unsigned long ext4_xflags_to_iflags(__u32 xflags)
 521 {
 522         unsigned long iflags = 0;
 523
 524         if (xflags & FS_XFLAG_SYNC)
 525                 iflags |= EXT4_SYNC_FL;
 526         if (xflags & FS_XFLAG_IMMUTABLE)
 527                 iflags |= EXT4_IMMUTABLE_FL;
 528         if (xflags & FS_XFLAG_APPEND)
 529                 iflags |= EXT4_APPEND_FL;
 530         if (xflags & FS_XFLAG_NODUMP)
 531                 iflags |= EXT4_NODUMP_FL;
 532         if (xflags & FS_XFLAG_NOATIME)
 533                 iflags |= EXT4_NOATIME_FL;
 534         if (xflags & FS_XFLAG_PROJINHERIT)
 535                 iflags |= EXT4_PROJINHERIT_FL;
 536
 537         return iflags;
 538 }
 539
 540 static int ext4_shutdown(struct super_block *sb, unsigned long arg)
 541 {
 542         struct ext4_sb_info *sbi = EXT4_SB(sb);
 543         __u32 flags;
 544
 545         if (!capable(CAP_SYS_ADMIN))
 546                 return -EPERM;
 547
 548         if (get_user(flags, (__u32 __user *)arg))
 549                 return -EFAULT;
 550
 551         if (flags > EXT4_GOING_FLAGS_NOLOGFLUSH)
 552                 return -EINVAL;
 553
 554         if (ext4_forced_shutdown(sbi))
 555                 return 0;
 556
 557         ext4_msg(sb, KERN_ALERT, "shut down requested (%d)", flags);
 558         trace_ext4_shutdown(sb, flags);
 559
 560         switch (flags) {
 561         case EXT4_GOING_FLAGS_DEFAULT:
 562                 freeze_bdev(sb->s_bdev);
 563                 set_bit(EXT4_FLAGS_SHUTDOWN, &sbi->s_ext4_flags);
 564                 thaw_bdev(sb->s_bdev, sb);
 565                 break;
 566         case EXT4_GOING_FLAGS_LOGFLUSH:
 567                 set_bit(EXT4_FLAGS_SHUTDOWN, &sbi->s_ext4_flags);
 568                 if (sbi->s_journal && !is_journal_aborted(sbi->s_journal)) {
 569                         (void) ext4_force_commit(sb);
 570                         jbd2_journal_abort(sbi->s_journal, -ESHUTDOWN);
 571                 }
 572                 break;
 573         case EXT4_GOING_FLAGS_NOLOGFLUSH:
 574                 set_bit(EXT4_FLAGS_SHUTDOWN, &sbi->s_ext4_flags);
 575                 if (sbi->s_journal && !is_journal_aborted(sbi->s_journal))
 576                         jbd2_journal_abort(sbi->s_journal, -ESHUTDOWN);
 577                 break;
 578         default:
 579                 return -EINVAL;
 580         }
 581         clear_opt(sb, DISCARD);
 582         return 0;
 583 }
 584
 585 struct getfsmap_info {
 586         struct super_block      *gi_sb;
 587         struct fsmap_head __user *gi_data;
 588         unsigned int            gi_idx;
 589         __u32                   gi_last_flags;
 590 };
 591
 592 static int ext4_getfsmap_format(struct ext4_fsmap *xfm, void *priv)
 593 {
 594         struct getfsmap_info *info = priv;
 595         struct fsmap fm;
 596
 597         trace_ext4_getfsmap_mapping(info->gi_sb, xfm);
 598
 599         info->gi_last_flags = xfm->fmr_flags;
 600         ext4_fsmap_from_internal(info->gi_sb, &fm, xfm);
 601         if (copy_to_user(&info->gi_data->fmh_recs[info->gi_idx++], &fm,
 602                         sizeof(struct fsmap)))
 603                 return -EFAULT;
 604
 605         return 0;
 606 }
 607
 608 static int ext4_ioc_getfsmap(struct super_block *sb,
 609                              struct fsmap_head __user *arg)
 610 {
 611         struct getfsmap_info info = { NULL };
 612         struct ext4_fsmap_head xhead = {0};
 613         struct fsmap_head head;
 614         bool aborted = false;
 615         int error;
 616
 617         if (copy_from_user(&head, arg, sizeof(struct fsmap_head)))
 618                 return -EFAULT;
 619         if (memchr_inv(head.fmh_reserved, 0, sizeof(head.fmh_reserved)) ||
 620             memchr_inv(head.fmh_keys[0].fmr_reserved, 0,
 621                        sizeof(head.fmh_keys[0].fmr_reserved)) ||
 622             memchr_inv(head.fmh_keys[1].fmr_reserved, 0,
 623                        sizeof(head.fmh_keys[1].fmr_reserved)))
 624                 return -EINVAL;
 625         /*
 626          * ext4 doesn't report file extents at all, so the only valid
 627          * file offsets are the magic ones (all zeroes or all ones).
