1 .. SPDX-License-Identifier: GPL-2.0
5 =======================================================
6 fs-verity: read-only file-based authenticity protection
7 =======================================================
12 fs-verity (``fs/verity/``) is a support layer that filesystems can
13 hook into to support transparent integrity and authenticity protection
14 of read-only files. Currently, it is supported by the ext4 and f2fs
15 filesystems. Like fscrypt, not too much filesystem-specific code is
16 needed to support fs-verity.
18 fs-verity is similar to `dm-verity
19 <https://www.kernel.org/doc/Documentation/device-mapper/verity.txt>`_
20 but works on files rather than block devices. On regular files on
21 filesystems supporting fs-verity, userspace can execute an ioctl that
22 causes the filesystem to build a Merkle tree for the file and persist
23 it to a filesystem-specific location associated with the file.
25 After this, the file is made readonly, and all reads from the file are
26 automatically verified against the file's Merkle tree. Reads of any
27 corrupted data, including mmap reads, will fail.
29 Userspace can use another ioctl to retrieve the root hash (actually
30 the "fs-verity file digest", which is a hash that includes the Merkle
31 tree root hash) that fs-verity is enforcing for the file. This ioctl
32 executes in constant time, regardless of the file size.
34 fs-verity is essentially a way to hash a file in constant time,
35 subject to the caveat that reads which would violate the hash will
41 By itself, the base fs-verity feature only provides integrity
42 protection, i.e. detection of accidental (non-malicious) corruption.
44 However, because fs-verity makes retrieving the file hash extremely
45 efficient, it's primarily meant to be used as a tool to support
46 authentication (detection of malicious modifications) or auditing
47 (logging file hashes before use).
49 Trusted userspace code (e.g. operating system code running on a
50 read-only partition that is itself authenticated by dm-verity) can
51 authenticate the contents of an fs-verity file by using the
52 `FS_IOC_MEASURE_VERITY`_ ioctl to retrieve its hash, then verifying a
53 digital signature of it.
55 A standard file hash could be used instead of fs-verity. However,
56 this is inefficient if the file is large and only a small portion may
57 be accessed. This is often the case for Android application package
58 (APK) files, for example. These typically contain many translations,
59 classes, and other resources that are infrequently or even never
60 accessed on a particular device. It would be slow and wasteful to
61 read and hash the entire file before starting the application.
63 Unlike an ahead-of-time hash, fs-verity also re-verifies data each
64 time it's paged in. This ensures that malicious disk firmware can't
65 undetectably change the contents of the file at runtime.
67 fs-verity does not replace or obsolete dm-verity. dm-verity should
68 still be used on read-only filesystems. fs-verity is for files that
69 must live on a read-write filesystem because they are independently
70 updated and potentially user-installed, so dm-verity cannot be used.
72 The base fs-verity feature is a hashing mechanism only; actually
73 authenticating the files is up to userspace. However, to meet some
74 users' needs, fs-verity optionally supports a simple signature
75 verification mechanism where users can configure the kernel to require
76 that all fs-verity files be signed by a key loaded into a keyring; see
77 `Built-in signature verification`_. Support for fs-verity file hashes
78 in IMA (Integrity Measurement Architecture) policies is also planned.
86 The FS_IOC_ENABLE_VERITY ioctl enables fs-verity on a file. It takes
87 in a pointer to a struct fsverity_enable_arg, defined as
90 struct fsverity_enable_arg {
99 __u64 __reserved2[11];
102 This structure contains the parameters of the Merkle tree to build for
103 the file, and optionally contains a signature. It must be initialized
106 - ``version`` must be 1.
107 - ``hash_algorithm`` must be the identifier for the hash algorithm to
108 use for the Merkle tree, such as FS_VERITY_HASH_ALG_SHA256. See
109 ``include/uapi/linux/fsverity.h`` for the list of possible values.
110 - ``block_size`` must be the Merkle tree block size. Currently, this
111 must be equal to the system page size, which is usually 4096 bytes.
