[QUESTION] Does fs-verity support large folios / large block size in ext4?

[QUESTION] Does fs-verity support large folios / large block size in ext4?

[Baokun Li]

I have recently been working on enabling large block size support in ext4.
I had assumed that fs-verity does not support large folios / large block
size. However, when testing with 64k LBS using:

    `kvm-xfstests -c ext4/64k -g verity -x encrypt`

the tests passed successfully.

Is this just a coincidence, or has fs-verity code already been updated to
support large block size / large folios?


Thanks,
Baokun


-------------------- Summary report
KERNEL:    kernel 6.18.0-rc4-next-20251103-dirty #880 SMP PREEMPT_DYNAMIC
Thu Nov  6 19:27:12 CST 2025 x86_64
CMDLINE:   -c ext4/64k -g verity -x encrypt
CPUS:      2
MEM:       7944.47

ext4/64k: 9 tests, 59 seconds
  generic/572  Pass     3s
  generic/573  Pass     2s
  generic/574  Pass     6s
  generic/575  Pass     4s
  generic/577  Pass     6s
  generic/579  Pass     25s
  generic/624  Pass     1s
  generic/625  Pass     2s
  generic/692  Pass     1s
Totals: 9 tests, 0 skipped, 0 failures, 0 errors, 50s

FSTESTVER: blktests 698f1a0 (Mon, 27 May 2024 11:30:36 +0900)
FSTESTVER: fio  fio-3.28 (Wed, 8 Sep 2021 08:59:48 -0600)
FSTESTVER: fsverity v1.6 (Wed, 20 Mar 2024 21:21:46 -0700)
FSTESTVER: libaio   libaio-0.3.108-81-g1b18bfa (Mon, 28 Mar 2022 11:30:33
-0400)
FSTESTVER: quota  v4.05-43-gd2256ac (Fri, 17 Sep 2021 14:04:16 +0200)
FSTESTVER: xfsprogs v5.13.0 (Fri, 20 Aug 2021 12:03:57 -0400)
FSTESTVER: xfstests-bld 1bdd10a-dirty (Fri, 3 May 2024 16:14:41 -0400)
FSTESTVER: xfstests v2024.05.12 (Sun, 12 May 2024 20:28:48 +0800)
FSTESTCFG: ext4/64k
FSTESTSET: -g verity -x encrypt
FSTESTOPT: aex

Re: [QUESTION] Does fs-verity support large folios / large block size in ext4?

[Eric Biggers]

On Thu, Nov 06, 2025 at 09:02:53PM +0800, Baokun Li wrote:
> I have recently been working on enabling large block size support in ext4.
> I had assumed that fs-verity does not support large folios / large block
> size. However, when testing with 64k LBS using:
> 
>     `kvm-xfstests -c ext4/64k -g verity -x encrypt`
> 
> the tests passed successfully.
> 
> Is this just a coincidence, or has fs-verity code already been updated to
> support large block size / large folios?

It might work already.  I already added support for verifying data from
large folios several years ago:

    commit 5d0f0e57ed900917836385527ce5b122fa1425a3
    Author: Eric Biggers <ebiggers@google.com>
    Date:   Fri Jan 27 14:15:29 2023 -0800

        fsverity: support verifying data from large folios

Note that the Merkle tree block size is still limited to <= PAGE_SIZE.
But that's fine for now; generally people want 4K Merkle tree blocks
anyway.  So the configuration in question, which should be getting
tested when you run that command and should theoretically work, should
be: filesystem_block_size=64K, PAGE_SIZE=4K, merkle_tree_block_size=4K.

Is there a git tree with your work that I can take a look at and confirm
that this is happening?

- Eric

Re: [QUESTION] Does fs-verity support large folios / large block size in ext4?

[Baokun Li]

On 2025-11-07 04:44, Eric Biggers wrote:
> On Thu, Nov 06, 2025 at 09:02:53PM +0800, Baokun Li wrote:
>> I have recently been working on enabling large block size support in ext4.
>> I had assumed that fs-verity does not support large folios / large block
>> size. However, when testing with 64k LBS using:
>>
>>     `kvm-xfstests -c ext4/64k -g verity -x encrypt`
>>
>> the tests passed successfully.
>>
>> Is this just a coincidence, or has fs-verity code already been updated to
>> support large block size / large folios?
> It might work already.  I already added support for verifying data from
> large folios several years ago:
>
>     commit 5d0f0e57ed900917836385527ce5b122fa1425a3
>     Author: Eric Biggers <ebiggers@google.com>
>     Date:   Fri Jan 27 14:15:29 2023 -0800
>
>         fsverity: support verifying data from large folios
>
> Note that the Merkle tree block size is still limited to <= PAGE_SIZE.
> But that's fine for now; generally people want 4K Merkle tree blocks
> anyway.  So the configuration in question, which should be getting
> tested when you run that command and should theoretically work, should
> be: filesystem_block_size=64K, PAGE_SIZE=4K, merkle_tree_block_size=4K.
>
> Is there a git tree with your work that I can take a look at and confirm
> that this is happening?
>
> - Eric

I sent out v1 without enabling LBS support in verity. Since you have
already done some adaptation work, I plan to enable it in v2. v2 will be
sent out today, and I will cc you.

Thank you very much for helping to confirm!


-- 
With Best Regards,
Baokun Li