1.. SPDX-License-Identifier: GPL-2.0 2 3.. _fsverity: 4 5======================================================= 6fs-verity: read-only file-based authenticity protection 7======================================================= 8 9Introduction 10============ 11 12fs-verity (``fs/verity/``) is a support layer that filesystems can 13hook into to support transparent integrity and authenticity protection 14of read-only files. Currently, it is supported by the ext4, f2fs, and 15btrfs filesystems. Like fscrypt, not too much filesystem-specific 16code is needed to support fs-verity. 17 18fs-verity is similar to `dm-verity 19<https://www.kernel.org/doc/Documentation/device-mapper/verity.txt>`_ 20but works on files rather than block devices. On regular files on 21filesystems supporting fs-verity, userspace can execute an ioctl that 22causes the filesystem to build a Merkle tree for the file and persist 23it to a filesystem-specific location associated with the file. 24 25After this, the file is made readonly, and all reads from the file are 26automatically verified against the file's Merkle tree. Reads of any 27corrupted data, including mmap reads, will fail. 28 29Userspace can use another ioctl to retrieve the root hash (actually 30the "fs-verity file digest", which is a hash that includes the Merkle 31tree root hash) that fs-verity is enforcing for the file. This ioctl 32executes in constant time, regardless of the file size. 33 34fs-verity is essentially a way to hash a file in constant time, 35subject to the caveat that reads which would violate the hash will 36fail at runtime. 37 38Use cases 39========= 40 41By itself, fs-verity only provides integrity protection, i.e. 42detection of accidental (non-malicious) corruption. 43 44However, because fs-verity makes retrieving the file hash extremely 45efficient, it's primarily meant to be used as a tool to support 46authentication (detection of malicious modifications) or auditing 47(logging file hashes before use). 48 49A standard file hash could be used instead of fs-verity. However, 50this is inefficient if the file is large and only a small portion may 51be accessed. This is often the case for Android application package 52(APK) files, for example. These typically contain many translations, 53classes, and other resources that are infrequently or even never 54accessed on a particular device. It would be slow and wasteful to 55read and hash the entire file before starting the application. 56 57Unlike an ahead-of-time hash, fs-verity also re-verifies data each 58time it's paged in. This ensures that malicious disk firmware can't 59undetectably change the contents of the file at runtime. 60 61fs-verity does not replace or obsolete dm-verity. dm-verity should 62still be used on read-only filesystems. fs-verity is for files that 63must live on a read-write filesystem because they are independently 64updated and potentially user-installed, so dm-verity cannot be used. 65 66fs-verity does not mandate a particular scheme for authenticating its 67file hashes. (Similarly, dm-verity does not mandate a particular 68scheme for authenticating its block device root hashes.) Options for 69authenticating fs-verity file hashes include: 70 71- Trusted userspace code. Often, the userspace code that accesses 72 files can be trusted to authenticate them. Consider e.g. an 73 application that wants to authenticate data files before using them, 74 or an application loader that is part of the operating system (which 75 is already authenticated in a different way, such as by being loaded 76 from a read-only partition that uses dm-verity) and that wants to 77 authenticate applications before loading them. In these cases, this 78 trusted userspace code can authenticate a file's contents by 79 retrieving its fs-verity digest using `FS_IOC_MEASURE_VERITY`_, then 80 verifying a signature of it using any userspace cryptographic 81 library that supports digital signatures. 82 83- Integrity Measurement Architecture (IMA). IMA supports fs-verity 84 file digests as an alternative to its traditional full file digests. 85 "IMA appraisal" enforces that files contain a valid, matching 86 signature in their "security.ima" extended attribute, as controlled 87 by the IMA policy. For more information, see the IMA documentation. 88 89- Trusted userspace code in combination with `Built-in signature 90 verification`_. This approach should be used only with great care. 91 92User API 93======== 94 95FS_IOC_ENABLE_VERITY 96-------------------- 97 98The FS_IOC_ENABLE_VERITY ioctl enables fs-verity on a file. It takes 99in a pointer to a struct fsverity_enable_arg, defined as 100follows:: 101 102 struct fsverity_enable_arg { 103 __u32 version; 104 __u32 hash_algorithm; 105 __u32 block_size; 106 __u32 salt_size; 107 __u64 salt_ptr; 108 __u32 sig_size; 109 __u32 __reserved1; 110 __u64 sig_ptr; 111 __u64 __reserved2[11]; 112 }; 113 114This structure contains the parameters of the Merkle tree to build for 115the file. It must be initialized as follows: 116 117- ``version`` must be 1. 118- ``hash_algorithm`` must be the identifier for the hash algorithm to 119 use for the Merkle tree, such as FS_VERITY_HASH_ALG_SHA256. See 120 ``include/uapi/linux/fsverity.h`` for the list of possible values. 