[v2,0/5] xfstests: verify ciphertext of IV_INO_LBLK_64 encryption policies
mbox series

Message ID 20191202230155.99071-1-ebiggers@kernel.org
Headers show
Series
  • xfstests: verify ciphertext of IV_INO_LBLK_64 encryption policies
Related show

Message

Eric Biggers Dec. 2, 2019, 11:01 p.m. UTC
Hello,

This series adds an xfstest which tests that the encryption for
IV_INO_LBLK_64 encryption policies is being done correctly.

IV_INO_LBLK_64 is a new fscrypt policy flag which modifies the
encryption to be optimized for inline encryption hardware compliant with
the UFS v2.1 standard or the upcoming version of the eMMC standard.  For
more information, see the kernel patchset:
https://lore.kernel.org/linux-fscrypt/20191024215438.138489-1-ebiggers@kernel.org/T/#u

The kernel patches have been merged into mainline and will be in v5.5.

In addition to the latest kernel, to run on ext4 this test also needs a
version of e2fsprogs built from the master branch, in order to get
support for formatting the filesystem with '-O stable_inodes'.

As usual, the test will skip itself if the prerequisites aren't met.

No real changes since v1; just rebased onto the latest xfstests master
branch and updated the cover letter.

Eric Biggers (5):
  fscrypt-crypt-util: create key_and_iv_params structure
  fscrypt-crypt-util: add HKDF context constants
  common/encrypt: create named variables for UAPI constants
  common/encrypt: support verifying ciphertext of IV_INO_LBLK_64
    policies
  generic: verify ciphertext of IV_INO_LBLK_64 encryption policies

 common/encrypt           | 126 +++++++++++++++++++++++++-------
 src/fscrypt-crypt-util.c | 151 ++++++++++++++++++++++++++++-----------
 tests/generic/805        |  43 +++++++++++
 tests/generic/805.out    |   6 ++
 tests/generic/group      |   1 +
 5 files changed, 259 insertions(+), 68 deletions(-)
 create mode 100644 tests/generic/805
 create mode 100644 tests/generic/805.out

Comments

Eric Biggers Dec. 9, 2019, 6:18 p.m. UTC | #1
On Mon, Dec 02, 2019 at 03:01:50PM -0800, Eric Biggers wrote:
> Hello,
> 
> This series adds an xfstest which tests that the encryption for
> IV_INO_LBLK_64 encryption policies is being done correctly.
> 
> IV_INO_LBLK_64 is a new fscrypt policy flag which modifies the
> encryption to be optimized for inline encryption hardware compliant with
> the UFS v2.1 standard or the upcoming version of the eMMC standard.  For
> more information, see the kernel patchset:
> https://lore.kernel.org/linux-fscrypt/20191024215438.138489-1-ebiggers@kernel.org/T/#u
> 
> The kernel patches have been merged into mainline and will be in v5.5.
> 
> In addition to the latest kernel, to run on ext4 this test also needs a
> version of e2fsprogs built from the master branch, in order to get
> support for formatting the filesystem with '-O stable_inodes'.
> 
> As usual, the test will skip itself if the prerequisites aren't met.
> 
> No real changes since v1; just rebased onto the latest xfstests master
> branch and updated the cover letter.
> 
> Eric Biggers (5):
>   fscrypt-crypt-util: create key_and_iv_params structure
>   fscrypt-crypt-util: add HKDF context constants
>   common/encrypt: create named variables for UAPI constants
>   common/encrypt: support verifying ciphertext of IV_INO_LBLK_64
>     policies
>   generic: verify ciphertext of IV_INO_LBLK_64 encryption policies
> 
>  common/encrypt           | 126 +++++++++++++++++++++++++-------
>  src/fscrypt-crypt-util.c | 151 ++++++++++++++++++++++++++++-----------
>  tests/generic/805        |  43 +++++++++++
>  tests/generic/805.out    |   6 ++
>  tests/generic/group      |   1 +
>  5 files changed, 259 insertions(+), 68 deletions(-)
>  create mode 100644 tests/generic/805
>  create mode 100644 tests/generic/805.out
> 

Ping.  Does anyone want to take a look at this?  Satya?