112 Other sizes may be supported in the future. This value is not
113 necessarily the same as the filesystem block size.
114 - ``salt_size`` is the size of the salt in bytes, or 0 if no salt is
115 provided. The salt is a value that is prepended to every hashed
116 block; it can be used to personalize the hashing for a particular
117 file or device. Currently the maximum salt size is 32 bytes.
118 - ``salt_ptr`` is the pointer to the salt, or NULL if no salt is
120 - ``sig_size`` is the size of the signature in bytes, or 0 if no
121 signature is provided. Currently the signature is (somewhat
122 arbitrarily) limited to 16128 bytes. See `Built-in signature
123 verification`_ for more information.
124 - ``sig_ptr`` is the pointer to the signature, or NULL if no
125 signature is provided.
126 - All reserved fields must be zeroed.
128 FS_IOC_ENABLE_VERITY causes the filesystem to build a Merkle tree for
129 the file and persist it to a filesystem-specific location associated
130 with the file, then mark the file as a verity file. This ioctl may
131 take a long time to execute on large files, and it is interruptible by
134 FS_IOC_ENABLE_VERITY checks for write access to the inode. However,
135 it must be executed on an O_RDONLY file descriptor and no processes
136 can have the file open for writing. Attempts to open the file for
137 writing while this ioctl is executing will fail with ETXTBSY. (This
138 is necessary to guarantee that no writable file descriptors will exist
139 after verity is enabled, and to guarantee that the file's contents are
140 stable while the Merkle tree is being built over it.)
142 On success, FS_IOC_ENABLE_VERITY returns 0, and the file becomes a
143 verity file. On failure (including the case of interruption by a
144 fatal signal), no changes are made to the file.
146 FS_IOC_ENABLE_VERITY can fail with the following errors:
148 - ``EACCES``: the process does not have write access to the file
149 - ``EBADMSG``: the signature is malformed
150 - ``EBUSY``: this ioctl is already running on the file
151 - ``EEXIST``: the file already has verity enabled
152 - ``EFAULT``: the caller provided inaccessible memory
153 - ``EINTR``: the operation was interrupted by a fatal signal
154 - ``EINVAL``: unsupported version, hash algorithm, or block size; or
155 reserved bits are set; or the file descriptor refers to neither a
156 regular file nor a directory.
157 - ``EISDIR``: the file descriptor refers to a directory
158 - ``EKEYREJECTED``: the signature doesn't match the file
159 - ``EMSGSIZE``: the salt or signature is too long
160 - ``ENOKEY``: the fs-verity keyring doesn't contain the certificate
161 needed to verify the signature
162 - ``ENOPKG``: fs-verity recognizes the hash algorithm, but it's not
163 available in the kernel's crypto API as currently configured (e.g.
164 for SHA-512, missing CONFIG_CRYPTO_SHA512).
165 - ``ENOTTY``: this type of filesystem does not implement fs-verity
166 - ``EOPNOTSUPP``: the kernel was not configured with fs-verity
167 support; or the filesystem superblock has not had the 'verity'
168 feature enabled on it; or the filesystem does not support fs-verity
169 on this file. (See `Filesystem support`_.)
170 - ``EPERM``: the file is append-only; or, a signature is required and
171 one was not provided.
172 - ``EROFS``: the filesystem is read-only
173 - ``ETXTBSY``: someone has the file open for writing. This can be the
174 caller's file descriptor, another open file descriptor, or the file
175 reference held by a writable memory map.
177 FS_IOC_MEASURE_VERITY
178 ---------------------
180 The FS_IOC_MEASURE_VERITY ioctl retrieves the digest of a verity file.
181 The fs-verity file digest is a cryptographic digest that identifies
182 the file contents that are being enforced on reads; it is computed via
183 a Merkle tree and is different from a traditional full-file digest.