121- ``block_size`` is the Merkle tree block size, in bytes. In Linux 122 v6.3 and later, this can be any power of 2 between (inclusively) 123 1024 and the minimum of the system page size and the filesystem 124 block size. In earlier versions, the page size was the only allowed 125 value. 126- ``salt_size`` is the size of the salt in bytes, or 0 if no salt is 127 provided. The salt is a value that is prepended to every hashed 128 block; it can be used to personalize the hashing for a particular 129 file or device. Currently the maximum salt size is 32 bytes. 130- ``salt_ptr`` is the pointer to the salt, or NULL if no salt is 131 provided. 132- ``sig_size`` is the size of the builtin signature in bytes, or 0 if no 133 builtin signature is provided. Currently the builtin signature is 134 (somewhat arbitrarily) limited to 16128 bytes. 135- ``sig_ptr`` is the pointer to the builtin signature, or NULL if no 136 builtin signature is provided. A builtin signature is only needed 137 if the `Built-in signature verification`_ feature is being used. It 138 is not needed for IMA appraisal, and it is not needed if the file 139 signature is being handled entirely in userspace. 140- All reserved fields must be zeroed. 141 142FS_IOC_ENABLE_VERITY causes the filesystem to build a Merkle tree for 143the file and persist it to a filesystem-specific location associated 144with the file, then mark the file as a verity file. This ioctl may 145take a long time to execute on large files, and it is interruptible by 146fatal signals. 147 148FS_IOC_ENABLE_VERITY checks for write access to the inode. However, 149it must be executed on an O_RDONLY file descriptor and no processes 150can have the file open for writing. Attempts to open the file for 151writing while this ioctl is executing will fail with ETXTBSY. (This 152is necessary to guarantee that no writable file descriptors will exist 153after verity is enabled, and to guarantee that the file's contents are 154stable while the Merkle tree is being built over it.) 155 156On success, FS_IOC_ENABLE_VERITY returns 0, and the file becomes a 157verity file. On failure (including the case of interruption by a 158fatal signal), no changes are made to the file. 159 160FS_IOC_ENABLE_VERITY can fail with the following errors: 161 162- ``EACCES``: the process does not have write access to the file 163- ``EBADMSG``: the builtin signature is malformed 164- ``EBUSY``: this ioctl is already running on the file 165- ``EEXIST``: the file already has verity enabled 166- ``EFAULT``: the caller provided inaccessible memory 167- ``EFBIG``: the file is too large to enable verity on 168- ``EINTR``: the operation was interrupted by a fatal signal 169- ``EINVAL``: unsupported version, hash algorithm, or block size; or 170 reserved bits are set; or the file descriptor refers to neither a 171 regular file nor a directory. 172- ``EISDIR``: the file descriptor refers to a directory 173- ``EKEYREJECTED``: the builtin signature doesn't match the file 174- ``EMSGSIZE``: the salt or builtin signature is too long 175- ``ENOKEY``: the ".fs-verity" keyring doesn't contain the certificate 176 needed to verify the builtin signature 177- ``ENOPKG``: fs-verity recognizes the hash algorithm, but it's not 178 available in the kernel's crypto API as currently configured (e.g. 179 for SHA-512, missing CONFIG_CRYPTO_SHA512). 180- ``ENOTTY``: this type of filesystem does not implement fs-verity 181- ``EOPNOTSUPP``: the kernel was not configured with fs-verity 182 support; or the filesystem superblock has not had the 'verity' 183 feature enabled on it; or the filesystem does not support fs-verity 184 on this file. (See `Filesystem support`_.) 185- ``EPERM``: the file is append-only; or, a builtin signature is 186 required and one was not provided. 187- ``EROFS``: the filesystem is read-only 188- ``ETXTBSY``: someone has the file open for writing. This can be the 189 caller's file descriptor, another open file descriptor, or the file 190 reference held by a writable memory map. 191 192FS_IOC_MEASURE_VERITY 193--------------------- 194 195The FS_IOC_MEASURE_VERITY ioctl retrieves the digest of a verity file. 196The fs-verity file digest is a cryptographic digest that identifies 197the file contents that are being enforced on reads; it is computed via 198a Merkle tree and is different from a traditional full-file digest. 199 200This ioctl takes in a pointer to a variable-length structure:: 201 202 struct fsverity_digest { 203 __u16 digest_algorithm; 204 __u16 digest_size; /* input/output */ 205 __u8 digest[]; 206 }; 207 208``digest_size`` is an input/output field. On input, it must be 209initialized to the number of bytes allocated for the variable-length 210``digest`` field. 211 212On success, 0 is returned and the kernel fills in the structure as 213follows: 214 215- ``digest_algorithm`` will be the hash algorithm used for the file 216 digest. It will match ``fsverity_enable_arg::hash_algorithm``. 217- ``digest_size`` will be the size of the digest in bytes, e.g. 32 218 for SHA-256. (This can be redundant with ``digest_algorithm``.) 219- ``digest`` will be the actual bytes of the digest. 