Compute AES CMAC using OpenSSL/libcrypto

When I googled on how to perform the AES CMAC calculation using OpenSSL/libcrypto I couldn’t find any code example. I was no even sure that there was support for it of libcrypto so I downloaded the OpenSSL sources and dig into them until I found [crypto/cmac/cmac.c](/cmac/cmac.c]( https://github.com/openssl/openssl/blob/master/crypto/cmac/cmac.c). From there it was obvious that there is support for CMAC and that it follow the *{Init,Update,Final} pattern found in EVP api. So I give it I try and I’m sharing the basic source code here for later reference:

	// compile with
	// cc `PKG_CONFIG_PATH="/usr/local/opt/openssl@1.1/lib/pkgconfig" pkg-config --cflags --libs openssl` mac_example.c -o mac_example
	// cc -I/usr/local/Cellar/openssl@1.1/1.1.1h/include -L/usr/local/Cellar/openssl@1.1/1.1.1h/lib -lssl -lcrypto mac_example.c
	#include <stdio.h>
	#include <openssl/cmac.h>

	void printBytes(unsigned char *buf, size_t len) {
	for(int i=0; i<len; i++) {
	printf("%02x ", buf[i]);
	}
	printf("\n");
	}

	int main(int argc, char *argv[])
	{
	// https://tools.ietf.org/html/rfc4493

	// K, M and T from
	// http://csrc.nist.gov/publications/nistpubs/800-38B/Updated_CMAC_Examples.pdf
	// D.1 AES-128

	// K: 2b7e1516 28aed2a6 abf71588 09cf4f3c
	unsigned char key[] = { 0x2b,0x7e,0x15,0x16,
	0x28,0xae,0xd2,0xa6,
	0xab,0xf7,0x15,0x88,
	0x09,0xcf,0x4f,0x3c};

	// M: 6bc1bee2 2e409f96 e93d7e11 7393172a Mlen: 128
	unsigned char message[] = { 0x6b,0xc1,0xbe,0xe2,
	0x2e,0x40,0x9f,0x96,
	0xe9,0x3d,0x7e,0x11,
	0x73,0x93,0x17,0x2a };

	unsigned char mact[16] = {0};
	size_t mactlen;

	CMAC_CTX *ctx = CMAC_CTX_new();
	CMAC_Init(ctx, key, 16, EVP_aes_128_cbc(), NULL);
	printf("message length = %lu bytes (%lu bits)\n",sizeof(message), sizeof(message)*8);

	CMAC_Update(ctx, message, sizeof(message));
	CMAC_Final(ctx, mact, &mactlen);

	printBytes(mact, mactlen);
	/* expected result T = 070a16b4 6b4d4144 f79bdd9d d04a287c */

	CMAC_CTX_free(ctx);
	return 0;
	}

view raw mac_example.c hosted with ❤ by GitHub

I used the example data from NIST doc to validate the results.

The functions used are:

CMAC_CTX_new: allocates a context
CMAC_Init: configure the context to use AES-128-CBC
CMAC_Update: Input the message, you can have several calls to it.
CMAC_Final: Generate the CMAC

Unfortunately I believe that the CMAC implementation doesn’t make use of AES-NI. In other words it doesn’t make use of the hardware acceleration on Intel x86-64.