crypto: arm/aes - use native endiannes for key schedule

Align ARM's hw instruction based AES implementation with other versions
that keep the key schedule in native endianness. This will allow us to
merge the various implementations going forward.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

arch/arm/crypto/aes-ce-core.S

@@ -88,19 +88,19 @@
 
 	.macro		do_block, dround, fround
 	cmp		r3, #12			@ which key size?
-	vld1.8		{q10-q11}, [ip]!
+	vld1.32		{q10-q11}, [ip]!
 	\dround		q8, q9
-	vld1.8		{q12-q13}, [ip]!
+	vld1.32		{q12-q13}, [ip]!
 	\dround		q10, q11
-	vld1.8		{q10-q11}, [ip]!
+	vld1.32		{q10-q11}, [ip]!
 	\dround		q12, q13
-	vld1.8		{q12-q13}, [ip]!
+	vld1.32		{q12-q13}, [ip]!
 	\dround		q10, q11
 	blo		0f			@ AES-128: 10 rounds
-	vld1.8		{q10-q11}, [ip]!
+	vld1.32		{q10-q11}, [ip]!
 	\dround		q12, q13
 	beq		1f			@ AES-192: 12 rounds
-	vld1.8		{q12-q13}, [ip]
+	vld1.32		{q12-q13}, [ip]
 	\dround		q10, q11
 0:	\fround		q12, q13, q14
 	bx		lr
@@ -149,8 +149,8 @@ ENDPROC(aes_decrypt_3x)
 
 	.macro		prepare_key, rk, rounds
 	add		ip, \rk, \rounds, lsl #4
-	vld1.8		{q8-q9}, [\rk]		@ load first 2 round keys
-	vld1.8		{q14}, [ip]		@ load last round key
+	vld1.32		{q8-q9}, [\rk]		@ load first 2 round keys
+	vld1.32		{q14}, [ip]		@ load last round key
 	.endm
 
 	/*
@@ -505,8 +505,8 @@ ENDPROC(ce_aes_sub)
 	 *                                        operation on round key *src
 	 */
 ENTRY(ce_aes_invert)
-	vld1.8		{q0}, [r1]
+	vld1.32		{q0}, [r1]
 	aesimc.8	q0, q0
-	vst1.8		{q0}, [r0]
+	vst1.32		{q0}, [r0]
 	bx		lr
 ENDPROC(ce_aes_invert)

arch/arm/crypto/aes-ce-glue.c

@@ -7,6 +7,7 @@
 
 #include <asm/hwcap.h>
 #include <asm/neon.h>
+#include <asm/unaligned.h>
 #include <crypto/aes.h>
 #include <crypto/internal/simd.h>
 #include <crypto/internal/skcipher.h>
@@ -77,21 +78,17 @@ static int ce_aes_expandkey(struct crypto_aes_ctx *ctx, const u8 *in_key,
 	    key_len != AES_KEYSIZE_256)
 		return -EINVAL;
 
-	memcpy(ctx->key_enc, in_key, key_len);
 	ctx->key_length = key_len;
+	for (i = 0; i < kwords; i++)
+		ctx->key_enc[i] = get_unaligned_le32(in_key + i * sizeof(u32));
 
 	kernel_neon_begin();
 	for (i = 0; i < sizeof(rcon); i++) {
 		u32 *rki = ctx->key_enc + (i * kwords);
 		u32 *rko = rki + kwords;
 
-#ifndef CONFIG_CPU_BIG_ENDIAN
 		rko[0] = ror32(ce_aes_sub(rki[kwords - 1]), 8);
 		rko[0] = rko[0] ^ rki[0] ^ rcon[i];
-#else
-		rko[0] = rol32(ce_aes_sub(rki[kwords - 1]), 8);
-		rko[0] = rko[0] ^ rki[0] ^ (rcon[i] << 24);
-#endif
 		rko[1] = rko[0] ^ rki[1];
 		rko[2] = rko[1] ^ rki[2];
 		rko[3] = rko[2] ^ rki[3];