220 221FS_IOC_MEASURE_VERITY is guaranteed to execute in constant time, 222regardless of the size of the file. 223 224FS_IOC_MEASURE_VERITY can fail with the following errors: 225 226- ``EFAULT``: the caller provided inaccessible memory 227- ``ENODATA``: the file is not a verity file 228- ``ENOTTY``: this type of filesystem does not implement fs-verity 229- ``EOPNOTSUPP``: the kernel was not configured with fs-verity 230 support, or the filesystem superblock has not had the 'verity' 231 feature enabled on it. (See `Filesystem support`_.) 232- ``EOVERFLOW``: the digest is longer than the specified 233 ``digest_size`` bytes. Try providing a larger buffer. 234 235FS_IOC_READ_VERITY_METADATA 236--------------------------- 237 238The FS_IOC_READ_VERITY_METADATA ioctl reads verity metadata from a 239verity file. This ioctl is available since Linux v5.12. 240 241This ioctl allows writing a server program that takes a verity file 242and serves it to a client program, such that the client can do its own 243fs-verity compatible verification of the file. This only makes sense 244if the client doesn't trust the server and if the server needs to 245provide the storage for the client. 246 247This is a fairly specialized use case, and most fs-verity users won't 248need this ioctl. 249 250This ioctl takes in a pointer to the following structure:: 251 252 #define FS_VERITY_METADATA_TYPE_MERKLE_TREE 1 253 #define FS_VERITY_METADATA_TYPE_DESCRIPTOR 2 254 #define FS_VERITY_METADATA_TYPE_SIGNATURE 3 255 256 struct fsverity_read_metadata_arg { 257 __u64 metadata_type; 258 __u64 offset; 259 __u64 length; 260 __u64 buf_ptr; 261 __u64 __reserved; 262 }; 263 264``metadata_type`` specifies the type of metadata to read: 265 266- ``FS_VERITY_METADATA_TYPE_MERKLE_TREE`` reads the blocks of the 267 Merkle tree. The blocks are returned in order from the root level 268 to the leaf level. Within each level, the blocks are returned in 269 the same order that their hashes are themselves hashed. 270 See `Merkle tree`_ for more information. 271 272- ``FS_VERITY_METADATA_TYPE_DESCRIPTOR`` reads the fs-verity 273 descriptor. See `fs-verity descriptor`_. 274 275- ``FS_VERITY_METADATA_TYPE_SIGNATURE`` reads the builtin signature 276 which was passed to FS_IOC_ENABLE_VERITY, if any. See `Built-in 277 signature verification`_. 278 279The semantics are similar to those of ``pread()``. ``offset`` 280specifies the offset in bytes into the metadata item to read from, and 281``length`` specifies the maximum number of bytes to read from the 282metadata item. ``buf_ptr`` is the pointer to the buffer to read into, 283cast to a 64-bit integer. ``__reserved`` must be 0. On success, the 284number of bytes read is returned. 0 is returned at the end of the 285metadata item. The returned length may be less than ``length``, for 286example if the ioctl is interrupted. 287 288The metadata returned by FS_IOC_READ_VERITY_METADATA isn't guaranteed 289to be authenticated against the file digest that would be returned by 290`FS_IOC_MEASURE_VERITY`_, as the metadata is expected to be used to 291implement fs-verity compatible verification anyway (though absent a 292malicious disk, the metadata will indeed match). E.g. to implement 293this ioctl, the filesystem is allowed to just read the Merkle tree 294blocks from disk without actually verifying the path to the root node. 295 296FS_IOC_READ_VERITY_METADATA can fail with the following errors: 297 298- ``EFAULT``: the caller provided inaccessible memory 299- ``EINTR``: the ioctl was interrupted before any data was read 300- ``EINVAL``: reserved fields were set, or ``offset + length`` 301 overflowed 302- ``ENODATA``: the file is not a verity file, or 303 FS_VERITY_METADATA_TYPE_SIGNATURE was requested but the file doesn't 304 have a builtin signature 305- ``ENOTTY``: this type of filesystem does not implement fs-verity, or 306 this ioctl is not yet implemented on it 307- ``EOPNOTSUPP``: the kernel was not configured with fs-verity 308 support, or the filesystem superblock has not had the 'verity' 309 feature enabled on it. (See `Filesystem support`_.) 310 311FS_IOC_GETFLAGS 312--------------- 313 314The existing ioctl FS_IOC_GETFLAGS (which isn't specific to fs-verity) 315can also be used to check whether a file has fs-verity enabled or not. 316To do so, check for FS_VERITY_FL (0x00100000) in the returned flags. 317 318The verity flag is not settable via FS_IOC_SETFLAGS. You must use 319FS_IOC_ENABLE_VERITY instead, since parameters must be provided. 320 321statx 322----- 323 324Since Linux v5.5, the statx() system call sets STATX_ATTR_VERITY if 325the file has fs-verity enabled. This can perform better than 326FS_IOC_GETFLAGS and FS_IOC_MEASURE_VERITY because it doesn't require 327opening the file, and opening verity files can be expensive. 328 329.. _accessing_verity_files: 330 331Accessing verity files 332====================== 333 334Applications can transparently access a verity file just like a 335non-verity one, with the following exceptions: 336 337- Verity files are readonly. They cannot be opened for writing or 338 truncate()d, even if the file mode bits allow it. Attempts to do 339 one of these things will fail with EPERM. However, changes to 340 metadata such as owner, mode, timestamps, and xattrs are still 341 allowed, since these are not measured by fs-verity. Verity files 342 can also still be renamed, deleted, and linked to. 343 344- Direct I/O is not supported on verity files. Attempts to use direct 345 I/O on such files will fall back to buffered I/O. 346 347- DAX (Direct Access) is not supported on verity files, because this 348 would circumvent the data verification. 