crypto: aesni - Add support for 192 & 256 bit keys to AESNI RFC4106

These patches fix the RFC4106 implementation in the aesni-intel
module so it supports 192 & 256 bit keys.

Since the AVX support that was added to this module also only
supports 128 bit keys, and this patch only affects the SSE
implementation, changes were also made to use the SSE version
if key sizes other than 128 are specified.

RFC4106 specifies that 192 & 256 bit keys must be supported (section
8.4).

Also, this should fix Strongswan issue 341 where the aesni module
needs to be unloaded if 256 bit keys are used:

http://wiki.strongswan.org/issues/341

This patch has been tested with Sandy Bridge and Haswell processors.
With 128 bit keys and input buffers > 512 bytes a slight performance
degradation was noticed (~1%).  For input buffers of less than 512
bytes there was no performance impact.  Compared to 128 bit keys,
256 bit key size performance is approx. .5 cycles per byte slower
on Sandy Bridge, and .37 cycles per byte slower on Haswell (vs.
SSE code).

This patch has also been tested with StrongSwan IPSec connections
where it worked correctly.

I created this diff from a git clone of crypto-2.6.git.

Any questions, please feel free to contact me.

Signed-off-by: Timothy McCaffrey <timothy.mccaffrey@unisys.com>
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

1 files changed, 28 insertions, 6 deletions

diff --git a/arch/x86/crypto/aesni-intel_glue.c b/arch/x86/crypto/aesni-intel_glue.c
index ae855f4f64b7..947c6bf52c33 100644
--- a/arch/x86/crypto/aesni-intel_glue.c
+++ b/arch/x86/crypto/aesni-intel_glue.c
@@ -43,6 +43,7 @@
 #include <asm/crypto/glue_helper.h>
 #endif
 