185 This ioctl takes in a pointer to a variable-length structure::
187 struct fsverity_digest {
188 __u16 digest_algorithm;
189 __u16 digest_size; /* input/output */
193 ``digest_size`` is an input/output field. On input, it must be
194 initialized to the number of bytes allocated for the variable-length
197 On success, 0 is returned and the kernel fills in the structure as
200 - ``digest_algorithm`` will be the hash algorithm used for the file
201 digest. It will match ``fsverity_enable_arg::hash_algorithm``.
202 - ``digest_size`` will be the size of the digest in bytes, e.g. 32
203 for SHA-256. (This can be redundant with ``digest_algorithm``.)
204 - ``digest`` will be the actual bytes of the digest.
206 FS_IOC_MEASURE_VERITY is guaranteed to execute in constant time,
207 regardless of the size of the file.
209 FS_IOC_MEASURE_VERITY can fail with the following errors:
211 - ``EFAULT``: the caller provided inaccessible memory
212 - ``ENODATA``: the file is not a verity file
213 - ``ENOTTY``: this type of filesystem does not implement fs-verity
214 - ``EOPNOTSUPP``: the kernel was not configured with fs-verity
215 support, or the filesystem superblock has not had the 'verity'
216 feature enabled on it. (See `Filesystem support`_.)
217 - ``EOVERFLOW``: the digest is longer than the specified
218 ``digest_size`` bytes. Try providing a larger buffer.
223 The existing ioctl FS_IOC_GETFLAGS (which isn't specific to fs-verity)
224 can also be used to check whether a file has fs-verity enabled or not.
225 To do so, check for FS_VERITY_FL (0x00100000) in the returned flags.
227 The verity flag is not settable via FS_IOC_SETFLAGS. You must use
228 FS_IOC_ENABLE_VERITY instead, since parameters must be provided.
233 Since Linux v5.5, the statx() system call sets STATX_ATTR_VERITY if
234 the file has fs-verity enabled. This can perform better than
235 FS_IOC_GETFLAGS and FS_IOC_MEASURE_VERITY because it doesn't require
236 opening the file, and opening verity files can be expensive.
238 Accessing verity files
239 ======================
241 Applications can transparently access a verity file just like a
242 non-verity one, with the following exceptions:
244 - Verity files are readonly. They cannot be opened for writing or
245 truncate()d, even if the file mode bits allow it. Attempts to do
246 one of these things will fail with EPERM. However, changes to
247 metadata such as owner, mode, timestamps, and xattrs are still
248 allowed, since these are not measured by fs-verity. Verity files
249 can also still be renamed, deleted, and linked to.
251 - Direct I/O is not supported on verity files. Attempts to use direct
252 I/O on such files will fall back to buffered I/O.
254 - DAX (Direct Access) is not supported on verity files, because this
255 would circumvent the data verification.
257 - Reads of data that doesn't match the verity Merkle tree will fail
258 with EIO (for read()) or SIGBUS (for mmap() reads).
260 - If the sysctl "fs.verity.require_signatures" is set to 1 and the
261 file is not signed by a key in the fs-verity keyring, then opening
262 the file will fail. See `Built-in signature verification`_.
264 Direct access to the Merkle tree is not supported. Therefore, if a
265 verity file is copied, or is backed up and restored, then it will lose
266 its "verity"-ness. fs-verity is primarily meant for files like
267 executables that are managed by a package manager.
269 File digest computation
270 =======================
272 This section describes how fs-verity hashes the file contents using a
273 Merkle tree to produce the digest which cryptographically identifies
274 the file contents. This algorithm is the same for all filesystems
275 that support fs-verity.
277 Userspace only needs to be aware of this algorithm if it needs to
278 compute fs-verity file digests itself, e.g. in order to sign files.
280 .. _fsverity_merkle_tree:
285 The file contents is divided into blocks, where the block size is
286 configurable but is usually 4096 bytes. The end of the last block is
287 zero-padded if needed. Each block is then hashed, producing the first
288 level of hashes. Then, the hashes in this first level are grouped
289 into 'blocksize'-byte blocks (zero-padding the ends as needed) and
290 these blocks are hashed, producing the second level of hashes. This
291 proceeds up the tree until only a single block remains. The hash of
292 this block is the "Merkle tree root hash".