 628          */
 629         if (head.fmh_keys[0].fmr_offset ||
 630             (head.fmh_keys[1].fmr_offset != 0 &&
 631              head.fmh_keys[1].fmr_offset != -1ULL))
 632                 return -EINVAL;
 633
 634         xhead.fmh_iflags = head.fmh_iflags;
 635         xhead.fmh_count = head.fmh_count;
 636         ext4_fsmap_to_internal(sb, &xhead.fmh_keys[0], &head.fmh_keys[0]);
 637         ext4_fsmap_to_internal(sb, &xhead.fmh_keys[1], &head.fmh_keys[1]);
 638
 639         trace_ext4_getfsmap_low_key(sb, &xhead.fmh_keys[0]);
 640         trace_ext4_getfsmap_high_key(sb, &xhead.fmh_keys[1]);
 641
 642         info.gi_sb = sb;
 643         info.gi_data = arg;
 644         error = ext4_getfsmap(sb, &xhead, ext4_getfsmap_format, &info);
 645         if (error == EXT4_QUERY_RANGE_ABORT) {
 646                 error = 0;
 647                 aborted = true;
 648         } else if (error)
 649                 return error;
 650
 651         /* If we didn't abort, set the "last" flag in the last fmx */
 652         if (!aborted && info.gi_idx) {
 653                 info.gi_last_flags |= FMR_OF_LAST;
 654                 if (copy_to_user(&info.gi_data->fmh_recs[info.gi_idx - 1].fmr_flags,
 655                                  &info.gi_last_flags,
 656                                  sizeof(info.gi_last_flags)))
 657                         return -EFAULT;
 658         }
 659
 660         /* copy back header */
 661         head.fmh_entries = xhead.fmh_entries;
 662         head.fmh_oflags = xhead.fmh_oflags;
 663         if (copy_to_user(arg, &head, sizeof(struct fsmap_head)))
 664                 return -EFAULT;
 665
 666         return 0;
 667 }
 668
 669 static long ext4_ioctl_group_add(struct file *file,
 670                                  struct ext4_new_group_data *input)
 671 {
 672         struct super_block *sb = file_inode(file)->i_sb;
 673         int err, err2=0;
 674
 675         err = ext4_resize_begin(sb);
 676         if (err)
 677                 return err;
 678
 679         if (ext4_has_feature_bigalloc(sb)) {
 680                 ext4_msg(sb, KERN_ERR,
 681                          "Online resizing not supported with bigalloc");
 682                 err = -EOPNOTSUPP;
 683                 goto group_add_out;
 684         }
 685
 686         err = mnt_want_write_file(file);
 687         if (err)
 688                 goto group_add_out;
 689
 690         err = ext4_group_add(sb, input);
 691         if (EXT4_SB(sb)->s_journal) {
 692                 jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal);
 693                 err2 = jbd2_journal_flush(EXT4_SB(sb)->s_journal);
 694                 jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal);
 695         }
 696         if (err == 0)
 697                 err = err2;
 698         mnt_drop_write_file(file);
 699         if (!err && ext4_has_group_desc_csum(sb) &&
 700             test_opt(sb, INIT_INODE_TABLE))
 701                 err = ext4_register_li_request(sb, input->group);
 702 group_add_out:
 703         ext4_resize_end(sb);
 704         return err;
 705 }
 706
 707 static int ext4_ioctl_check_project(struct inode *inode, struct fsxattr *fa)
 708 {
 709         /*
 710          * Project Quota ID state is only allowed to change from within the init
 711          * namespace. Enforce that restriction only if we are trying to change
 712          * the quota ID state. Everything else is allowed in user namespaces.