349 350- Reads of data that doesn't match the verity Merkle tree will fail 351 with EIO (for read()) or SIGBUS (for mmap() reads). 352 353- If the sysctl "fs.verity.require_signatures" is set to 1 and the 354 file is not signed by a key in the ".fs-verity" keyring, then 355 opening the file will fail. See `Built-in signature verification`_. 356 357Direct access to the Merkle tree is not supported. Therefore, if a 358verity file is copied, or is backed up and restored, then it will lose 359its "verity"-ness. fs-verity is primarily meant for files like 360executables that are managed by a package manager. 361 362File digest computation 363======================= 364 365This section describes how fs-verity hashes the file contents using a 366Merkle tree to produce the digest which cryptographically identifies 367the file contents. This algorithm is the same for all filesystems 368that support fs-verity. 369 370Userspace only needs to be aware of this algorithm if it needs to 371compute fs-verity file digests itself, e.g. in order to sign files. 372 373.. _fsverity_merkle_tree: 374 375Merkle tree 376----------- 377 378The file contents is divided into blocks, where the block size is 379configurable but is usually 4096 bytes. The end of the last block is 380zero-padded if needed. Each block is then hashed, producing the first 381level of hashes. Then, the hashes in this first level are grouped 382into 'blocksize'-byte blocks (zero-padding the ends as needed) and 383these blocks are hashed, producing the second level of hashes. This 384proceeds up the tree until only a single block remains. The hash of 385this block is the "Merkle tree root hash". 386 387If the file fits in one block and is nonempty, then the "Merkle tree 388root hash" is simply the hash of the single data block. If the file 389is empty, then the "Merkle tree root hash" is all zeroes. 390 391The "blocks" here are not necessarily the same as "filesystem blocks". 392 393If a salt was specified, then it's zero-padded to the closest multiple 394of the input size of the hash algorithm's compression function, e.g. 39564 bytes for SHA-256 or 128 bytes for SHA-512. The padded salt is 396prepended to every data or Merkle tree block that is hashed. 397 398The purpose of the block padding is to cause every hash to be taken 399over the same amount of data, which simplifies the implementation and 400keeps open more possibilities for hardware acceleration. The purpose 401of the salt padding is to make the salting "free" when the salted hash 402state is precomputed, then imported for each hash. 403 404Example: in the recommended configuration of SHA-256 and 4K blocks, 405128 hash values fit in each block. Thus, each level of the Merkle 406tree is approximately 128 times smaller than the previous, and for 407large files the Merkle tree's size converges to approximately 1/127 of 408the original file size. However, for small files, the padding is 409significant, making the space overhead proportionally more. 410 411.. _fsverity_descriptor: 412 413fs-verity descriptor 414-------------------- 415 416By itself, the Merkle tree root hash is ambiguous. For example, it 417can't a distinguish a large file from a small second file whose data 418is exactly the top-level hash block of the first file. Ambiguities 419also arise from the convention of padding to the next block boundary. 420 421To solve this problem, the fs-verity file digest is actually computed 422as a hash of the following structure, which contains the Merkle tree 423root hash as well as other fields such as the file size:: 424 425 struct fsverity_descriptor { 426 __u8 version; /* must be 1 */ 427 __u8 hash_algorithm; /* Merkle tree hash algorithm */ 428 __u8 log_blocksize; /* log2 of size of data and tree blocks */ 429 __u8 salt_size; /* size of salt in bytes; 0 if none */ 430 __le32 __reserved_0x04; /* must be 0 */ 431 __le64 data_size; /* size of file the Merkle tree is built over */ 432 __u8 root_hash[64]; /* Merkle tree root hash */ 433 __u8 salt[32]; /* salt prepended to each hashed block */ 434 __u8 __reserved[144]; /* must be 0's */ 435 }; 436 437Built-in signature verification 438=============================== 439 440CONFIG_FS_VERITY_BUILTIN_SIGNATURES=y adds supports for in-kernel 441verification of fs-verity builtin signatures. 442 443**IMPORTANT**! Please take great care before using this feature. 444It is not the only way to do signatures with fs-verity, and the 445alternatives (such as userspace signature verification, and IMA 446appraisal) can be much better. It's also easy to fall into a trap 447of thinking this feature solves more problems than it actually does. 448 449Enabling this option adds the following: 450 4511. At boot time, the kernel creates a keyring named ".fs-verity". The 452 root user can add trusted X.509 certificates to this keyring using 453 the add_key() system call. 454 4552. `FS_IOC_ENABLE_VERITY`_ accepts a pointer to a PKCS#7 formatted 456 detached signature in DER format of the file's fs-verity digest. 457 On success, the ioctl persists the signature alongside the Merkle 458 tree. Then, any time the file is opened, the kernel verifies the 459 file's actual digest against this signature, using the certificates 460 in the ".fs-verity" keyring. 