294 If the file fits in one block and is nonempty, then the "Merkle tree
295 root hash" is simply the hash of the single data block. If the file
296 is empty, then the "Merkle tree root hash" is all zeroes.
298 The "blocks" here are not necessarily the same as "filesystem blocks".
300 If a salt was specified, then it's zero-padded to the closest multiple
301 of the input size of the hash algorithm's compression function, e.g.
302 64 bytes for SHA-256 or 128 bytes for SHA-512. The padded salt is
303 prepended to every data or Merkle tree block that is hashed.
305 The purpose of the block padding is to cause every hash to be taken
306 over the same amount of data, which simplifies the implementation and
307 keeps open more possibilities for hardware acceleration. The purpose
308 of the salt padding is to make the salting "free" when the salted hash
309 state is precomputed, then imported for each hash.
311 Example: in the recommended configuration of SHA-256 and 4K blocks,
312 128 hash values fit in each block. Thus, each level of the Merkle
313 tree is approximately 128 times smaller than the previous, and for
314 large files the Merkle tree's size converges to approximately 1/127 of
315 the original file size. However, for small files, the padding is
316 significant, making the space overhead proportionally more.
318 .. _fsverity_descriptor:
323 By itself, the Merkle tree root hash is ambiguous. For example, it
324 can't a distinguish a large file from a small second file whose data
325 is exactly the top-level hash block of the first file. Ambiguities
326 also arise from the convention of padding to the next block boundary.
328 To solve this problem, the fs-verity file digest is actually computed
329 as a hash of the following structure, which contains the Merkle tree
330 root hash as well as other fields such as the file size::
332 struct fsverity_descriptor {
333 __u8 version; /* must be 1 */
334 __u8 hash_algorithm; /* Merkle tree hash algorithm */
335 __u8 log_blocksize; /* log2 of size of data and tree blocks */
336 __u8 salt_size; /* size of salt in bytes; 0 if none */
337 __le32 __reserved_0x04; /* must be 0 */
338 __le64 data_size; /* size of file the Merkle tree is built over */
339 __u8 root_hash[64]; /* Merkle tree root hash */
340 __u8 salt[32]; /* salt prepended to each hashed block */
341 __u8 __reserved[144]; /* must be 0's */
344 Built-in signature verification
345 ===============================
347 With CONFIG_FS_VERITY_BUILTIN_SIGNATURES=y, fs-verity supports putting
348 a portion of an authentication policy (see `Use cases`_) in the
349 kernel. Specifically, it adds support for:
351 1. At fs-verity module initialization time, a keyring ".fs-verity" is
352 created. The root user can add trusted X.509 certificates to this
353 keyring using the add_key() system call, then (when done)
354 optionally use keyctl_restrict_keyring() to prevent additional
355 certificates from being added.
357 2. `FS_IOC_ENABLE_VERITY`_ accepts a pointer to a PKCS#7 formatted
358 detached signature in DER format of the file's fs-verity digest.
359 On success, this signature is persisted alongside the Merkle tree.
360 Then, any time the file is opened, the kernel will verify the
361 file's actual digest against this signature, using the certificates
362 in the ".fs-verity" keyring.
364 3. A new sysctl "fs.verity.require_signatures" is made available.
365 When set to 1, the kernel requires that all verity files have a
366 correctly signed digest as described in (2).
368 fs-verity file digests must be signed in the following format, which
369 is similar to the structure used by `FS_IOC_MEASURE_VERITY`_::
371 struct fsverity_formatted_digest {
372 char magic[8]; /* must be "FSVerity" */
373 __le16 digest_algorithm;
378 fs-verity's built-in signature verification support is meant as a
379 relatively simple mechanism that can be used to provide some level of
380 authenticity protection for verity files, as an alternative to doing
381 the signature verification in userspace or using IMA-appraisal.