 713          */
 714         if (current_user_ns() == &init_user_ns)
 715                 return 0;
 716
 717         if (__kprojid_val(EXT4_I(inode)->i_projid) != fa->fsx_projid)
 718                 return -EINVAL;
 719
 720         if (ext4_test_inode_flag(inode, EXT4_INODE_PROJINHERIT)) {
 721                 if (!(fa->fsx_xflags & FS_XFLAG_PROJINHERIT))
 722                         return -EINVAL;
 723         } else {
 724                 if (fa->fsx_xflags & FS_XFLAG_PROJINHERIT)
 725                         return -EINVAL;
 726         }
 727
 728         return 0;
 729 }
 730
 731 long ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 732 {
 733         struct inode *inode = file_inode(filp);
 734         struct super_block *sb = inode->i_sb;
 735         struct ext4_inode_info *ei = EXT4_I(inode);
 736         unsigned int flags;
 737
 738         ext4_debug("cmd = %u, arg = %lu\n", cmd, arg);
 739
 740         switch (cmd) {
 741         case FS_IOC_GETFSMAP:
 742                 return ext4_ioc_getfsmap(sb, (void __user *)arg);
 743         case EXT4_IOC_GETFLAGS:
 744                 flags = ei->i_flags & EXT4_FL_USER_VISIBLE;
 745                 return put_user(flags, (int __user *) arg);
 746         case EXT4_IOC_SETFLAGS: {
 747                 int err;
 748
 749                 if (!inode_owner_or_capable(inode))
 750                         return -EACCES;
 751
 752                 if (get_user(flags, (int __user *) arg))
 753                         return -EFAULT;
 754
 755                 if (flags & ~EXT4_FL_USER_VISIBLE)
 756                         return -EOPNOTSUPP;
 757                 /*
 758                  * chattr(1) grabs flags via GETFLAGS, modifies the result and
 759                  * passes that to SETFLAGS. So we cannot easily make SETFLAGS
 760                  * more restrictive than just silently masking off visible but
 761                  * not settable flags as we always did.
 762                  */
 763                 flags &= EXT4_FL_USER_MODIFIABLE;
 764                 if (ext4_mask_flags(inode->i_mode, flags) != flags)
 765                         return -EOPNOTSUPP;
 766
 767                 err = mnt_want_write_file(filp);
 768                 if (err)
 769                         return err;
 770
 771                 inode_lock(inode);
 772                 err = ext4_ioctl_setflags(inode, flags);
 773                 inode_unlock(inode);
 774                 mnt_drop_write_file(filp);
 775                 return err;
 776         }
 777         case EXT4_IOC_GETVERSION:
 778         case EXT4_IOC_GETVERSION_OLD:
 779                 return put_user(inode->i_generation, (int __user *) arg);
 780         case EXT4_IOC_SETVERSION:
 781         case EXT4_IOC_SETVERSION_OLD: {
 782                 handle_t *handle;
 783                 struct ext4_iloc iloc;
 784                 __u32 generation;
 785                 int err;
 786
 787                 if (!inode_owner_or_capable(inode))
 788                         return -EPERM;
 789
 790                 if (ext4_has_metadata_csum(inode->i_sb)) {
 791                         ext4_warning(sb, "Setting inode version is not "
 792                                      "supported with metadata_csum enabled.");
 793                         return -ENOTTY;
 794                 }
 795
 796                 err = mnt_want_write_file(filp);
 797                 if (err)
 798                         return err;
 799                 if (get_user(generation, (int __user *) arg)) {
 800                         err = -EFAULT;
 801                         goto setversion_out;
 802                 }
 803
 804                 inode_lock(inode);
 805                 handle = ext4_journal_start(inode, EXT4_HT_INODE, 1);
 806                 if (IS_ERR(handle)) {
 807                         err = PTR_ERR(handle);
 808                         goto unlock_out;
 809                 }
 810                 err = ext4_reserve_inode_write(handle, inode, &iloc);
 811                 if (err == 0) {
 812                         inode->i_ctime = current_time(inode);
 813                         inode->i_generation = generation;
 814                         err = ext4_mark_iloc_dirty(handle, inode, &iloc);
 815                 }
 816                 ext4_journal_stop(handle);
 817
 818 unlock_out:
 819                 inode_unlock(inode);
 820 setversion_out:
 821                 mnt_drop_write_file(filp);