+
 /* This data is stored at the end of the crypto_tfm struct.
  * It's a type of per "session" data storage location.
  * This needs to be 16 byte aligned.
@@ -182,7 +183,8 @@ static void aesni_gcm_enc_avx(void *ctx, u8 *out,
 			u8 *hash_subkey, const u8 *aad, unsigned long aad_len,
 			u8 *auth_tag, unsigned long auth_tag_len)
 {
-	if (plaintext_len < AVX_GEN2_OPTSIZE) {
+        struct crypto_aes_ctx *aes_ctx = (struct crypto_aes_ctx*)ctx;
+	if ((plaintext_len < AVX_GEN2_OPTSIZE) || (aes_ctx-> key_length != AES_KEYSIZE_128)){
 		aesni_gcm_enc(ctx, out, in, plaintext_len, iv, hash_subkey, aad,
 				aad_len, auth_tag, auth_tag_len);
 	} else {
@@ -197,7 +199,8 @@ static void aesni_gcm_dec_avx(void *ctx, u8 *out,
 			u8 *hash_subkey, const u8 *aad, unsigned long aad_len,
 			u8 *auth_tag, unsigned long auth_tag_len)
 {
-	if (ciphertext_len < AVX_GEN2_OPTSIZE) {
+        struct crypto_aes_ctx *aes_ctx = (struct crypto_aes_ctx*)ctx;
+	if ((ciphertext_len < AVX_GEN2_OPTSIZE) || (aes_ctx-> key_length != AES_KEYSIZE_128)) {
 		aesni_gcm_dec(ctx, out, in, ciphertext_len, iv, hash_subkey, aad,
 				aad_len, auth_tag, auth_tag_len);
 	} else {
@@ -231,7 +234,8 @@ static void aesni_gcm_enc_avx2(void *ctx, u8 *out,
 			u8 *hash_subkey, const u8 *aad, unsigned long aad_len,
 			u8 *auth_tag, unsigned long auth_tag_len)
 {
-	if (plaintext_len < AVX_GEN2_OPTSIZE) {
+       struct crypto_aes_ctx *aes_ctx = (struct crypto_aes_ctx*)ctx;
+	if ((plaintext_len < AVX_GEN2_OPTSIZE) || (aes_ctx-> key_length != AES_KEYSIZE_128)) {
 		aesni_gcm_enc(ctx, out, in, plaintext_len, iv, hash_subkey, aad,
 				aad_len, auth_tag, auth_tag_len);
 	} else if (plaintext_len < AVX_GEN4_OPTSIZE) {
@@ -250,7 +254,8 @@ static void aesni_gcm_dec_avx2(void *ctx, u8 *out,
 			u8 *hash_subkey, const u8 *aad, unsigned long aad_len,
 			u8 *auth_tag, unsigned long auth_tag_len)
 {
-	if (ciphertext_len < AVX_GEN2_OPTSIZE) {
+       struct crypto_aes_ctx *aes_ctx = (struct crypto_aes_ctx*)ctx;
+	if ((ciphertext_len < AVX_GEN2_OPTSIZE) || (aes_ctx-> key_length != AES_KEYSIZE_128)) {
 		aesni_gcm_dec(ctx, out, in, ciphertext_len, iv, hash_subkey,
 				aad, aad_len, auth_tag, auth_tag_len);
 	} else if (ciphertext_len < AVX_GEN4_OPTSIZE) {
@@ -511,7 +516,7 @@ static int ctr_crypt(struct blkcipher_desc *desc,
 	kernel_fpu_begin();
 	while ((nbytes = walk.nbytes) >= AES_BLOCK_SIZE) {
 		aesni_ctr_enc_tfm(ctx, walk.dst.virt.addr, walk.src.virt.addr,
-				  nbytes & AES_BLOCK_MASK, walk.iv);
+			              nbytes & AES_BLOCK_MASK, walk.iv);
 		nbytes &= AES_BLOCK_SIZE - 1;
 		err = blkcipher_walk_done(desc, &walk, nbytes);
 	}
@@ -902,7 +907,8 @@ static int rfc4106_set_key(struct crypto_aead *parent, const u8 *key,
 	}
 	/*Account for 4 byte nonce at the end.*/
 	key_len -= 4;
-	if (key_len != AES_KEYSIZE_128) {
+	if (key_len != AES_KEYSIZE_128 && key_len != AES_KEYSIZE_192 &&
+	    key_len != AES_KEYSIZE_256) {
 		crypto_tfm_set_flags(tfm, CRYPTO_TFM_RES_BAD_KEY_LEN);
 		return -EINVAL;
 	}
@@ -1013,6 +1019,7 @@ static int __driver_rfc4106_encrypt(struct aead_request *req)
 	__be32 counter = cpu_to_be32(1);
 	struct crypto_aead *tfm = crypto_aead_reqtfm(req);
 	struct aesni_rfc4106_gcm_ctx *ctx = aesni_rfc4106_gcm_ctx_get(tfm);
+	u32 key_len = ctx->aes_key_expanded.key_length;
 	void *aes_ctx = &(ctx->aes_key_expanded);
 	unsigned long auth_tag_len = crypto_aead_authsize(tfm);
 	u8 iv_tab[16+AESNI_ALIGN];
@@ -1027,6 +1034,13 @@ static int __driver_rfc4106_encrypt(struct aead_request *req)
 	/* to 8 or 12 bytes */
 	if (unlikely(req->assoclen != 8 && req->assoclen != 12))
 		return -EINVAL;
+	if (unlikely(auth_tag_len != 8 && auth_tag_len != 12 && auth_tag_len != 16))
+	        return -EINVAL;
+	if (unlikely(key_len != AES_KEYSIZE_128 &&
+	             key_len != AES_KEYSIZE_192 &&
+	             key_len != AES_KEYSIZE_256))
+	        return -EINVAL;
+
 	/* IV below built */
 	for (i = 0; i < 4; i++)
 		*(iv+i) = ctx->nonce[i];
@@ -1091,6 +1105,7 @@ static int __driver_rfc4106_decrypt(struct aead_request *req)
 	int retval = 0;
 	struct crypto_aead *tfm = crypto_aead_reqtfm(req);
 	struct aesni_rfc4106_gcm_ctx *ctx = aesni_rfc4106_gcm_ctx_get(tfm);
+	u32 key_len = ctx->aes_key_expanded.key_length;
 	void *aes_ctx = &(ctx->aes_key_expanded);
 	unsigned long auth_tag_len = crypto_aead_authsize(tfm);
 	u8 iv_and_authTag[32+AESNI_ALIGN];
@@ -1104,6 +1119,13 @@ static int __driver_rfc4106_decrypt(struct aead_request *req)
 	if (unlikely((req->cryptlen < auth_tag_len) ||
 		(req->assoclen != 8 && req->assoclen != 12)))
 		return -EINVAL;
+	if (unlikely(auth_tag_len != 8 && auth_tag_len != 12 && auth_tag_len != 16))
+	        return -EINVAL;
+	if (unlikely(key_len != AES_KEYSIZE_128 &&
+	             key_len != AES_KEYSIZE_192 &&
+	             key_len != AES_KEYSIZE_256))
+	        return -EINVAL;
+
 	/* Assuming we are supporting rfc4106 64-bit extended */
 	/* sequence numbers We need to have the AAD length */
 	/* equal to 8 or 12 bytes */