crypto: blowfish - use unaligned accessors instead of alignmask

Instead of using an alignmask of 0x3 to ensure 32-bit alignment of
the Blowfish input and output blocks, which propagates to mode drivers,
and results in pointless copying on architectures that don't care about
alignment, use the unaligned accessors, which will do the right thing on
each respective architecture, avoiding the need for double buffering.
Signed-off-by: NArd Biesheuvel <ardb@kernel.org>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

crypto/blowfish_generic.c

@@ -14,7 +14,7 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/mm.h>
-#include <asm/byteorder.h>
+#include <asm/unaligned.h>
 #include <linux/crypto.h>
 #include <linux/types.h>
 #include <crypto/blowfish.h>
@@ -36,12 +36,10 @@
 static void bf_encrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src)
 {
 	struct bf_ctx *ctx = crypto_tfm_ctx(tfm);
-	const __be32 *in_blk = (const __be32 *)src;
-	__be32 *const out_blk = (__be32 *)dst;
 	const u32 *P = ctx->p;
 	const u32 *S = ctx->s;
-	u32 yl = be32_to_cpu(in_blk[0]);
-	u32 yr = be32_to_cpu(in_blk[1]);
+	u32 yl = get_unaligned_be32(src);
+	u32 yr = get_unaligned_be32(src + 4);
 
 	ROUND(yr, yl, 0);
 	ROUND(yl, yr, 1);
@@ -63,19 +61,17 @@ static void bf_encrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src)
 	yl ^= P[16];
 	yr ^= P[17];
 
-	out_blk[0] = cpu_to_be32(yr);
-	out_blk[1] = cpu_to_be32(yl);
+	put_unaligned_be32(yr, dst);
+	put_unaligned_be32(yl, dst + 4);
 }
 
 static void bf_decrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src)
 {
 	struct bf_ctx *ctx = crypto_tfm_ctx(tfm);
-	const __be32 *in_blk = (const __be32 *)src;
-	__be32 *const out_blk = (__be32 *)dst;
 	const u32 *P = ctx->p;
 	const u32 *S = ctx->s;
-	u32 yl = be32_to_cpu(in_blk[0]);
-	u32 yr = be32_to_cpu(in_blk[1]);
+	u32 yl = get_unaligned_be32(src);
+	u32 yr = get_unaligned_be32(src + 4);
 
 	ROUND(yr, yl, 17);
 	ROUND(yl, yr, 16);
@@ -97,8 +93,8 @@ static void bf_decrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src)
 	yl ^= P[1];
 	yr ^= P[0];
 
-	out_blk[0] = cpu_to_be32(yr);
-	out_blk[1] = cpu_to_be32(yl);
+	put_unaligned_be32(yr, dst);
+	put_unaligned_be32(yl, dst + 4);
 }
 
 static struct crypto_alg alg = {
@@ -108,7 +104,6 @@ static struct crypto_alg alg = {
 	.cra_flags		=	CRYPTO_ALG_TYPE_CIPHER,
 	.cra_blocksize		=	BF_BLOCK_SIZE,
 	.cra_ctxsize		=	sizeof(struct bf_ctx),
-	.cra_alignmask		=	3,
 	.cra_module		=	THIS_MODULE,
 	.cra_u			=	{ .cipher = {
 	.cia_min_keysize	=	BF_MIN_KEY_SIZE,