RE: [dpdk-dev v3] crypto/qat: use intel-ipsec-mb for partial hash & aes

["Zhang, Roy Fan" <roy.fan.zhang@intel.com>]

Hi Kai,

> -----Original Message-----
> From: Ji, Kai <kai.ji@intel.com>
> Sent: Wednesday, May 25, 2022 5:52 PM
> To: dev@dpdk.org
> Cc: Zhang, Roy Fan <roy.fan.zhang@intel.com>; Richardson, Bruce
> <bruce.richardson@intel.com>; gakhil@marvell.com; Ji, Kai <kai.ji@intel.com>
> Subject: [dpdk-dev v3] crypto/qat: use intel-ipsec-mb for partial hash & aes
> 
> Since openssl 3.0 now deprecates the low level API QAT required to
> perform partial hash & aes operation when creating the session. This
> patch add in qat_ipsec_mb_lib driver parameter to allow QAT PMD to
> switch APIs between openssl and intel ipsec-mb library.
> 
> Signed-off-by: Kai Ji <kai.ji@intel.com>
> Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com>
> ---

<snip>

> +static int qat_sym_do_precomputes_ipsec_mb(enum
> icp_qat_hw_auth_algo hash_alg,
> +				const uint8_t *auth_key,
> +				uint16_t auth_keylen,
> +				uint8_t *p_state_buf,
> +				uint16_t *p_state_len,
> +				uint8_t aes_cmac)
> +{
> +	int block_size;
> +	uint8_t
> ipad[qat_hash_get_block_size(ICP_QAT_HW_AUTH_ALGO_DELIMITER)];
> +	uint8_t
> opad[qat_hash_get_block_size(ICP_QAT_HW_AUTH_ALGO_DELIMITER)];
> +	int i;
> +
> +	IMB_MGR *m;
> +	m = alloc_mb_mgr(0);
> +	if (m == NULL)
> +		return -ENOMEM;
> +
> +	init_mb_mgr_auto(m, NULL);
> +
> +	if (hash_alg == ICP_QAT_HW_AUTH_ALGO_AES_XCBC_MAC) {
> +
> +		/* CMAC */
> +		if (aes_cmac) {
> +			uint8_t *in = NULL;
> +			uint8_t k0[ICP_QAT_HW_AES_128_KEY_SZ];
> +			uint8_t *k1, *k2;
> +
> +			auth_keylen = ICP_QAT_HW_AES_128_KEY_SZ;
> +
> +			in = rte_zmalloc("AES CMAC K1",
> +					 ICP_QAT_HW_AES_128_KEY_SZ, 16);
> +
> +			if (in == NULL) {
> +				QAT_LOG(ERR, "Failed to alloc memory");
> +				return -ENOMEM;
> +			}
> +
> +			rte_memcpy(in, AES_CMAC_SEED,
> +				   ICP_QAT_HW_AES_128_KEY_SZ);
> +			rte_memcpy(p_state_buf, auth_key, auth_keylen);
> +
> +			DECLARE_ALIGNED(uint32_t expkey[4*15], 16);
> +			DECLARE_ALIGNED(uint32_t dust[4*15], 16);
> +			IMB_AES_KEYEXP_128(m, p_state_buf, expkey,
> dust);
> +			k1 = p_state_buf +
> ICP_QAT_HW_AES_XCBC_MAC_STATE1_SZ;
> +			k2 = k1 + ICP_QAT_HW_AES_XCBC_MAC_STATE1_SZ;
> +
> +			IMB_AES_CMAC_SUBKEY_GEN_128(m, expkey, k1,
> k2);
> +			memset(k0, 0, ICP_QAT_HW_AES_128_KEY_SZ);
> +			*p_state_len =
> ICP_QAT_HW_AES_XCBC_MAC_STATE2_SZ;
> +			rte_free(in);
> +			free_mb_mgr(m);
> +			return 0;
> +		}
> +
> +		static uint8_t qat_aes_xcbc_key_seed[
> +				ICP_QAT_HW_AES_XCBC_MAC_STATE2_SZ]
> = {
> +			0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
> +			0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
> +			0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02,
> +			0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02,
> +			0x03, 0x03, 0x03, 0x03, 0x03, 0x03, 0x03, 0x03,
> +			0x03, 0x03, 0x03, 0x03, 0x03, 0x03, 0x03, 0x03,
> +		};
> +
> +		uint8_t *in = NULL;
> +		uint8_t *out = p_state_buf;
> +		int x;
> +
> +		in = rte_zmalloc("working mem for key",
> +				ICP_QAT_HW_AES_XCBC_MAC_STATE2_SZ,
> 16);
> +		if (in == NULL) {
> +			QAT_LOG(ERR, "Failed to alloc memory");
> +			return -ENOMEM;
> +		}
> +
> +		rte_memcpy(in, qat_aes_xcbc_key_seed,
> +				ICP_QAT_HW_AES_XCBC_MAC_STATE2_SZ);
> +		for (x = 0; x < HASH_XCBC_PRECOMP_KEY_NUM; x++) {
> +			if (aes_ipsecmb_job(in, out, m, auth_key,
> auth_keylen)) {
> +				rte_free(in -
> +				  (x *
> ICP_QAT_HW_AES_XCBC_MAC_KEY_SZ));
> +				memset(out -
> +				   (x *
> ICP_QAT_HW_AES_XCBC_MAC_KEY_SZ),
> +				  0,
> ICP_QAT_HW_AES_XCBC_MAC_STATE2_SZ);
> +				return -EFAULT;
> +			}
> +
> +			in += ICP_QAT_HW_AES_XCBC_MAC_KEY_SZ;
> +			out += ICP_QAT_HW_AES_XCBC_MAC_KEY_SZ;
> +		}
> +		*p_state_len = ICP_QAT_HW_AES_XCBC_MAC_STATE2_SZ;
> +		rte_free(in - x*ICP_QAT_HW_AES_XCBC_MAC_KEY_SZ);
> +		free_mb_mgr(m);
> +		return 0;
> +
> +	} else if ((hash_alg == ICP_QAT_HW_AUTH_ALGO_GALOIS_128) ||
> +		(hash_alg == ICP_QAT_HW_AUTH_ALGO_GALOIS_64)) {
> +		uint8_t *in = NULL;
> +		uint8_t *out = p_state_buf;
> +
> +		memset(p_state_buf, 0, ICP_QAT_HW_GALOIS_H_SZ +
> +				ICP_QAT_HW_GALOIS_LEN_A_SZ +
> +				ICP_QAT_HW_GALOIS_E_CTR0_SZ);
> +		in = rte_zmalloc("working mem for key",
> +				ICP_QAT_HW_GALOIS_H_SZ, 16);
> +		if (in == NULL) {
> +			QAT_LOG(ERR, "Failed to alloc memory");
> +			return -ENOMEM;
> +		}
> +
> +		memset(in, 0, ICP_QAT_HW_GALOIS_H_SZ);
> +		if (aes_ipsecmb_job(in, out, m, auth_key, auth_keylen))
> +			return -EFAULT;
> +
> +		*p_state_len = ICP_QAT_HW_GALOIS_H_SZ +
> +				ICP_QAT_HW_GALOIS_LEN_A_SZ +
> +				ICP_QAT_HW_GALOIS_E_CTR0_SZ;
> +		rte_free(in);
> +		free_mb_mgr(m);
> +		return 0;
> +	}
> +