382 However, with this mechanism, userspace programs still need to check
383 that the verity bit is set, and there is no protection against verity
384 files being swapped around.
389 fs-verity is currently supported by the ext4 and f2fs filesystems.
390 The CONFIG_FS_VERITY kconfig option must be enabled to use fs-verity
391 on either filesystem.
393 ``include/linux/fsverity.h`` declares the interface between the
394 ``fs/verity/`` support layer and filesystems. Briefly, filesystems
395 must provide an ``fsverity_operations`` structure that provides
396 methods to read and write the verity metadata to a filesystem-specific
397 location, including the Merkle tree blocks and
398 ``fsverity_descriptor``. Filesystems must also call functions in
399 ``fs/verity/`` at certain times, such as when a file is opened or when
400 pages have been read into the pagecache. (See `Verifying data`_.)
405 ext4 supports fs-verity since Linux v5.4 and e2fsprogs v1.45.2.
407 To create verity files on an ext4 filesystem, the filesystem must have
408 been formatted with ``-O verity`` or had ``tune2fs -O verity`` run on
409 it. "verity" is an RO_COMPAT filesystem feature, so once set, old
410 kernels will only be able to mount the filesystem readonly, and old
411 versions of e2fsck will be unable to check the filesystem. Moreover,
412 currently ext4 only supports mounting a filesystem with the "verity"
413 feature when its block size is equal to PAGE_SIZE (often 4096 bytes).
415 ext4 sets the EXT4_VERITY_FL on-disk inode flag on verity files. It
416 can only be set by `FS_IOC_ENABLE_VERITY`_, and it cannot be cleared.
418 ext4 also supports encryption, which can be used simultaneously with
419 fs-verity. In this case, the plaintext data is verified rather than
420 the ciphertext. This is necessary in order to make the fs-verity file
421 digest meaningful, since every file is encrypted differently.
423 ext4 stores the verity metadata (Merkle tree and fsverity_descriptor)
424 past the end of the file, starting at the first 64K boundary beyond
425 i_size. This approach works because (a) verity files are readonly,
426 and (b) pages fully beyond i_size aren't visible to userspace but can
427 be read/written internally by ext4 with only some relatively small
428 changes to ext4. This approach avoids having to depend on the
429 EA_INODE feature and on rearchitecturing ext4's xattr support to
430 support paging multi-gigabyte xattrs into memory, and to support
431 encrypting xattrs. Note that the verity metadata *must* be encrypted
432 when the file is, since it contains hashes of the plaintext data.
434 Currently, ext4 verity only supports the case where the Merkle tree
435 block size, filesystem block size, and page size are all the same. It
436 also only supports extent-based files.
441 f2fs supports fs-verity since Linux v5.4 and f2fs-tools v1.11.0.
443 To create verity files on an f2fs filesystem, the filesystem must have
444 been formatted with ``-O verity``.
446 f2fs sets the FADVISE_VERITY_BIT on-disk inode flag on verity files.
447 It can only be set by `FS_IOC_ENABLE_VERITY`_, and it cannot be
450 Like ext4, f2fs stores the verity metadata (Merkle tree and
451 fsverity_descriptor) past the end of the file, starting at the first
452 64K boundary beyond i_size. See explanation for ext4 above.
453 Moreover, f2fs supports at most 4096 bytes of xattr entries per inode
454 which wouldn't be enough for even a single Merkle tree block.
456 Currently, f2fs verity only supports a Merkle tree block size of 4096.
457 Also, f2fs doesn't support enabling verity on files that currently
458 have atomic or volatile writes pending.
460 Implementation details
461 ======================
466 fs-verity ensures that all reads of a verity file's data are verified,
467 regardless of which syscall is used to do the read (e.g. mmap(),
468 read(), pread()) and regardless of whether it's the first read or a
469 later read (unless the later read can return cached data that was
470 already verified). Below, we describe how filesystems implement this.
