From: Eric Biggers <ebiggers@xxxxxxxxxx> Instead of specifying a nonzero alignmask, use the unaligned access helpers. This eliminates unnecessary alignment operations on most CPUs, which can handle unaligned accesses efficiently, and brings us a step closer to eventually removing support for the alignmask field. Signed-off-by: Eric Biggers <ebiggers@xxxxxxxxxx> --- crypto/aria_generic.c | 37 +++++++++++++++++-------------------- 1 file changed, 17 insertions(+), 20 deletions(-) diff --git a/crypto/aria_generic.c b/crypto/aria_generic.c index d96dfc4fdde67..bd359d3313c22 100644 --- a/crypto/aria_generic.c +++ b/crypto/aria_generic.c @@ -13,10 +13,11 @@ * * Public domain version is distributed above. */ #include <crypto/aria.h> +#include <linux/unaligned.h> static const u32 key_rc[20] = { 0x517cc1b7, 0x27220a94, 0xfe13abe8, 0xfa9a6ee0, 0x6db14acc, 0x9e21c820, 0xff28b1d5, 0xef5de2b0, 0xdb92371d, 0x2126e970, 0x03249775, 0x04e8c90e, @@ -25,36 +26,35 @@ static const u32 key_rc[20] = { }; static void aria_set_encrypt_key(struct aria_ctx *ctx, const u8 *in_key, unsigned int key_len) { - const __be32 *key = (const __be32 *)in_key; u32 w0[4], w1[4], w2[4], w3[4]; u32 reg0, reg1, reg2, reg3; const u32 *ck; int rkidx = 0; ck = &key_rc[(key_len - 16) / 2]; - w0[0] = be32_to_cpu(key[0]); - w0[1] = be32_to_cpu(key[1]); - w0[2] = be32_to_cpu(key[2]); - w0[3] = be32_to_cpu(key[3]); + w0[0] = get_unaligned_be32(&in_key[0]); + w0[1] = get_unaligned_be32(&in_key[4]); + w0[2] = get_unaligned_be32(&in_key[8]); + w0[3] = get_unaligned_be32(&in_key[12]); reg0 = w0[0] ^ ck[0]; reg1 = w0[1] ^ ck[1]; reg2 = w0[2] ^ ck[2]; reg3 = w0[3] ^ ck[3]; aria_subst_diff_odd(®0, ®1, ®2, ®3); if (key_len > 16) { - w1[0] = be32_to_cpu(key[4]); - w1[1] = be32_to_cpu(key[5]); + w1[0] = get_unaligned_be32(&in_key[16]); + w1[1] = get_unaligned_be32(&in_key[20]); if (key_len > 24) { - w1[2] = be32_to_cpu(key[6]); - w1[3] = be32_to_cpu(key[7]); + w1[2] = get_unaligned_be32(&in_key[24]); + w1[3] = get_unaligned_be32(&in_key[28]); } else { w1[2] = 0; w1[3] = 0; } } else { @@ -193,21 +193,19 @@ int aria_set_key(struct crypto_tfm *tfm, const u8 *in_key, unsigned int key_len) EXPORT_SYMBOL_GPL(aria_set_key); static void __aria_crypt(struct aria_ctx *ctx, u8 *out, const u8 *in, u32 key[][ARIA_RD_KEY_WORDS]) { - const __be32 *src = (const __be32 *)in; - __be32 *dst = (__be32 *)out; u32 reg0, reg1, reg2, reg3; int rounds, rkidx = 0; rounds = ctx->rounds; - reg0 = be32_to_cpu(src[0]); - reg1 = be32_to_cpu(src[1]); - reg2 = be32_to_cpu(src[2]); - reg3 = be32_to_cpu(src[3]); + reg0 = get_unaligned_be32(&in[0]); + reg1 = get_unaligned_be32(&in[4]); + reg2 = get_unaligned_be32(&in[8]); + reg3 = get_unaligned_be32(&in[12]); aria_add_round_key(key[rkidx], ®0, ®1, ®2, ®3); rkidx++; aria_subst_diff_odd(®0, ®1, ®2, ®3); @@ -239,14 +237,14 @@ static void __aria_crypt(struct aria_ctx *ctx, u8 *out, const u8 *in, reg3 = key[rkidx][3] ^ make_u32((u8)(x1[get_u8(reg3, 0)]), (u8)(x2[get_u8(reg3, 1)] >> 8), (u8)(s1[get_u8(reg3, 2)]), (u8)(s2[get_u8(reg3, 3)])); - dst[0] = cpu_to_be32(reg0); - dst[1] = cpu_to_be32(reg1); - dst[2] = cpu_to_be32(reg2); - dst[3] = cpu_to_be32(reg3); + put_unaligned_be32(reg0, &out[0]); + put_unaligned_be32(reg1, &out[4]); + put_unaligned_be32(reg2, &out[8]); + put_unaligned_be32(reg3, &out[12]); } void aria_encrypt(void *_ctx, u8 *out, const u8 *in) { struct aria_ctx *ctx = (struct aria_ctx *)_ctx; @@ -282,11 +280,10 @@ static struct crypto_alg aria_alg = { .cra_driver_name = "aria-generic", .cra_priority = 100, .cra_flags = CRYPTO_ALG_TYPE_CIPHER, .cra_blocksize = ARIA_BLOCK_SIZE, .cra_ctxsize = sizeof(struct aria_ctx), - .cra_alignmask = 3, .cra_module = THIS_MODULE, .cra_u = { .cipher = { .cia_min_keysize = ARIA_MIN_KEY_SIZE, .cia_max_keysize = ARIA_MAX_KEY_SIZE, -- 2.47.1