arm64/crypto: use crypto instructions to generate AES key schedule

This patch implements the AES key schedule generation using ARMv8
Crypto Instructions. It replaces the table based C implementation
in aes_generic.ko, which means we can drop the dependency on that
module.

Tested-by: Steve Capper <steve.capper@linaro.org>
Acked-by: Steve Capper <steve.capper@linaro.org>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Will Deacon <will.deacon@arm.com>

arch/arm64/crypto/Kconfig

@@ -27,20 +27,19 @@ config CRYPTO_AES_ARM64_CE
 	tristate "AES core cipher using ARMv8 Crypto Extensions"
 	depends on ARM64 && KERNEL_MODE_NEON
 	select CRYPTO_ALGAPI
-	select CRYPTO_AES
 
 config CRYPTO_AES_ARM64_CE_CCM
 	tristate "AES in CCM mode using ARMv8 Crypto Extensions"
 	depends on ARM64 && KERNEL_MODE_NEON
 	select CRYPTO_ALGAPI
-	select CRYPTO_AES
+	select CRYPTO_AES_ARM64_CE
 	select CRYPTO_AEAD
 
 config CRYPTO_AES_ARM64_CE_BLK
 	tristate "AES in ECB/CBC/CTR/XTS modes using ARMv8 Crypto Extensions"
 	depends on ARM64 && KERNEL_MODE_NEON
 	select CRYPTO_BLKCIPHER
-	select CRYPTO_AES
+	select CRYPTO_AES_ARM64_CE
 	select CRYPTO_ABLK_HELPER
 
 config CRYPTO_AES_ARM64_NEON_BLK

arch/arm64/crypto/aes-ce-ccm-glue.c

@@ -16,6 +16,8 @@
 #include <linux/crypto.h>
 #include <linux/module.h>
 
+#include "aes-ce-setkey.h"
+
 static int num_rounds(struct crypto_aes_ctx *ctx)
 {
 	/*
@@ -48,7 +50,7 @@ static int ccm_setkey(struct crypto_aead *tfm, const u8 *in_key,
 	struct crypto_aes_ctx *ctx = crypto_aead_ctx(tfm);
 	int ret;
 
-	ret = crypto_aes_expand_key(ctx, in_key, key_len);
+	ret = ce_aes_expandkey(ctx, in_key, key_len);
 	if (!ret)
 		return 0;
 

arch/arm64/crypto/aes-ce-cipher.c

@@ -14,6 +14,8 @@
 #include <linux/crypto.h>
 #include <linux/module.h>
 
+#include "aes-ce-setkey.h"
+
 MODULE_DESCRIPTION("Synchronous AES cipher using ARMv8 Crypto Extensions");
 MODULE_AUTHOR("Ard Biesheuvel <ard.biesheuvel@linaro.org>");
 MODULE_LICENSE("GPL v2");
@@ -124,6 +126,114 @@ static void aes_cipher_decrypt(struct crypto_tfm *tfm, u8 dst[], u8 const src[])
 	kernel_neon_end();
 }
 
+/*
+ * aes_sub() - use the aese instruction to perform the AES sbox substitution
+ *             on each byte in 'input'
+ */
+static u32 aes_sub(u32 input)
+{
+	u32 ret;
+
+	__asm__("dup	v1.4s, %w[in]		;"
+		"movi	v0.16b, #0		;"
+		"aese	v0.16b, v1.16b		;"
+		"umov	%w[out], v0.4s[0]	;"
+
+	:	[out]	"=r"(ret)
+	:	[in]	"r"(input)
+	:		"v0","v1");
+
+	return ret;
+}
+
+int ce_aes_expandkey(struct crypto_aes_ctx *ctx, const u8 *in_key,
+		     unsigned int key_len)
+{
+	/*
+	 * The AES key schedule round constants
+	 */
+	static u8 const rcon[] = {
+		0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80, 0x1b, 0x36,
+	};
+
+	u32 kwords = key_len / sizeof(u32);
+	struct aes_block *key_enc, *key_dec;
+	int i, j;
+
+	if (key_len != AES_KEYSIZE_128 &&
+	    key_len != AES_KEYSIZE_192 &&
+	    key_len != AES_KEYSIZE_256)
+		return -EINVAL;
+
+	memcpy(ctx->key_enc, in_key, key_len);
+	ctx->key_length = key_len;
+
+	kernel_neon_begin_partial(2);
+	for (i = 0; i < sizeof(rcon); i++) {
+		u32 *rki = ctx->key_enc + (i * kwords);
+		u32 *rko = rki + kwords;
+
+		rko[0] = ror32(aes_sub(rki[kwords - 1]), 8) ^ rcon[i] ^ rki[0];
+		rko[1] = rko[0] ^ rki[1];
+		rko[2] = rko[1] ^ rki[2];
+		rko[3] = rko[2] ^ rki[3];
+
+		if (key_len == AES_KEYSIZE_192) {
+			if (i >= 7)
+				break;
+			rko[4] = rko[3] ^ rki[4];
+			rko[5] = rko[4] ^ rki[5];
+		} else if (key_len == AES_KEYSIZE_256) {
+			if (i >= 6)
+				break;
+			rko[4] = aes_sub(rko[3]) ^ rki[4];
+			rko[5] = rko[4] ^ rki[5];
+			rko[6] = rko[5] ^ rki[6];
+			rko[7] = rko[6] ^ rki[7];
+		}
+	}
+
+	/*
+	 * Generate the decryption keys for the Equivalent Inverse Cipher.
+	 * This involves reversing the order of the round keys, and applying
+	 * the Inverse Mix Columns transformation on all but the first and
+	 * the last one.
+	 */
+	key_enc = (struct aes_block *)ctx->key_enc;
+	key_dec = (struct aes_block *)ctx->key_dec;
+	j = num_rounds(ctx);
+
+	key_dec[0] = key_enc[j];
+	for (i = 1, j--; j > 0; i++, j--)
+		__asm__("ld1	{v0.16b}, %[in]		;"
+			"aesimc	v1.16b, v0.16b		;"
+			"st1	{v1.16b}, %[out]	;"
+
+		:	[out]	"=Q"(key_dec[i])
+		:	[in]	"Q"(key_enc[j])
+		:		"v0","v1");
+	key_dec[i] = key_enc[0];
+
+	kernel_neon_end();
+	return 0;
+}
+EXPORT_SYMBOL(ce_aes_expandkey);
+
+int ce_aes_setkey(struct crypto_tfm *tfm, const u8 *in_key,
+		  unsigned int key_len)
+{
+	struct crypto_aes_ctx *ctx = crypto_tfm_ctx(tfm);
+	int ret;
+
+	ret = ce_aes_expandkey(ctx, in_key, key_len);
+	if (!ret)
+		return 0;
+
+	tfm->crt_flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
+	return -EINVAL;
+}
+EXPORT_SYMBOL(ce_aes_setkey);
+
 static struct crypto_alg aes_alg = {
 	.cra_name		= "aes",
 	.cra_driver_name	= "aes-ce",
@@ -135,7 +245,7 @@ static struct crypto_alg aes_alg = {
 	.cra_cipher = {
 		.cia_min_keysize	= AES_MIN_KEY_SIZE,
 		.cia_max_keysize	= AES_MAX_KEY_SIZE,
-		.cia_setkey		= crypto_aes_set_key,
+		.cia_setkey		= ce_aes_setkey,
 		.cia_encrypt		= aes_cipher_encrypt,
 		.cia_decrypt		= aes_cipher_decrypt
 	}

