diff options
author | David Howells <dhowells@redhat.com> | 2016-03-03 21:49:27 +0000 |
---|---|---|
committer | David Howells <dhowells@redhat.com> | 2016-03-03 21:49:27 +0000 |
commit | d43de6c780a84def056afaf4fb3e66bdaa1efc00 (patch) | |
tree | f6221a2761cd7fbdb9c5bdc3e82b6dbfef90430a /crypto/authencesn.c | |
parent | a49de377e051eac5bc50a3b838614a05671da4e7 (diff) | |
download | linux-d43de6c780a84def056afaf4fb3e66bdaa1efc00.tar.bz2 |
akcipher: Move the RSA DER encoding check to the crypto layer
Move the RSA EMSA-PKCS1-v1_5 encoding from the asymmetric-key public_key
subtype to the rsa crypto module's pkcs1pad template. This means that the
public_key subtype no longer has any dependencies on public key type.
To make this work, the following changes have been made:
(1) The rsa pkcs1pad template is now used for RSA keys. This strips off the
padding and returns just the message hash.
(2) In a previous patch, the pkcs1pad template gained an optional second
parameter that, if given, specifies the hash used. We now give this,
and pkcs1pad checks the encoded message E(M) for the EMSA-PKCS1-v1_5
encoding and verifies that the correct digest OID is present.
(3) The crypto driver in crypto/asymmetric_keys/rsa.c is now reduced to
something that doesn't care about what the encryption actually does
and and has been merged into public_key.c.
(4) CONFIG_PUBLIC_KEY_ALGO_RSA is gone. Module signing must set
CONFIG_CRYPTO_RSA=y instead.
Thoughts:
(*) Should the encoding style (eg. raw, EMSA-PKCS1-v1_5) also be passed to
the padding template? Should there be multiple padding templates
registered that share most of the code?
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Tadeusz Struk <tadeusz.struk@intel.com>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Diffstat (limited to 'crypto/authencesn.c')
0 files changed, 0 insertions, 0 deletions