461 4623. A new sysctl "fs.verity.require_signatures" is made available. 463 When set to 1, the kernel requires that all verity files have a 464 correctly signed digest as described in (2). 465 466The data that the signature as described in (2) must be a signature of 467is the fs-verity file digest in the following format:: 468 469 struct fsverity_formatted_digest { 470 char magic[8]; /* must be "FSVerity" */ 471 __le16 digest_algorithm; 472 __le16 digest_size; 473 __u8 digest[]; 474 }; 475 476That's it. It should be emphasized again that fs-verity builtin 477signatures are not the only way to do signatures with fs-verity. See 478`Use cases`_ for an overview of ways in which fs-verity can be used. 479fs-verity builtin signatures have some major limitations that should 480be carefully considered before using them: 481 482- Builtin signature verification does *not* make the kernel enforce 483 that any files actually have fs-verity enabled. Thus, it is not a 484 complete authentication policy. Currently, if it is used, the only 485 way to complete the authentication policy is for trusted userspace 486 code to explicitly check whether files have fs-verity enabled with a 487 signature before they are accessed. (With 488 fs.verity.require_signatures=1, just checking whether fs-verity is 489 enabled suffices.) But, in this case the trusted userspace code 490 could just store the signature alongside the file and verify it 491 itself using a cryptographic library, instead of using this feature. 492 493- A file's builtin signature can only be set at the same time that 494 fs-verity is being enabled on the file. Changing or deleting the 495 builtin signature later requires re-creating the file. 496 497- Builtin signature verification uses the same set of public keys for 498 all fs-verity enabled files on the system. Different keys cannot be 499 trusted for different files; each key is all or nothing. 500 501- The sysctl fs.verity.require_signatures applies system-wide. 502 Setting it to 1 only works when all users of fs-verity on the system 503 agree that it should be set to 1. This limitation can prevent 504 fs-verity from being used in cases where it would be helpful. 505 506- Builtin signature verification can only use signature algorithms 507 that are supported by the kernel. For example, the kernel does not 508 yet support Ed25519, even though this is often the signature 509 algorithm that is recommended for new cryptographic designs. 510 511- fs-verity builtin signatures are in PKCS#7 format, and the public 512 keys are in X.509 format. These formats are commonly used, 513 including by some other kernel features (which is why the fs-verity 514 builtin signatures use them), and are very feature rich. 515 Unfortunately, history has shown that code that parses and handles 516 these formats (which are from the 1990s and are based on ASN.1) 517 often has vulnerabilities as a result of their complexity. This 518 complexity is not inherent to the cryptography itself. 519 520 fs-verity users who do not need advanced features of X.509 and 521 PKCS#7 should strongly consider using simpler formats, such as plain 522 Ed25519 keys and signatures, and verifying signatures in userspace. 523 524 fs-verity users who choose to use X.509 and PKCS#7 anyway should 525 still consider that verifying those signatures in userspace is more 526 flexible (for other reasons mentioned earlier in this document) and 527 eliminates the need to enable CONFIG_FS_VERITY_BUILTIN_SIGNATURES 528 and its associated increase in kernel attack surface. In some cases 529 it can even be necessary, since advanced X.509 and PKCS#7 features 530 do not always work as intended with the kernel. For example, the 531 kernel does not check X.509 certificate validity times. 532 533 Note: IMA appraisal, which supports fs-verity, does not use PKCS#7 534 for its signatures, so it partially avoids the issues discussed 535 here. IMA appraisal does use X.509. 536 537Filesystem support 538================== 539 540fs-verity is supported by several filesystems, described below. The 541CONFIG_FS_VERITY kconfig option must be enabled to use fs-verity on 542any of these filesystems. 543 544``include/linux/fsverity.h`` declares the interface between the 545``fs/verity/`` support layer and filesystems. Briefly, filesystems 546must provide an ``fsverity_operations`` structure that provides 547methods to read and write the verity metadata to a filesystem-specific 548location, including the Merkle tree blocks and 549``fsverity_descriptor``. Filesystems must also call functions in 550``fs/verity/`` at certain times, such as when a file is opened or when 551pages have been read into the pagecache. (See `Verifying data`_.) 552 553ext4 554---- 555 556ext4 supports fs-verity since Linux v5.4 and e2fsprogs v1.45.2. 557 558To create verity files on an ext4 filesystem, the filesystem must have 559been formatted with ``-O verity`` or had ``tune2fs -O verity`` run on 560it. "verity" is an RO_COMPAT filesystem feature, so once set, old 561kernels will only be able to mount the filesystem readonly, and old 562versions of e2fsck will be unable to check the filesystem. 563 564Originally, an ext4 filesystem with the "verity" feature could only be 565mounted when its block size was equal to the system page size 566(typically 4096 bytes). In Linux v6.3, this limitation was removed. 