kernel/linux.git/include/linux/fscrypt.h, branch v6.12.80

fscrypt: Drop d_revalidate once the key is added

2024-02-27T21:55:34+00:00

When a key is added, existing directory dentries in the DCACHE_NOKEY_NAME form are moved by the VFS to the plaintext version. But, since they have the DCACHE_OP_REVALIDATE flag set, revalidation will be done at each lookup only to return immediately, since plaintext dentries can't go stale until eviction. This patch optimizes this case, by dropping the flag once the nokey_name dentry becomes plain-text. Note that non-directory dentries are not moved this way, so they won't be affected. Of course, this can only be done if fscrypt is the only thing requiring revalidation for a dentry. For this reason, we only disable d_revalidate if the .d_revalidate hook is fscrypt_d_revalidate itself. It is safe to do it here because when moving the dentry to the plain-text version, we are holding the d_lock. We might race with a concurrent RCU lookup but this is harmless because, at worst, we will get an extra d_revalidate on the keyed dentry, which will still find the dentry to be valid. Finally, now that we do more than just clear the DCACHE_NOKEY_NAME in fscrypt_handle_d_move, skip it entirely for plaintext dentries, to avoid extra costs. Reviewed-by: Eric Biggers Link: https://lore.kernel.org/r/20240221171412.10710-5-krisman@suse.de Signed-off-by: Gabriel Krisman Bertazi

fscrypt: Drop d_revalidate for valid dentries during lookup

2024-02-27T21:55:34+00:00

Unencrypted and encrypted-dentries where the key is available don't need to be revalidated by fscrypt, since they don't go stale from under VFS and the key cannot be removed for the encrypted case without evicting the dentry. Disable their d_revalidate hook on the first lookup, to avoid repeated revalidation later. This is done in preparation to always configuring d_op through sb->s_d_op. The only part detail is that, since the filesystem might have other features that require revalidation, we only apply this optimization if the d_revalidate handler is fscrypt_d_revalidate itself. Finally, we need to clean the dentry->flags even for unencrypted dentries, so the ->d_lock might be acquired even for them. In order to avoid doing it for filesystems that don't care about fscrypt at all, we peek ->d_flags without the lock at first, and only acquire it if we actually need to write the flag. Reviewed-by: Eric Biggers Link: https://lore.kernel.org/r/20240221171412.10710-4-krisman@suse.de Signed-off-by: Gabriel Krisman Bertazi

fscrypt: Factor out a helper to configure the lookup dentry

2024-02-27T21:55:34+00:00

Both fscrypt_prepare_lookup_partial and fscrypt_prepare_lookup will set DCACHE_NOKEY_NAME for dentries when the key is not available. Extract out a helper to set this flag in a single place, in preparation to also add the optimization that will disable ->d_revalidate if possible. Reviewed-by: Eric Biggers Link: https://lore.kernel.org/r/20240221171412.10710-3-krisman@suse.de Signed-off-by: Gabriel Krisman Bertazi

fscrypt: rename fscrypt_info => fscrypt_inode_info

2023-10-09T03:44:26+00:00

We are going to track per-extent information, so it'll be necessary to distinguish between inode infos and extent infos. Rename fscrypt_info to fscrypt_inode_info, adjusting any lines that now exceed 80 characters. Signed-off-by: Josef Bacik [ebiggers: rebased onto fscrypt tree, renamed fscrypt_get_info(), adjusted two comments, and fixed some lines over 80 characters] Link: https://lore.kernel.org/r/20231005025757.33521-1-ebiggers@kernel.org Reviewed-by: Neal Gompa Signed-off-by: Eric Biggers

fscrypt: support crypto data unit size less than filesystem block size

2023-09-26T05:34:33+00:00

Until now, fscrypt has always used the filesystem block size as the granularity of file contents encryption. Two scenarios have come up where a sub-block granularity of contents encryption would be useful: 1. Inline crypto hardware that only supports a crypto data unit size that is less than the filesystem block size. 2. Support for direct I/O at a granularity less than the filesystem block size, for example at the block device's logical block size in order to match the traditional direct I/O alignment requirement. (1) first came up with older eMMC inline crypto hardware that only supports a crypto data unit size of 512 bytes. That specific case ultimately went away because all systems with that hardware continued using out of tree code and never actually upgraded to the upstream inline crypto framework. But, now it's coming back in a new way: some current UFS controllers only support a data unit size of 4096 bytes, and there is a proposal to increase the filesystem block size to 16K. (2) was discussed as a "nice to have" feature, though not essential, when support for direct I/O on encrypted files was being upstreamed. Still, the fact that this feature has come up several times does suggest it would be wise to have available. Therefore, this patch implements it by using one of the reserved bytes in fscrypt_policy_v2 to allow users to select a sub-block data unit size. Supported data unit sizes are powers of 2 between 512 and the filesystem block size, inclusively. Support is implemented for both the FS-layer and inline crypto cases. This patch focuses on the basic support for sub-block data units. Some things are out of scope for this patch but may be addressed later: - Supporting sub-block data units in combination with FSCRYPT_POLICY_FLAG_IV_INO_LBLK_64, in most cases. Unfortunately this combination usually causes data unit indices to exceed 32 bits, and thus fscrypt_supported_policy() correctly disallows it. The users who potentially need this combination are using f2fs. To support it, f2fs would need to provide an option to slightly reduce its max file size. - Supporting sub-block data units in combination with FSCRYPT_POLICY_FLAG_IV_INO_LBLK_32. This has the same problem described above, but also it will need special code to make DUN wraparound still happen on a FS block boundary. - Supporting use case (2) mentioned above. The encrypted direct I/O code will need to stop requiring and assuming FS block alignment. This won't be hard, but it belongs in a separate patch. - Supporting this feature on filesystems other than ext4 and f2fs. (Filesystems declare support for it via their fscrypt_operations.) On UBIFS, sub-block data units don't make sense because UBIFS encrypts variable-length blocks as a result of compression. CephFS could support it, but a bit more work would be needed to make the fscrypt_*_block_inplace functions play nicely with sub-block data units. I don't think there's a use case for this on CephFS anyway. Link: https://lore.kernel.org/r/20230925055451.59499-6-ebiggers@kernel.org Signed-off-by: Eric Biggers