 822                 return err;
 823         }
 824         case EXT4_IOC_GROUP_EXTEND: {
 825                 ext4_fsblk_t n_blocks_count;
 826                 int err, err2=0;
 827
 828                 err = ext4_resize_begin(sb);
 829                 if (err)
 830                         return err;
 831
 832                 if (get_user(n_blocks_count, (__u32 __user *)arg)) {
 833                         err = -EFAULT;
 834                         goto group_extend_out;
 835                 }
 836
 837                 if (ext4_has_feature_bigalloc(sb)) {
 838                         ext4_msg(sb, KERN_ERR,
 839                                  "Online resizing not supported with bigalloc");
 840                         err = -EOPNOTSUPP;
 841                         goto group_extend_out;
 842                 }
 843
 844                 err = mnt_want_write_file(filp);
 845                 if (err)
 846                         goto group_extend_out;
 847
 848                 err = ext4_group_extend(sb, EXT4_SB(sb)->s_es, n_blocks_count);
 849                 if (EXT4_SB(sb)->s_journal) {
 850                         jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal);
 851                         err2 = jbd2_journal_flush(EXT4_SB(sb)->s_journal);
 852                         jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal);
 853                 }
 854                 if (err == 0)
 855                         err = err2;
 856                 mnt_drop_write_file(filp);
 857 group_extend_out:
 858                 ext4_resize_end(sb);
 859                 return err;
 860         }
 861
 862         case EXT4_IOC_MOVE_EXT: {
 863                 struct move_extent me;
 864                 struct fd donor;
 865                 int err;
 866
 867                 if (!(filp->f_mode & FMODE_READ) ||
 868                     !(filp->f_mode & FMODE_WRITE))
 869                         return -EBADF;
 870
 871                 if (copy_from_user(&me,
 872                         (struct move_extent __user *)arg, sizeof(me)))
 873                         return -EFAULT;
 874                 me.moved_len = 0;
 875
 876                 donor = fdget(me.donor_fd);
 877                 if (!donor.file)
 878                         return -EBADF;
 879
 880                 if (!(donor.file->f_mode & FMODE_WRITE)) {
 881                         err = -EBADF;
 882                         goto mext_out;
 883                 }
 884
 885                 if (ext4_has_feature_bigalloc(sb)) {
 886                         ext4_msg(sb, KERN_ERR,
 887                                  "Online defrag not supported with bigalloc");
 888                         err = -EOPNOTSUPP;
 889                         goto mext_out;
 890                 } else if (IS_DAX(inode)) {
 891                         ext4_msg(sb, KERN_ERR,
 892                                  "Online defrag not supported with DAX");
 893                         err = -EOPNOTSUPP;
 894                         goto mext_out;
 895                 }
 896
 897                 err = mnt_want_write_file(filp);
 898                 if (err)
 899                         goto mext_out;
 900
 901                 err = ext4_move_extents(filp, donor.file, me.orig_start,
 902                                         me.donor_start, me.len, &me.moved_len);
 903                 mnt_drop_write_file(filp);
 904
 905                 if (copy_to_user((struct move_extent __user *)arg,
 906                                  &me, sizeof(me)))
 907                         err = -EFAULT;
 908 mext_out:
 909                 fdput(donor);
 910                 return err;
 911         }
 912
 913         case EXT4_IOC_GROUP_ADD: {
 914                 struct ext4_new_group_data input;
 915
 916                 if (copy_from_user(&input, (struct ext4_new_group_input __user *)arg,
 917                                 sizeof(input)))
 918                         return -EFAULT;
 919
 920                 return ext4_ioctl_group_add(filp, &input);
 921         }
 922
 923         case EXT4_IOC_MIGRATE:
 924         {
 925                 int err;
 926                 if (!inode_owner_or_capable(inode))
 927                         return -EACCES;
 928
 929                 err = mnt_want_write_file(filp);
 930                 if (err)
 931                         return err;
 932                 /*
 933                  * inode_mutex prevent write and truncate on the file.
