]> git.ipfire.org Git - thirdparty/kernel/linux.git/commit
fsverity: generate and store zero-block hash
authorAndrey Albershteyn <aalbersh@kernel.org>
Wed, 20 May 2026 12:37:02 +0000 (14:37 +0200)
committerChristian Brauner <brauner@kernel.org>
Fri, 5 Jun 2026 11:36:22 +0000 (13:36 +0200)
commit07d09774e2bfa21dedcee3ef45892bb20827b12c
tree4de26142f568ea71cfe2067930ec607a17faaaaa
parent198c3b90e600d442f8333d254a23c5da989c5cb4
fsverity: generate and store zero-block hash

Compute the hash of one filesystem block's worth of zeros. A filesystem
implementation can decide to elide merkle tree blocks containing only
this hash and synthesize the contents at read time.

Let's pretend that there's a file containing 131 data block and whose
merkle tree looks roughly like this:

root
 +--leaf0
 |   +--data0
 |   +--data1
 |   +--...
 |   `--data128
 `--leaf1
     +--data129
     +--data130
     `--data131

If data[0-128] are sparse holes, then leaf0 will contain a repeating
sequence of @zero_digest.  Therefore, leaf0 need not be written to disk
because its contents can be synthesized.

A subsequent xfs patch will use this to reduce the size of the merkle
tree when dealing with sparse gold master disk images and the like.

Note that this works only on the first-level (data holes). fsverity
doesn't store/generate zero_digest for any higher levels.

Add a helper to pre-fill folio with hashes of empty blocks. This will be
used by iomap to synthesize blocks full of zero hashes on the fly.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Acked-by: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: Andrey Albershteyn <aalbersh@kernel.org>
Link: https://patch.msgid.link/20260520123722.405752-5-aalbersh@kernel.org
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
fs/verity/fsverity_private.h
fs/verity/measure.c
fs/verity/open.c
fs/verity/pagecache.c
include/linux/fsverity.h