- Eric
Eric Biggers Jan. 3, 2020, 4:46 p.m. UTC | #2
On Mon, Dec 09, 2019 at 10:18:27AM -0800, Eric Biggers wrote:
> On Mon, Dec 02, 2019 at 03:01:50PM -0800, Eric Biggers wrote:
> > Hello,
> > 
> > This series adds an xfstest which tests that the encryption for
> > IV_INO_LBLK_64 encryption policies is being done correctly.
> > 
> > IV_INO_LBLK_64 is a new fscrypt policy flag which modifies the
> > encryption to be optimized for inline encryption hardware compliant with
> > the UFS v2.1 standard or the upcoming version of the eMMC standard.  For
> > more information, see the kernel patchset:
> > https://lore.kernel.org/linux-fscrypt/20191024215438.138489-1-ebiggers@kernel.org/T/#u
> > 
> > The kernel patches have been merged into mainline and will be in v5.5.
> > 
> > In addition to the latest kernel, to run on ext4 this test also needs a
> > version of e2fsprogs built from the master branch, in order to get
> > support for formatting the filesystem with '-O stable_inodes'.
> > 
> > As usual, the test will skip itself if the prerequisites aren't met.
> > 
> > No real changes since v1; just rebased onto the latest xfstests master
> > branch and updated the cover letter.
> > 
> > Eric Biggers (5):
> >   fscrypt-crypt-util: create key_and_iv_params structure
> >   fscrypt-crypt-util: add HKDF context constants
> >   common/encrypt: create named variables for UAPI constants
> >   common/encrypt: support verifying ciphertext of IV_INO_LBLK_64
> >     policies
> >   generic: verify ciphertext of IV_INO_LBLK_64 encryption policies
> > 
> >  common/encrypt           | 126 +++++++++++++++++++++++++-------
> >  src/fscrypt-crypt-util.c | 151 ++++++++++++++++++++++++++++-----------
> >  tests/generic/805        |  43 +++++++++++
> >  tests/generic/805.out    |   6 ++
> >  tests/generic/group      |   1 +
> >  5 files changed, 259 insertions(+), 68 deletions(-)
> >  create mode 100644 tests/generic/805
> >  create mode 100644 tests/generic/805.out
> > 
> 
> Ping.  Does anyone want to take a look at this?  Satya?
> 

Eryu, can you review and consider applying this series?  It doesn't look like
anyone else is going to formally review it.

Thanks,

- Eric
Eryu Guan Jan. 6, 2020, 2:55 a.m. UTC | #3
On Fri, Jan 03, 2020 at 08:46:26AM -0800, Eric Biggers wrote:
> On Mon, Dec 09, 2019 at 10:18:27AM -0800, Eric Biggers wrote:
> > On Mon, Dec 02, 2019 at 03:01:50PM -0800, Eric Biggers wrote:
> > > Hello,
> > > 
> > > This series adds an xfstest which tests that the encryption for
> > > IV_INO_LBLK_64 encryption policies is being done correctly.
> > > 
> > > IV_INO_LBLK_64 is a new fscrypt policy flag which modifies the
> > > encryption to be optimized for inline encryption hardware compliant with
> > > the UFS v2.1 standard or the upcoming version of the eMMC standard.  For
> > > more information, see the kernel patchset:
> > > https://lore.kernel.org/linux-fscrypt/20191024215438.138489-1-ebiggers@kernel.org/T/#u
> > > 
> > > The kernel patches have been merged into mainline and will be in v5.5.
> > > 
> > > In addition to the latest kernel, to run on ext4 this test also needs a
> > > version of e2fsprogs built from the master branch, in order to get
> > > support for formatting the filesystem with '-O stable_inodes'.
> > > 
> > > As usual, the test will skip itself if the prerequisites aren't met.
> > > 
> > > No real changes since v1; just rebased onto the latest xfstests master
> > > branch and updated the cover letter.
> > > 
> > > Eric Biggers (5):
> > >   fscrypt-crypt-util: create key_and_iv_params structure
> > >   fscrypt-crypt-util: add HKDF context constants
> > >   common/encrypt: create named variables for UAPI constants
> > >   common/encrypt: support verifying ciphertext of IV_INO_LBLK_64
> > >     policies
> > >   generic: verify ciphertext of IV_INO_LBLK_64 encryption policies
> > > 
> > >  common/encrypt           | 126 +++++++++++++++++++++++++-------
> > >  src/fscrypt-crypt-util.c | 151 ++++++++++++++++++++++++++++-----------
> > >  tests/generic/805        |  43 +++++++++++
> > >  tests/generic/805.out    |   6 ++
> > >  tests/generic/group      |   1 +
> > >  5 files changed, 259 insertions(+), 68 deletions(-)
> > >  create mode 100644 tests/generic/805
> > >  create mode 100644 tests/generic/805.out
> > > 
> > 
> > Ping.  Does anyone want to take a look at this?  Satya?
> > 
> 
> Eryu, can you review and consider applying this series?  It doesn't look like
> anyone else is going to formally review it.

I've reviewed & applied this patchset in my local tree, will push them
out soon. So sorry for the delay!

Thanks,
Eryu