[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: RE: [pkcs11] FIPS and GCM
Daniel, Thank you for the pointer to the 3.0 spec. It’s pretty clear to me how to do it using that version. Sincerely, Jonathan From: Daniel Minder <Daniel.Minder@utimaco.com> Jonathan, actually a good question that is addressed in the PKCS #11 3.0 standard by the introduction of the message based functions (C_MessageEncryptInit, C_EncryptMessage,
…) and of a new parameter structure for GCM (CK_GCM_MESSAGE_PARAMS). Together, they allow to specify that an IV of a certain length shall be generated by the token and allow to return this IV. Returning the IV in non-message-based functions (C_EncryptInit, C_Encrypt,
…) are vendor defined extensions and not covered by the standard. Regards, Daniel From: pkcs11@lists.oasis-open.org <pkcs11@lists.oasis-open.org>
On Behalf Of Jonathan Schulze-Hewett All, Likely a stupid question, but as FIPS validation appears to prohibit the use of External IVs for GCM what is the expected behavior of C_EncryptInit with GCM as a mechanism? Is it expected that the IV field of the input parameters will be
changed by the C_EncryptInit call so that the module generated IV can be communicated? Is there or should there be a value in the parameters that indicates what form of IV generation is needed? Thanks, Jonathan Jonathan Schulze-Hewett Director of Development Information Security Corp. schulze-hewett@infoseccorp.com 708-445-1704
|
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]