567 568ext4 sets the EXT4_VERITY_FL on-disk inode flag on verity files. It 569can only be set by `FS_IOC_ENABLE_VERITY`_, and it cannot be cleared. 570 571ext4 also supports encryption, which can be used simultaneously with 572fs-verity. In this case, the plaintext data is verified rather than 573the ciphertext. This is necessary in order to make the fs-verity file 574digest meaningful, since every file is encrypted differently. 575 576ext4 stores the verity metadata (Merkle tree and fsverity_descriptor) 577past the end of the file, starting at the first 64K boundary beyond 578i_size. This approach works because (a) verity files are readonly, 579and (b) pages fully beyond i_size aren't visible to userspace but can 580be read/written internally by ext4 with only some relatively small 581changes to ext4. This approach avoids having to depend on the 582EA_INODE feature and on rearchitecturing ext4's xattr support to 583support paging multi-gigabyte xattrs into memory, and to support 584encrypting xattrs. Note that the verity metadata *must* be encrypted 585when the file is, since it contains hashes of the plaintext data. 586 587ext4 only allows verity on extent-based files. 588 589f2fs 590---- 591 592f2fs supports fs-verity since Linux v5.4 and f2fs-tools v1.11.0. 593 594To create verity files on an f2fs filesystem, the filesystem must have 595been formatted with ``-O verity``. 596 597f2fs sets the FADVISE_VERITY_BIT on-disk inode flag on verity files. 598It can only be set by `FS_IOC_ENABLE_VERITY`_, and it cannot be 599cleared. 600 601Like ext4, f2fs stores the verity metadata (Merkle tree and 602fsverity_descriptor) past the end of the file, starting at the first 60364K boundary beyond i_size. See explanation for ext4 above. 604Moreover, f2fs supports at most 4096 bytes of xattr entries per inode 605which usually wouldn't be enough for even a single Merkle tree block. 606 607f2fs doesn't support enabling verity on files that currently have 608atomic or volatile writes pending. 609 610btrfs 611----- 612 613btrfs supports fs-verity since Linux v5.15. Verity-enabled inodes are 614marked with a RO_COMPAT inode flag, and the verity metadata is stored 615in separate btree items. 616 617Implementation details 618====================== 619 620Verifying data 621-------------- 622 623fs-verity ensures that all reads of a verity file's data are verified, 624regardless of which syscall is used to do the read (e.g. mmap(), 625read(), pread()) and regardless of whether it's the first read or a 626later read (unless the later read can return cached data that was 627already verified). Below, we describe how filesystems implement this. 628 629Pagecache 630~~~~~~~~~ 631 632For filesystems using Linux's pagecache, the ``->read_folio()`` and 633``->readahead()`` methods must be modified to verify folios before 634they are marked Uptodate. Merely hooking ``->read_iter()`` would be 635insufficient, since ``->read_iter()`` is not used for memory maps. 636 637Therefore, fs/verity/ provides the function fsverity_verify_blocks() 638which verifies data that has been read into the pagecache of a verity 639inode. The containing folio must still be locked and not Uptodate, so 640it's not yet readable by userspace. As needed to do the verification, 641fsverity_verify_blocks() will call back into the filesystem to read 642hash blocks via fsverity_operations::read_merkle_tree_page(). 643 644fsverity_verify_blocks() returns false if verification failed; in this 645case, the filesystem must not set the folio Uptodate. Following this, 646as per the usual Linux pagecache behavior, attempts by userspace to 647read() from the part of the file containing the folio will fail with 648EIO, and accesses to the folio within a memory map will raise SIGBUS. 649 650In principle, verifying a data block requires verifying the entire 651path in the Merkle tree from the data block to the root hash. 652However, for efficiency the filesystem may cache the hash blocks. 653Therefore, fsverity_verify_blocks() only ascends the tree reading hash 654blocks until an already-verified hash block is seen. It then verifies 655the path to that block. 656 657This optimization, which is also used by dm-verity, results in 658excellent sequential read performance. This is because usually (e.g. 659127 in 128 times for 4K blocks and SHA-256) the hash block from the 660bottom level of the tree will already be cached and checked from 661reading a previous data block. However, random reads perform worse. 662 663Block device based filesystems 664~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 665 666Block device based filesystems (e.g. ext4 and f2fs) in Linux also use 667the pagecache, so the above subsection applies too. However, they 668also usually read many data blocks from a file at once, grouped into a 669structure called a "bio". To make it easier for these types of 670filesystems to support fs-verity, fs/verity/ also provides a function 671fsverity_verify_bio() which verifies all data blocks in a bio. 672 673ext4 and f2fs also support encryption. If a verity file is also 674encrypted, the data must be decrypted before being verified. To 675support this, these filesystems allocate a "post-read context" for 676each bio and store it in ``->bi_private``:: 677 678 struct bio_post_read_ctx { 679 struct bio *bio; 680 struct work_struct work; 681 unsigned int cur_step; 682 unsigned int enabled_steps; 683 }; 684 685``enabled_steps`` is a bitmask that specifies whether decryption, 686verity, or both is enabled. After the bio completes, for each needed 687postprocessing step the filesystem enqueues the bio_post_read_ctx on a 688workqueue, and then the workqueue work does the decryption or 689verification. Finally, folios where no decryption or verity error 690occurred are marked Uptodate, and the folios are unlocked. 