Tomáš Mráz
Tomáš Mráz
I wanted to push you towards doing the right fix. However as I did it myself, and other assembler modules basically do the same (i.e. no-op for less than block...
Merged to all the active branches. Thank you for your contribution.
Actually all the checks required for the common prime curves are already happening as part of importing the public key into the provider. The only thing missing is the full...
The check is being done when the EC_POINT coordinates are set internally. https://github.com/openssl/openssl/blob/6dacee485fad2c4d334e08af48891636205ddb6b/crypto/ec/ec_lib.c#L878
In ECDH the check is actually done twice, but it is not expensive (well, only after https://github.com/openssl/openssl/commit/b916940752e4de5922553b1cf482687dfc653f7a) so it is OK.
IMO verification 1 and 2 are implicit in the way the public key is decoded as well. And I do not know what verification 4 are you talking about. Also...
@levitte there are some unresolved comments - please open the Files changed view and look at the unresolved comments.
Merged to the master branch. Thank you for your contribution.
We cannot change the OID without breaking backwards compatibility with existing (albeit wrongly) encoded data. We could add the correct OID as an alias for the algorithm IMO but we...
The conflict in libcrypto.num is trivial.