 934                  * Read still goes through. We take i_data_sem in
 935                  * ext4_ext_swap_inode_data before we switch the
 936                  * inode format to prevent read.
 937                  */
 938                 inode_lock((inode));
 939                 err = ext4_ext_migrate(inode);
 940                 inode_unlock((inode));
 941                 mnt_drop_write_file(filp);
 942                 return err;
 943         }
 944
 945         case EXT4_IOC_ALLOC_DA_BLKS:
 946         {
 947                 int err;
 948                 if (!inode_owner_or_capable(inode))
 949                         return -EACCES;
 950
 951                 err = mnt_want_write_file(filp);
 952                 if (err)
 953                         return err;
 954                 err = ext4_alloc_da_blocks(inode);
 955                 mnt_drop_write_file(filp);
 956                 return err;
 957         }
 958
 959         case EXT4_IOC_SWAP_BOOT:
 960         {
 961                 int err;
 962                 if (!(filp->f_mode & FMODE_WRITE))
 963                         return -EBADF;
 964                 err = mnt_want_write_file(filp);
 965                 if (err)
 966                         return err;
 967                 err = swap_inode_boot_loader(sb, inode);
 968                 mnt_drop_write_file(filp);
 969                 return err;
 970         }
 971
 972         case EXT4_IOC_RESIZE_FS: {
 973                 ext4_fsblk_t n_blocks_count;
 974                 int err = 0, err2 = 0;
 975                 ext4_group_t o_group = EXT4_SB(sb)->s_groups_count;
 976
 977                 if (copy_from_user(&n_blocks_count, (__u64 __user *)arg,
 978                                    sizeof(__u64))) {
 979                         return -EFAULT;
 980                 }
 981
 982                 err = ext4_resize_begin(sb);
 983                 if (err)
 984                         return err;
 985
 986                 err = mnt_want_write_file(filp);
 987                 if (err)
 988                         goto resizefs_out;
 989
 990                 err = ext4_resize_fs(sb, n_blocks_count);
 991                 if (EXT4_SB(sb)->s_journal) {
 992                         jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal);
 993                         err2 = jbd2_journal_flush(EXT4_SB(sb)->s_journal);
 994                         jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal);
 995                 }
 996                 if (err == 0)
 997                         err = err2;
 998                 mnt_drop_write_file(filp);
 999                 if (!err && (o_group < EXT4_SB(sb)->s_groups_count) &&
1000                     ext4_has_group_desc_csum(sb) &&
1001                     test_opt(sb, INIT_INODE_TABLE))
1002                         err = ext4_register_li_request(sb, o_group);
1003
1004 resizefs_out:
1005                 ext4_resize_end(sb);
1006                 return err;
1007         }
1008
1009         case FITRIM:
1010         {
1011                 struct request_queue *q = bdev_get_queue(sb->s_bdev);
1012                 struct fstrim_range range;
1013                 int ret = 0;
1014
1015                 if (!capable(CAP_SYS_ADMIN))
1016                         return -EPERM;
1017
1018                 if (!blk_queue_discard(q))
1019                         return -EOPNOTSUPP;
1020
1021                 /*
1022                  * We haven't replayed the journal, so we cannot use our
1023                  * block-bitmap-guided storage zapping commands.