arch/arm64/crypto/aes-ce-setkey.h

@@ -0,0 +1,5 @@
+
+int ce_aes_setkey(struct crypto_tfm *tfm, const u8 *in_key,
+		  unsigned int key_len);
+int ce_aes_expandkey(struct crypto_aes_ctx *ctx, const u8 *in_key,
+		     unsigned int key_len);

arch/arm64/crypto/aes-glue.c

@@ -16,9 +16,13 @@
 #include <linux/module.h>
 #include <linux/cpufeature.h>
 
+#include "aes-ce-setkey.h"
+
 #ifdef USE_V8_CRYPTO_EXTENSIONS
 #define MODE			"ce"
 #define PRIO			300
+#define aes_setkey		ce_aes_setkey
+#define aes_expandkey		ce_aes_expandkey
 #define aes_ecb_encrypt		ce_aes_ecb_encrypt
 #define aes_ecb_decrypt		ce_aes_ecb_decrypt
 #define aes_cbc_encrypt		ce_aes_cbc_encrypt
@@ -30,6 +34,8 @@ MODULE_DESCRIPTION("AES-ECB/CBC/CTR/XTS using ARMv8 Crypto Extensions");
 #else
 #define MODE			"neon"
 #define PRIO			200
+#define aes_setkey		crypto_aes_set_key
+#define aes_expandkey		crypto_aes_expand_key
 #define aes_ecb_encrypt		neon_aes_ecb_encrypt
 #define aes_ecb_decrypt		neon_aes_ecb_decrypt
 #define aes_cbc_encrypt		neon_aes_cbc_encrypt
@@ -79,10 +85,10 @@ static int xts_set_key(struct crypto_tfm *tfm, const u8 *in_key,
 	struct crypto_aes_xts_ctx *ctx = crypto_tfm_ctx(tfm);
 	int ret;
 
-	ret = crypto_aes_expand_key(&ctx->key1, in_key, key_len / 2);
+	ret = aes_expandkey(&ctx->key1, in_key, key_len / 2);
 	if (!ret)
-		ret = crypto_aes_expand_key(&ctx->key2, &in_key[key_len / 2],
-					    key_len / 2);
+		ret = aes_expandkey(&ctx->key2, &in_key[key_len / 2],
+				    key_len / 2);
 	if (!ret)
 		return 0;
 
@@ -288,7 +294,7 @@ static struct crypto_alg aes_algs[] = { {
 		.min_keysize	= AES_MIN_KEY_SIZE,
 		.max_keysize	= AES_MAX_KEY_SIZE,
 		.ivsize		= AES_BLOCK_SIZE,
-		.setkey		= crypto_aes_set_key,
+		.setkey		= aes_setkey,
 		.encrypt	= ecb_encrypt,
 		.decrypt	= ecb_decrypt,
 	},
@@ -306,7 +312,7 @@ static struct crypto_alg aes_algs[] = { {
 		.min_keysize	= AES_MIN_KEY_SIZE,
 		.max_keysize	= AES_MAX_KEY_SIZE,
 		.ivsize		= AES_BLOCK_SIZE,
-		.setkey		= crypto_aes_set_key,
+		.setkey		= aes_setkey,
 		.encrypt	= cbc_encrypt,
 		.decrypt	= cbc_decrypt,
 	},
@@ -324,7 +330,7 @@ static struct crypto_alg aes_algs[] = { {
 		.min_keysize	= AES_MIN_KEY_SIZE,
 		.max_keysize	= AES_MAX_KEY_SIZE,
 		.ivsize		= AES_BLOCK_SIZE,
-		.setkey		= crypto_aes_set_key,
+		.setkey		= aes_setkey,
 		.encrypt	= ctr_encrypt,
 		.decrypt	= ctr_encrypt,
 	},