691 692On many filesystems, files can contain holes. Normally, 693``->readahead()`` simply zeroes hole blocks and considers the 694corresponding data to be up-to-date; no bios are issued. To prevent 695this case from bypassing fs-verity, filesystems use 696fsverity_verify_blocks() to verify hole blocks. 697 698Filesystems also disable direct I/O on verity files, since otherwise 699direct I/O would bypass fs-verity. 700 701Userspace utility 702================= 703 704This document focuses on the kernel, but a userspace utility for 705fs-verity can be found at: 706 707 https://git.kernel.org/pub/scm/fs/fsverity/fsverity-utils.git 708 709See the README.md file in the fsverity-utils source tree for details, 710including examples of setting up fs-verity protected files. 711 712Tests 713===== 714 715To test fs-verity, use xfstests. For example, using `kvm-xfstests 716<https://github.com/tytso/xfstests-bld/blob/master/Documentation/kvm-quickstart.md>`_:: 717 718 kvm-xfstests -c ext4,f2fs,btrfs -g verity 719 720FAQ 721=== 722 723This section answers frequently asked questions about fs-verity that 724weren't already directly answered in other parts of this document. 725 726:Q: Why isn't fs-verity part of IMA? 727:A: fs-verity and IMA (Integrity Measurement Architecture) have 728 different focuses. fs-verity is a filesystem-level mechanism for 729 hashing individual files using a Merkle tree. In contrast, IMA 730 specifies a system-wide policy that specifies which files are 731 hashed and what to do with those hashes, such as log them, 732 authenticate them, or add them to a measurement list. 733 734 IMA supports the fs-verity hashing mechanism as an alternative 735 to full file hashes, for those who want the performance and 736 security benefits of the Merkle tree based hash. However, it 737 doesn't make sense to force all uses of fs-verity to be through 738 IMA. fs-verity already meets many users' needs even as a 739 standalone filesystem feature, and it's testable like other 740 filesystem features e.g. with xfstests. 741 742:Q: Isn't fs-verity useless because the attacker can just modify the 743 hashes in the Merkle tree, which is stored on-disk? 744:A: To verify the authenticity of an fs-verity file you must verify 745 the authenticity of the "fs-verity file digest", which 746 incorporates the root hash of the Merkle tree. See `Use cases`_. 747 748:Q: Isn't fs-verity useless because the attacker can just replace a 749 verity file with a non-verity one? 750:A: See `Use cases`_. In the initial use case, it's really trusted 751 userspace code that authenticates the files; fs-verity is just a 752 tool to do this job efficiently and securely. The trusted 753 userspace code will consider non-verity files to be inauthentic. 754 755:Q: Why does the Merkle tree need to be stored on-disk? Couldn't you 756 store just the root hash? 757:A: If the Merkle tree wasn't stored on-disk, then you'd have to 758 compute the entire tree when the file is first accessed, even if 759 just one byte is being read. This is a fundamental consequence of 760 how Merkle tree hashing works. To verify a leaf node, you need to 761 verify the whole path to the root hash, including the root node 762 (the thing which the root hash is a hash of). But if the root 763 node isn't stored on-disk, you have to compute it by hashing its 764 children, and so on until you've actually hashed the entire file. 765 766 That defeats most of the point of doing a Merkle tree-based hash, 767 since if you have to hash the whole file ahead of time anyway, 768 then you could simply do sha256(file) instead. That would be much 769 simpler, and a bit faster too. 770 771 It's true that an in-memory Merkle tree could still provide the 772 advantage of verification on every read rather than just on the 773 first read. However, it would be inefficient because every time a 774 hash page gets evicted (you can't pin the entire Merkle tree into 775 memory, since it may be very large), in order to restore it you 776 again need to hash everything below it in the tree. This again 777 defeats most of the point of doing a Merkle tree-based hash, since 778 a single block read could trigger re-hashing gigabytes of data. 779 780:Q: But couldn't you store just the leaf nodes and compute the rest? 781:A: See previous answer; this really just moves up one level, since 782 one could alternatively interpret the data blocks as being the 783 leaf nodes of the Merkle tree. It's true that the tree can be 784 computed much faster if the leaf level is stored rather than just 785 the data, but that's only because each level is less than 1% the 786 size of the level below (assuming the recommended settings of 787 SHA-256 and 4K blocks). For the exact same reason, by storing 788 "just the leaf nodes" you'd already be storing over 99% of the 789 tree, so you might as well simply store the whole tree. 790 791:Q: Can the Merkle tree be built ahead of time, e.g. distributed as 792 part of a package that is installed to many computers? 