On 07/30/26 at 11:48am, Eric Biggers wrote:
> ext4 and f2fs don't prevent filesystem-level encrypted files from being
> set up directly as swap files. In this case, encryption is bypassed.
>
> No one should be doing this, vs. the methods of encrypted swap that
> actually do work (such as swapping to a dm-crypt device, or swapping to
> a loopback device on top of a filesystem-level encrypted file).
>
> Nevertheless, to prevent user error, make swapon() explicitly reject
> this case. Document this behavior in fscrypt.rst as well.
>
> Fixes: 9bd8212f981e ("ext4 crypto: add encryption policy and password salt
> support")
> Fixes: f424f664f0e8 ("f2fs crypto: add encryption policy and password salt
> support")
> Cc: [email protected]
> Signed-off-by: Eric Biggers <[email protected]>
> ---
> Documentation/filesystems/fscrypt.rst | 4 ++++
> mm/swapfile.c | 4 ++++
> 2 files changed, 8 insertions(+)
>
> diff --git a/Documentation/filesystems/fscrypt.rst
> b/Documentation/filesystems/fscrypt.rst
> index c0dd35f1af12..cba1989777da 100644
> --- a/Documentation/filesystems/fscrypt.rst
> +++ b/Documentation/filesystems/fscrypt.rst
> @@ -1238,6 +1238,10 @@ astute users may notice some differences in behavior:
>
> - DAX (Direct Access) is not supported on encrypted files.
>
> +- Encrypted files cannot be used directly as swap files. To swap to
> + an encrypted file, set up a loopback device on top of it.
> + Alternatively, encrypted swap can use a dm-crypt device instead.
> +
> - The maximum length of an encrypted symlink is 2 bytes shorter than
> the maximum length of an unencrypted symlink. For example, on an
> EXT4 filesystem with a 4K block size, unencrypted symlinks can be up
> diff --git a/mm/swapfile.c b/mm/swapfile.c
> index 78b49b0658ad..e4991da81b5f 100644
> --- a/mm/swapfile.c
> +++ b/mm/swapfile.c
> @@ -3650,6 +3650,10 @@ SYSCALL_DEFINE2(swapon, const char __user *,
> specialfile, int, swap_flags)
> error = -EBUSY;
> goto bad_swap_unlock_inode;
> }
> + if (IS_ENCRYPTED(inode)) {
> + error = -EINVAL;
> + goto bad_swap_unlock_inode;
> + }
Sounds reasonable to me.
One minor suggestion: it might be helpful to emit a pr_warn_once()
explaining *why* the swapon was rejected and pointing users to the
dm-crypt / loopback alternatives, so the -EINVAL isn't opaque.
Optional though -- the check itself is what matters.
And this patch updates fscrypt.rst. Would it be good to also
update the swapon(8) man page to document the new -EINVAL
condition?
Other than these,
Reviewed-by: Baoquan He <[email protected]>
>
> /*
> * The swap subsystem needs a major overhaul to support this.
>
> base-commit: f5098b6bae761e346ebcd9da7f95622c04733cff
> --
> 2.55.0
>