475 For filesystems using Linux's pagecache, the ``->readpage()`` and
476 ``->readpages()`` methods must be modified to verify pages before they
477 are marked Uptodate. Merely hooking ``->read_iter()`` would be
478 insufficient, since ``->read_iter()`` is not used for memory maps.
480 Therefore, fs/verity/ provides a function fsverity_verify_page() which
481 verifies a page that has been read into the pagecache of a verity
482 inode, but is still locked and not Uptodate, so it's not yet readable
483 by userspace. As needed to do the verification,
484 fsverity_verify_page() will call back into the filesystem to read
485 Merkle tree pages via fsverity_operations::read_merkle_tree_page().
487 fsverity_verify_page() returns false if verification failed; in this
488 case, the filesystem must not set the page Uptodate. Following this,
489 as per the usual Linux pagecache behavior, attempts by userspace to
490 read() from the part of the file containing the page will fail with
491 EIO, and accesses to the page within a memory map will raise SIGBUS.
493 fsverity_verify_page() currently only supports the case where the
494 Merkle tree block size is equal to PAGE_SIZE (often 4096 bytes).
496 In principle, fsverity_verify_page() verifies the entire path in the
497 Merkle tree from the data page to the root hash. However, for
498 efficiency the filesystem may cache the hash pages. Therefore,
499 fsverity_verify_page() only ascends the tree reading hash pages until
500 an already-verified hash page is seen, as indicated by the PageChecked
501 bit being set. It then verifies the path to that page.
503 This optimization, which is also used by dm-verity, results in
504 excellent sequential read performance. This is because usually (e.g.
505 127 in 128 times for 4K blocks and SHA-256) the hash page from the
506 bottom level of the tree will already be cached and checked from
507 reading a previous data page. However, random reads perform worse.
509 Block device based filesystems
510 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
512 Block device based filesystems (e.g. ext4 and f2fs) in Linux also use
513 the pagecache, so the above subsection applies too. However, they
514 also usually read many pages from a file at once, grouped into a
515 structure called a "bio". To make it easier for these types of
516 filesystems to support fs-verity, fs/verity/ also provides a function
517 fsverity_verify_bio() which verifies all pages in a bio.
519 ext4 and f2fs also support encryption. If a verity file is also
520 encrypted, the pages must be decrypted before being verified. To
521 support this, these filesystems allocate a "post-read context" for
522 each bio and store it in ``->bi_private``::
524 struct bio_post_read_ctx {
526 struct work_struct work;
527 unsigned int cur_step;
528 unsigned int enabled_steps;
531 ``enabled_steps`` is a bitmask that specifies whether decryption,
532 verity, or both is enabled. After the bio completes, for each needed
533 postprocessing step the filesystem enqueues the bio_post_read_ctx on a
534 workqueue, and then the workqueue work does the decryption or
535 verification. Finally, pages where no decryption or verity error
536 occurred are marked Uptodate, and the pages are unlocked.
538 Files on ext4 and f2fs may contain holes. Normally, ``->readpages()``
539 simply zeroes holes and sets the corresponding pages Uptodate; no bios
540 are issued. To prevent this case from bypassing fs-verity, these
541 filesystems use fsverity_verify_page() to verify hole pages.
543 ext4 and f2fs disable direct I/O on verity files, since otherwise
544 direct I/O would bypass fs-verity. (They also do the same for
550 This document focuses on the kernel, but a userspace utility for
551 fs-verity can be found at:
553 https://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/fsverity-utils.git
555 See the README.md file in the fsverity-utils source tree for details,
556 including examples of setting up fs-verity protected files.
561 To test fs-verity, use xfstests. For example, using `kvm-xfstests
562 <https://github.com/tytso/xfstests-bld/blob/master/Documentation/kvm-quickstart.md>`_::
564 kvm-xfstests -c ext4,f2fs -g verity
569 This section answers frequently asked questions about fs-verity that
570 weren't already directly answered in other parts of this document.