1024                  */
1025                 if (test_opt(sb, NOLOAD) && ext4_has_feature_journal(sb))
1026                         return -EROFS;
1027
1028                 if (copy_from_user(&range, (struct fstrim_range __user *)arg,
1029                     sizeof(range)))
1030                         return -EFAULT;
1031
1032                 range.minlen = max((unsigned int)range.minlen,
1033                                    q->limits.discard_granularity);
1034                 ret = ext4_trim_fs(sb, &range);
1035                 if (ret < 0)
1036                         return ret;
1037
1038                 if (copy_to_user((struct fstrim_range __user *)arg, &range,
1039                     sizeof(range)))
1040                         return -EFAULT;
1041
1042                 return 0;
1043         }
1044         case EXT4_IOC_PRECACHE_EXTENTS:
1045                 return ext4_ext_precache(inode);
1046
1047         case EXT4_IOC_SET_ENCRYPTION_POLICY:
1048                 if (!ext4_has_feature_encrypt(sb))
1049                         return -EOPNOTSUPP;
1050                 return fscrypt_ioctl_set_policy(filp, (const void __user *)arg);
1051
1052         case EXT4_IOC_GET_ENCRYPTION_PWSALT: {
1053 #ifdef CONFIG_FS_ENCRYPTION
1054                 int err, err2;
1055                 struct ext4_sb_info *sbi = EXT4_SB(sb);
1056                 handle_t *handle;
1057
1058                 if (!ext4_has_feature_encrypt(sb))
1059                         return -EOPNOTSUPP;
1060                 if (uuid_is_zero(sbi->s_es->s_encrypt_pw_salt)) {
1061                         err = mnt_want_write_file(filp);
1062                         if (err)
1063                                 return err;
1064                         handle = ext4_journal_start_sb(sb, EXT4_HT_MISC, 1);
1065                         if (IS_ERR(handle)) {
1066                                 err = PTR_ERR(handle);
1067                                 goto pwsalt_err_exit;
1068                         }
1069                         err = ext4_journal_get_write_access(handle, sbi->s_sbh);
1070                         if (err)
1071                                 goto pwsalt_err_journal;
1072                         generate_random_uuid(sbi->s_es->s_encrypt_pw_salt);
1073                         err = ext4_handle_dirty_metadata(handle, NULL,
1074                                                          sbi->s_sbh);
1075                 pwsalt_err_journal:
1076                         err2 = ext4_journal_stop(handle);
1077                         if (err2 && !err)
1078                                 err = err2;
1079                 pwsalt_err_exit:
1080                         mnt_drop_write_file(filp);
1081                         if (err)
1082                                 return err;
1083                 }
1084                 if (copy_to_user((void __user *) arg,
1085                                  sbi->s_es->s_encrypt_pw_salt, 16))
1086                         return -EFAULT;
1087                 return 0;
1088 #else
1089                 return -EOPNOTSUPP;
1090 #endif
1091         }
1092         case EXT4_IOC_GET_ENCRYPTION_POLICY:
1093                 return fscrypt_ioctl_get_policy(filp, (void __user *)arg);
1094
1095         case EXT4_IOC_FSGETXATTR:
1096         {
1097                 struct fsxattr fa;
1098
1099                 memset(&fa, 0, sizeof(struct fsxattr));
1100                 fa.fsx_xflags = ext4_iflags_to_xflags(ei->i_flags & EXT4_FL_USER_VISIBLE);
1101
1102                 if (ext4_has_feature_project(inode->i_sb)) {
1103                         fa.fsx_projid = (__u32)from_kprojid(&init_user_ns,
1104                                 EXT4_I(inode)->i_projid);
1105                 }
1106
1107                 if (copy_to_user((struct fsxattr __user *)arg,
1108                                  &fa, sizeof(fa)))
1109                         return -EFAULT;
1110                 return 0;
1111         }
1112         case EXT4_IOC_FSSETXATTR:
1113         {
1114                 struct fsxattr fa;
1115                 int err;
1116
1117                 if (copy_from_user(&fa, (struct fsxattr __user *)arg,
1118                                    sizeof(fa)))