From this point on, mb_mgr is not freed either error happens or the function exists
Normally.
 
> +	block_size = qat_hash_get_block_size(hash_alg);
> +	if (block_size < 0)
> +		return block_size;
> +	/* init ipad and opad from key and xor with fixed values */
> +	memset(ipad, 0, block_size);
> +	memset(opad, 0, block_size);
> +
> +	if (auth_keylen > (unsigned int)block_size) {
> +		QAT_LOG(ERR, "invalid keylen %u", auth_keylen);
> +		return -EFAULT;
> +	}
> +	rte_memcpy(ipad, auth_key, auth_keylen);
> +	rte_memcpy(opad, auth_key, auth_keylen);
> +
> +	for (i = 0; i < block_size; i++) {
> +		uint8_t *ipad_ptr = ipad + i;
> +		uint8_t *opad_ptr = opad + i;
> +		*ipad_ptr ^= HMAC_IPAD_VALUE;
> +		*opad_ptr ^= HMAC_OPAD_VALUE;
> +	}
> +
> +	/* do partial hash of ipad and copy to state1 */
> +	if (partial_hash_compute_ipsec_mb(hash_alg, ipad, p_state_buf)) {
> +		memset(ipad, 0, block_size);
> +		memset(opad, 0, block_size);
> +		QAT_LOG(ERR, "ipad precompute failed");
> +		return -EFAULT;
> +	}
> +
> +	/*
> +	 * State len is a multiple of 8, so may be larger than the digest.
> +	 * Put the partial hash of opad state_len bytes after state1
> +	 */
> +	*p_state_len = qat_hash_get_state1_size(hash_alg);
> +	if (partial_hash_compute_ipsec_mb(hash_alg, opad, p_state_buf +
> *p_state_len)) {
> +		memset(ipad, 0, block_size);
> +		memset(opad, 0, block_size);
> +		QAT_LOG(ERR, "opad precompute failed");
> +		return -EFAULT;
> +	}
> +
> +	/*  don't leave data lying around */
> +	memset(ipad, 0, block_size);
> +	memset(opad, 0, block_size);
> +	return 0;
> +}