With some x509 related issues fixed, advertise revision 155 now
so that IBM TSS 2 test suite passes the X509 tests.
Signed-off-by: Stefan Berger <stefanb@linux.ibm.com>
When a child key is derived from an Object, we need to know what
SEED_COMPAT_LEVEL the Object was created under (or what the its seed is)
so that we can later on derive child with the appropriate
SEED_COMPAT_LEVEL.
When an external object is loaded we set the SEED_COMPAT_LEVEL to
SEED_COMPAT_LEVEL_ORIGINAL so that any RSA key that could possibly be
derived from it will always create the same RSA key independent
of the SEED_COMPAT_LEVEL of the hierarchy it is loaded under. This
way any key derivation that could happen on an external key remains
the same before and after this code change.
Signed-off-by: Stefan Berger <stefanb@linux.ibm.com>
Pass the SEED_COMPAT_LEVEL, originating from the seed that's being used,
to the CryptAdjustPrimeCandidate function and use it to determine
whether the old code should be used or the new one.
Signed-off-by: Stefan Berger <stefanb@linux.ibm.com>
Add in the new RsaAdjustPrimeCandidate() function but do not use it
so far since it creates slightly different primes than the previous
code and we would get different derived keys if we were to use it with
'old' seeds.
Adjust the code to return the same results for 64 bit and 32 bit machines.
Signed-off-by: Stefan Berger <stefanb@linux.ibm.com>
Add a SEED_COMPAT_LEVEL to the DRBG state that is associated with the seed
and indicates the seed compatibility level we need to maintain when deriving
RSA keys from seeds. We only need to be able to handle RSA keys derived via
the DRBG state. Other keys, such as symmetric keys, are not affected. Also
RSA keys cannot be derived from a KDR, so the KDF does not need to carry
the SEED_COMPAT_LEVEL.
All functions that need to pass a value set SEED_COMPAT_LEVEL to
SEED_COMPAT_LEVEL_ORIGINAL (0) for now.
Signed-off-by: Stefan Berger <stefanb@linux.ibm.com>
Add SEED_COMPAT_LEVEL to the nullSeed in the state_reset data to track
its compatibility level. We need it for VM suspend and resume.
Signed-off-by: Stefan Berger <stefanb@linux.ibm.com>
Add a variable seedCompatLevel to the each seed in the PERSISTENT_DATA
that allows us to track the age of the seed. Whenever a new seed
is created the seedCompatLevel is also written and set to the latest
version. This seedCompatLevel then influences the crypto algorithm that
can be used for deriving keys so that previously derived keys
are now still generated in the same way. When the seed is changed
the old keys are all useless and newly derived keys can then use
the new algorithm.
This patch only sets the variables to the current compatibility
level SEED_COMPAT_LEVEL_ORIGINAL and writes it out as part of the state
file. This makes the state file not downgradeable.
Signed-off-by: Stefan Berger <stefanb@linux.ibm.com>
- Introduce Platform.h and replace usage of PlatformData.h and
Platform_fp.h
- Drop Implementation.h since we now use TpmProfile.h (which we
prepared previously); many #defines were moved to TpmTypes.h
Signed-off-by: Stefan Berger <stefanb@linux.ibm.com>