793:A: This isn't currently supported. It was part of the original 794 design, but was removed to simplify the kernel UAPI and because it 795 wasn't a critical use case. Files are usually installed once and 796 used many times, and cryptographic hashing is somewhat fast on 797 most modern processors. 798 799:Q: Why doesn't fs-verity support writes? 800:A: Write support would be very difficult and would require a 801 completely different design, so it's well outside the scope of 802 fs-verity. Write support would require: 803 804 - A way to maintain consistency between the data and hashes, 805 including all levels of hashes, since corruption after a crash 806 (especially of potentially the entire file!) is unacceptable. 807 The main options for solving this are data journalling, 808 copy-on-write, and log-structured volume. But it's very hard to 809 retrofit existing filesystems with new consistency mechanisms. 810 Data journalling is available on ext4, but is very slow. 811 812 - Rebuilding the Merkle tree after every write, which would be 813 extremely inefficient. Alternatively, a different authenticated 814 dictionary structure such as an "authenticated skiplist" could 815 be used. However, this would be far more complex. 816 817 Compare it to dm-verity vs. dm-integrity. dm-verity is very 818 simple: the kernel just verifies read-only data against a 819 read-only Merkle tree. In contrast, dm-integrity supports writes 820 but is slow, is much more complex, and doesn't actually support 821 full-device authentication since it authenticates each sector 822 independently, i.e. there is no "root hash". It doesn't really 823 make sense for the same device-mapper target to support these two 824 very different cases; the same applies to fs-verity. 825 826:Q: Since verity files are immutable, why isn't the immutable bit set? 827:A: The existing "immutable" bit (FS_IMMUTABLE_FL) already has a 828 specific set of semantics which not only make the file contents 829 read-only, but also prevent the file from being deleted, renamed, 830 linked to, or having its owner or mode changed. These extra 831 properties are unwanted for fs-verity, so reusing the immutable 832 bit isn't appropriate. 833 834:Q: Why does the API use ioctls instead of setxattr() and getxattr()? 835:A: Abusing the xattr interface for basically arbitrary syscalls is 836 heavily frowned upon by most of the Linux filesystem developers. 837 An xattr should really just be an xattr on-disk, not an API to 838 e.g. magically trigger construction of a Merkle tree. 839 840:Q: Does fs-verity support remote filesystems? 841:A: So far all filesystems that have implemented fs-verity support are 842 local filesystems, but in principle any filesystem that can store 843 per-file verity metadata can support fs-verity, regardless of 844 whether it's local or remote. Some filesystems may have fewer 845 options of where to store the verity metadata; one possibility is 846 to store it past the end of the file and "hide" it from userspace 847 by manipulating i_size. The data verification functions provided 848 by ``fs/verity/`` also assume that the filesystem uses the Linux 849 pagecache, but both local and remote filesystems normally do so. 850 851:Q: Why is anything filesystem-specific at all? Shouldn't fs-verity 852 be implemented entirely at the VFS level? 853:A: There are many reasons why this is not possible or would be very 854 difficult, including the following: 855 856 - To prevent bypassing verification, folios must not be marked 857 Uptodate until they've been verified. Currently, each 858 filesystem is responsible for marking folios Uptodate via 859 ``->readahead()``. Therefore, currently it's not possible for 860 the VFS to do the verification on its own. Changing this would 861 require significant changes to the VFS and all filesystems. 862 863 - It would require defining a filesystem-independent way to store 864 the verity metadata. Extended attributes don't work for this 865 because (a) the Merkle tree may be gigabytes, but many 866 filesystems assume that all xattrs fit into a single 4K 867 filesystem block, and (b) ext4 and f2fs encryption doesn't 868 encrypt xattrs, yet the Merkle tree *must* be encrypted when the 869 file contents are, because it stores hashes of the plaintext 870 file contents. 871 872 So the verity metadata would have to be stored in an actual 873 file. Using a separate file would be very ugly, since the 874 metadata is fundamentally part of the file to be protected, and 875 it could cause problems where users could delete the real file 876 but not the metadata file or vice versa. On the other hand, 877 having it be in the same file would break applications unless 878 filesystems' notion of i_size were divorced from the VFS's, 879 which would be complex and require changes to all filesystems. 880 881 - It's desirable that FS_IOC_ENABLE_VERITY uses the filesystem's 882 transaction mechanism so that either the file ends up with 883 verity enabled, or no changes were made. Allowing intermediate 884 states to occur after a crash may cause problems. 885