Developer Reference


Keyed Hash Functions

The Intel IPP HMAC primitive functions, described in this section, use various HMAC schemes based on one-way hash functions described in One-Way Hash Primitives.
Usage model of the generalized HMAC functions is similar to the model explained below.
Each HMAC scheme is implemented as a set of the primitive functions.
primitive implementing HMAC uses the
context as an operational vehicle to carry all necessary variables to manage computation of the chaining digest value.
The following example illustrates how the application code can apply the implemented HMAC-SHA1 hash standard to digest the input message stream:
  1. Call the function HMAC_GetSize to get the size required to configure the
  2. Ensure that the required memory space is properly allocated. With the allocated memory, call the function
    HMAC_Init with the value of
    equal to
    to set up key material and the initial context state with the SHA-1 specified initialization vectors.
  3. Keep calling the function HMAC_Update to digest incoming message stream in the queue till its completion. To determine the current value of the message digest, call HMAC_GetTag between the two calls to
  4. Call the function HMAC_Final to pad the partial block into a final SHA-1 message block and transform it into a resulting HMAC value.
  5. Clean up secret data stored in the context.
  6. Call the operating system memory free service function to release the
context is position-dependent. The
HMACPack, HMACUnpack
functions transform it to a position-independent form and vice versa:
The crypto community does not consider HMACSHA1 or HMACMD5 secure anymore.
Recommendation: use a more secure hash algorithm (for example, any algorithm from the SHA-2 family) instead of HMACSHA1 or HMACMD5.

Product and Performance Information


Performance varies by use, configuration and other factors. Learn more at