572 :Q: Why isn't fs-verity part of IMA?
573 :A: fs-verity and IMA (Integrity Measurement Architecture) have
574 different focuses. fs-verity is a filesystem-level mechanism for
575 hashing individual files using a Merkle tree. In contrast, IMA
576 specifies a system-wide policy that specifies which files are
577 hashed and what to do with those hashes, such as log them,
578 authenticate them, or add them to a measurement list.
580 IMA is planned to support the fs-verity hashing mechanism as an
581 alternative to doing full file hashes, for people who want the
582 performance and security benefits of the Merkle tree based hash.
583 But it doesn't make sense to force all uses of fs-verity to be
584 through IMA. As a standalone filesystem feature, fs-verity
585 already meets many users' needs, and it's testable like other
586 filesystem features e.g. with xfstests.
588 :Q: Isn't fs-verity useless because the attacker can just modify the
589 hashes in the Merkle tree, which is stored on-disk?
590 :A: To verify the authenticity of an fs-verity file you must verify
591 the authenticity of the "fs-verity file digest", which
592 incorporates the root hash of the Merkle tree. See `Use cases`_.
594 :Q: Isn't fs-verity useless because the attacker can just replace a
595 verity file with a non-verity one?
596 :A: See `Use cases`_. In the initial use case, it's really trusted
597 userspace code that authenticates the files; fs-verity is just a
598 tool to do this job efficiently and securely. The trusted
599 userspace code will consider non-verity files to be inauthentic.
601 :Q: Why does the Merkle tree need to be stored on-disk? Couldn't you
602 store just the root hash?
603 :A: If the Merkle tree wasn't stored on-disk, then you'd have to
604 compute the entire tree when the file is first accessed, even if
605 just one byte is being read. This is a fundamental consequence of
606 how Merkle tree hashing works. To verify a leaf node, you need to
607 verify the whole path to the root hash, including the root node
608 (the thing which the root hash is a hash of). But if the root
609 node isn't stored on-disk, you have to compute it by hashing its
610 children, and so on until you've actually hashed the entire file.
612 That defeats most of the point of doing a Merkle tree-based hash,
613 since if you have to hash the whole file ahead of time anyway,
614 then you could simply do sha256(file) instead. That would be much
615 simpler, and a bit faster too.
617 It's true that an in-memory Merkle tree could still provide the
618 advantage of verification on every read rather than just on the
619 first read. However, it would be inefficient because every time a
620 hash page gets evicted (you can't pin the entire Merkle tree into
621 memory, since it may be very large), in order to restore it you
622 again need to hash everything below it in the tree. This again
623 defeats most of the point of doing a Merkle tree-based hash, since
624 a single block read could trigger re-hashing gigabytes of data.
626 :Q: But couldn't you store just the leaf nodes and compute the rest?
627 :A: See previous answer; this really just moves up one level, since
628 one could alternatively interpret the data blocks as being the
629 leaf nodes of the Merkle tree. It's true that the tree can be
630 computed much faster if the leaf level is stored rather than just
631 the data, but that's only because each level is less than 1% the
632 size of the level below (assuming the recommended settings of
633 SHA-256 and 4K blocks). For the exact same reason, by storing
634 "just the leaf nodes" you'd already be storing over 99% of the
635 tree, so you might as well simply store the whole tree.
637 :Q: Can the Merkle tree be built ahead of time, e.g. distributed as
638 part of a package that is installed to many computers?
639 :A: This isn't currently supported. It was part of the original
640 design, but was removed to simplify the kernel UAPI and because it
641 wasn't a critical use case. Files are usually installed once and
642 used many times, and cryptographic hashing is somewhat fast on
643 most modern processors.
645 :Q: Why doesn't fs-verity support writes?
646 :A: Write support would be very difficult and would require a
647 completely different design, so it's well outside the scope of
648 fs-verity. Write support would require:
650 - A way to maintain consistency between the data and hashes,
651 including all levels of hashes, since corruption after a crash
652 (especially of potentially the entire file!) is unacceptable.