fscrypt: replace get_ino_and_lblk_bits with just has_32bit_inodes

2023-09-26T05:34:33+00:00

Now that fs/crypto/ computes the filesystem's lblk_bits from its maximum file size, it is no longer necessary for filesystems to provide lblk_bits via fscrypt_operations::get_ino_and_lblk_bits. It is still necessary for fs/crypto/ to retrieve ino_bits from the filesystem. However, this is used only to decide whether inode numbers fit in 32 bits. Also, ino_bits is static for all relevant filesystems, i.e. it doesn't depend on the filesystem instance. Therefore, in the interest of keeping things as simple as possible, replace 'get_ino_and_lblk_bits' with a flag 'has_32bit_inodes'. This can always be changed back to a function if a filesystem needs it to be dynamic, but for now a static flag is all that's needed. Link: https://lore.kernel.org/r/20230925055451.59499-5-ebiggers@kernel.org Signed-off-by: Eric Biggers

fscrypt: make the bounce page pool opt-in instead of opt-out

2023-09-25T06:03:09+00:00

Replace FS_CFLG_OWN_PAGES with a bit flag 'needs_bounce_pages' which has the opposite meaning. I.e., filesystems now opt into the bounce page pool instead of opt out. Make fscrypt_alloc_bounce_page() check that the bounce page pool has been initialized. I believe the opt-in makes more sense, since nothing else in fscrypt_operations is opt-out, and these days filesystems can choose to use blk-crypto which doesn't need the fscrypt bounce page pool. Also, I happen to be planning to add two more flags, and I wanted to fix the "FS_CFLG_" name anyway as it wasn't prefixed with "FSCRYPT_". Link: https://lore.kernel.org/r/20230925055451.59499-3-ebiggers@kernel.org Signed-off-by: Eric Biggers

fscrypt: make it clearer that key_prefix is deprecated

2023-09-25T06:03:09+00:00

fscrypt_operations::key_prefix should not be set by any filesystems that aren't setting it already. This is already documented, but apparently it's not sufficiently clear, as both ceph and btrfs have tried to set it. Rename the field to legacy_key_prefix and improve the documentation to hopefully make it clearer. Link: https://lore.kernel.org/r/20230925055451.59499-2-ebiggers@kernel.org Signed-off-by: Eric Biggers

Merge tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

2023-04-26T15:57:41+00:00

Pull ext4 updates from Ted Ts'o: "There are a number of major cleanups in ext4 this cycle: - The data=journal writepath has been significantly cleaned up and simplified, and reduces a large number of data=journal special cases by Jan Kara. - Ojaswin Muhoo has replaced linked list used to track extents that have been used for inode preallocation with a red-black tree in the multi-block allocator. This improves performance for workloads which do a large number of random allocating writes. - Thanks to Kemeng Shi for a lot of cleanup and bug fixes in the multi-block allocator. - Matthew wilcox has converted the code paths for reading and writing ext4 pages to use folios. - Jason Yan has continued to factor out ext4_fill_super() into smaller functions for improve ease of maintenance and comprehension. - Josh Triplett has created an uapi header for ext4 userspace API's" * tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (105 commits) ext4: Add a uapi header for ext4 userspace APIs ext4: remove useless conditional branch code ext4: remove unneeded check of nr_to_submit ext4: move dax and encrypt checking into ext4_check_feature_compatibility() ext4: factor out ext4_block_group_meta_init() ext4: move s_reserved_gdt_blocks and addressable checking into ext4_check_geometry() ext4: rename two functions with 'check' ext4: factor out ext4_flex_groups_free() ext4: use ext4_group_desc_free() in ext4_put_super() to save some duplicated code ext4: factor out ext4_percpu_param_init() and ext4_percpu_param_destroy() ext4: factor out ext4_hash_info_init() Revert "ext4: Fix warnings when freezing filesystem with journaled data" ext4: Update comment in mpage_prepare_extent_to_map() ext4: Simplify handling of journalled data in ext4_bmap() ext4: Drop special handling of journalled data from ext4_quota_on() ext4: Drop special handling of journalled data from ext4_evict_inode() ext4: Fix special handling of journalled data from extent zeroing ext4: Drop special handling of journalled data from extent shifting operations ext4: Drop special handling of journalled data from ext4_sync_file() ext4: Commit transaction before writing back pages in data=journal mode ...

fscrypt: Add some folio helper functions

2023-04-06T17:39:50+00:00

fscrypt_is_bounce_folio() is the equivalent of fscrypt_is_bounce_page() and fscrypt_pagecache_folio() is the equivalent of fscrypt_pagecache_page(). Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: Ritesh Harjani (IBM) Link: https://lore.kernel.org/r/20230324180129.1220691-3-willy@infradead.org Signed-off-by: Theodore Ts'o