1119                         return -EFAULT;
1120
1121                 /* Make sure caller has proper permission */
1122                 if (!inode_owner_or_capable(inode))
1123                         return -EACCES;
1124
1125                 if (fa.fsx_xflags & ~EXT4_SUPPORTED_FS_XFLAGS)
1126                         return -EOPNOTSUPP;
1127
1128                 flags = ext4_xflags_to_iflags(fa.fsx_xflags);
1129                 if (ext4_mask_flags(inode->i_mode, flags) != flags)
1130                         return -EOPNOTSUPP;
1131
1132                 err = mnt_want_write_file(filp);
1133                 if (err)
1134                         return err;
1135
1136                 inode_lock(inode);
1137                 err = ext4_ioctl_check_project(inode, &fa);
1138                 if (err)
1139                         goto out;
1140                 flags = (ei->i_flags & ~EXT4_FL_XFLAG_VISIBLE) |
1141                          (flags & EXT4_FL_XFLAG_VISIBLE);
1142                 err = ext4_ioctl_setflags(inode, flags);
1143                 if (err)
1144                         goto out;
1145                 err = ext4_ioctl_setproject(filp, fa.fsx_projid);
1146 out:
1147                 inode_unlock(inode);
1148                 mnt_drop_write_file(filp);
1149                 return err;
1150         }
1151         case EXT4_IOC_SHUTDOWN:
1152                 return ext4_shutdown(sb, arg);
1153         default:
1154                 return -ENOTTY;
1155         }
1156 }
1157
1158 #ifdef CONFIG_COMPAT
1159 long ext4_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
1160 {
1161         /* These are just misnamed, they actually get/put from/to user an int */
1162         switch (cmd) {
1163         case EXT4_IOC32_GETFLAGS:
1164                 cmd = EXT4_IOC_GETFLAGS;
1165                 break;
1166         case EXT4_IOC32_SETFLAGS:
1167                 cmd = EXT4_IOC_SETFLAGS;
1168                 break;
1169         case EXT4_IOC32_GETVERSION:
1170                 cmd = EXT4_IOC_GETVERSION;
1171                 break;
1172         case EXT4_IOC32_SETVERSION:
1173                 cmd = EXT4_IOC_SETVERSION;
1174                 break;
1175         case EXT4_IOC32_GROUP_EXTEND:
1176                 cmd = EXT4_IOC_GROUP_EXTEND;
1177                 break;
1178         case EXT4_IOC32_GETVERSION_OLD:
1179                 cmd = EXT4_IOC_GETVERSION_OLD;
1180                 break;
1181         case EXT4_IOC32_SETVERSION_OLD:
1182                 cmd = EXT4_IOC_SETVERSION_OLD;
1183                 break;
1184         case EXT4_IOC32_GETRSVSZ:
1185                 cmd = EXT4_IOC_GETRSVSZ;
1186                 break;
1187         case EXT4_IOC32_SETRSVSZ:
1188                 cmd = EXT4_IOC_SETRSVSZ;
1189                 break;
1190         case EXT4_IOC32_GROUP_ADD: {
1191                 struct compat_ext4_new_group_input __user *uinput;
1192                 struct ext4_new_group_data input;
1193                 int err;
1194
1195                 uinput = compat_ptr(arg);
1196                 err = get_user(input.group, &uinput->group);
1197                 err |= get_user(input.block_bitmap, &uinput->block_bitmap);
1198                 err |= get_user(input.inode_bitmap, &uinput->inode_bitmap);
1199                 err |= get_user(input.inode_table, &uinput->inode_table);
1200                 err |= get_user(input.blocks_count, &uinput->blocks_count);
1201                 err |= get_user(input.reserved_blocks,
1202                                 &uinput->reserved_blocks);
1203                 if (err)
1204                         return -EFAULT;
1205                 return ext4_ioctl_group_add(file, &input);
1206         }
1207         case EXT4_IOC_MOVE_EXT:
1208         case EXT4_IOC_RESIZE_FS:
1209         case EXT4_IOC_PRECACHE_EXTENTS:
1210         case EXT4_IOC_SET_ENCRYPTION_POLICY:
1211         case EXT4_IOC_GET_ENCRYPTION_PWSALT:
1212         case EXT4_IOC_GET_ENCRYPTION_POLICY:
1213         case EXT4_IOC_SHUTDOWN:
1214         case FS_IOC_GETFSMAP:
1215                 break;
1216         default:
1217                 return -ENOIOCTLCMD;
1218         }
1219         return ext4_ioctl(file, cmd, (unsigned long) compat_ptr(arg));
1220 }
1221 #endif