653 The main options for solving this are data journalling,
654 copy-on-write, and log-structured volume. But it's very hard to
655 retrofit existing filesystems with new consistency mechanisms.
656 Data journalling is available on ext4, but is very slow.
658 - Rebuilding the Merkle tree after every write, which would be
659 extremely inefficient. Alternatively, a different authenticated
660 dictionary structure such as an "authenticated skiplist" could
661 be used. However, this would be far more complex.
663 Compare it to dm-verity vs. dm-integrity. dm-verity is very
664 simple: the kernel just verifies read-only data against a
665 read-only Merkle tree. In contrast, dm-integrity supports writes
666 but is slow, is much more complex, and doesn't actually support
667 full-device authentication since it authenticates each sector
668 independently, i.e. there is no "root hash". It doesn't really
669 make sense for the same device-mapper target to support these two
670 very different cases; the same applies to fs-verity.
672 :Q: Since verity files are immutable, why isn't the immutable bit set?
673 :A: The existing "immutable" bit (FS_IMMUTABLE_FL) already has a
674 specific set of semantics which not only make the file contents
675 read-only, but also prevent the file from being deleted, renamed,
676 linked to, or having its owner or mode changed. These extra
677 properties are unwanted for fs-verity, so reusing the immutable
678 bit isn't appropriate.
680 :Q: Why does the API use ioctls instead of setxattr() and getxattr()?
681 :A: Abusing the xattr interface for basically arbitrary syscalls is
682 heavily frowned upon by most of the Linux filesystem developers.
683 An xattr should really just be an xattr on-disk, not an API to
684 e.g. magically trigger construction of a Merkle tree.
686 :Q: Does fs-verity support remote filesystems?
687 :A: Only ext4 and f2fs support is implemented currently, but in
688 principle any filesystem that can store per-file verity metadata
689 can support fs-verity, regardless of whether it's local or remote.
690 Some filesystems may have fewer options of where to store the
691 verity metadata; one possibility is to store it past the end of
692 the file and "hide" it from userspace by manipulating i_size. The
693 data verification functions provided by ``fs/verity/`` also assume
694 that the filesystem uses the Linux pagecache, but both local and
695 remote filesystems normally do so.
697 :Q: Why is anything filesystem-specific at all? Shouldn't fs-verity
698 be implemented entirely at the VFS level?
699 :A: There are many reasons why this is not possible or would be very
700 difficult, including the following:
702 - To prevent bypassing verification, pages must not be marked
703 Uptodate until they've been verified. Currently, each
704 filesystem is responsible for marking pages Uptodate via
705 ``->readpages()``. Therefore, currently it's not possible for
706 the VFS to do the verification on its own. Changing this would
707 require significant changes to the VFS and all filesystems.
709 - It would require defining a filesystem-independent way to store
710 the verity metadata. Extended attributes don't work for this
711 because (a) the Merkle tree may be gigabytes, but many
712 filesystems assume that all xattrs fit into a single 4K
713 filesystem block, and (b) ext4 and f2fs encryption doesn't
714 encrypt xattrs, yet the Merkle tree *must* be encrypted when the
715 file contents are, because it stores hashes of the plaintext
718 So the verity metadata would have to be stored in an actual
719 file. Using a separate file would be very ugly, since the
720 metadata is fundamentally part of the file to be protected, and
721 it could cause problems where users could delete the real file
722 but not the metadata file or vice versa. On the other hand,
723 having it be in the same file would break applications unless
724 filesystems' notion of i_size were divorced from the VFS's,
725 which would be complex and require changes to all filesystems.
727 - It's desirable that FS_IOC_ENABLE_VERITY uses the filesystem's
728 transaction mechanism so that either the file ends up with
729 verity enabled, or no changes were made. Allowing intermediate
730 states to occur after a crash may cause problems.
projects / linux-2.6-microblaze.git / blob