Hello,
This series adds Speck support to the crypto API, including the Speck128
and Speck64 variants. Speck is a lightweight block cipher that can be
much faster than AES on processors that don't have AES instructions.
We are planning to offer Speck-XTS (probably Speck128/256-XTS) as an
option for dm-crypt and fscrypt on Android, for low-end mobile devices
with older CPUs such as ARMv7 which don't have the Cryptography
Extensions. Currently, such devices are unencrypted because AES is not
fast enough, even when the NEON bit-sliced implementation of AES is
used. Other AES alternatives such as Twofish, Threefish, Camellia,
CAST6, and Serpent aren't fast enough either; it seems that only a
modern ARX cipher can provide sufficient performance on these devices.
This is a replacement for our original proposal
(https://patchwork.kernel.org/patch/10101451/) which was to offer
ChaCha20 for these devices. However, the use of a stream cipher for
disk/file encryption with no space to store nonces would have been much
more insecure than we thought initially, given that it would be used on
top of flash storage as well as potentially on top of F2FS, neither of
which is guaranteed to overwrite data in-place.
Speck has been somewhat controversial due to its origin. Nevertheless,
it has a straightforward design (it's an ARX cipher), and it appears to
be the leading software-optimized lightweight block cipher currently,
with the most cryptanalysis. It's also easy to implement without side
channels, unlike AES. Moreover, we only intend Speck to be used when
the status quo is no encryption, due to AES not being fast enough.
We've also considered a novel length-preserving encryption mode based on
ChaCha20 and Poly1305. While theoretically attractive, such a mode
would be a brand new crypto construction and would be more complicated
and difficult to implement efficiently in comparison to Speck-XTS.
Thus, patch 1 adds a generic implementation of Speck, and the following
patches add a 32-bit ARM NEON implementation of Speck-XTS. The
NEON-accelerated implementation is much faster than the generic
implementation and therefore is the implementation that would primarily
be used in practice on the devices we are targeting.
There is no AArch64 implementation included, since most such CPUs have
the Cryptography Extensions, allowing the use of AES. An AArch64
implementation can be added later if there is interest though.
Changed since v2:
- Fix __speck64_xts_crypt() to work on big endian CPUs.
Changed since v1:
- Use the word order recommended by the Speck authors. All test
vectors were updated.
Eric Biggers (5):
crypto: add support for the Speck block cipher
crypto: speck - export common helpers
crypto: arm/speck - add NEON-accelerated implementation of Speck-XTS
crypto: speck - add test vectors for Speck128-XTS
crypto: speck - add test vectors for Speck64-XTS
arch/arm/crypto/Kconfig | 6 +
arch/arm/crypto/Makefile | 2 +
arch/arm/crypto/speck-neon-core.S | 432 +++++++++
arch/arm/crypto/speck-neon-glue.c | 288 ++++++
crypto/Kconfig | 14 +
crypto/Makefile | 1 +
crypto/speck.c | 307 ++++++
crypto/testmgr.c | 36 +
crypto/testmgr.h | 1486 +++++++++++++++++++++++++++++
include/crypto/speck.h | 62 ++
10 files changed, 2634 insertions(+)
create mode 100644 arch/arm/crypto/speck-neon-core.S
create mode 100644 arch/arm/crypto/speck-neon-glue.c
create mode 100644 crypto/speck.c
create mode 100644 include/crypto/speck.h
--
2.16.1.291.g4437f3f132-goog
Export the Speck constants and transform context and the ->setkey(),
->encrypt(), and ->decrypt() functions so that they can be reused by the
ARM NEON implementation of Speck-XTS. The generic key expansion code
will be reused because it is not performance-critical and is not
vectorizable, while the generic encryption and decryption functions are
needed as fallbacks and for the XTS tweak encryption.
Signed-off-by: Eric Biggers <[email protected]>
---
crypto/speck.c | 90 +++++++++++++++++++++++-------------------
include/crypto/speck.h | 62 +++++++++++++++++++++++++++++
2 files changed, 111 insertions(+), 41 deletions(-)
create mode 100644 include/crypto/speck.h
diff --git a/crypto/speck.c b/crypto/speck.c
index 4e80ad76bcd7..58aa9f7f91f7 100644
--- a/crypto/speck.c
+++ b/crypto/speck.c
@@ -24,6 +24,7 @@
*/
#include <asm/unaligned.h>
+#include <crypto/speck.h>
#include <linux/bitops.h>
#include <linux/crypto.h>
#include <linux/init.h>
@@ -31,22 +32,6 @@
/* Speck128 */
-#define SPECK128_BLOCK_SIZE 16
-
-#define SPECK128_128_KEY_SIZE 16
-#define SPECK128_128_NROUNDS 32
-
-#define SPECK128_192_KEY_SIZE 24
-#define SPECK128_192_NROUNDS 33
-
-#define SPECK128_256_KEY_SIZE 32
-#define SPECK128_256_NROUNDS 34
-
-struct speck128_tfm_ctx {
- u64 round_keys[SPECK128_256_NROUNDS];
- int nrounds;
-};
-
static __always_inline void speck128_round(u64 *x, u64 *y, u64 k)
{
*x = ror64(*x, 8);
@@ -65,9 +50,9 @@ static __always_inline void speck128_unround(u64 *x, u64 *y, u64 k)
*x = rol64(*x, 8);
}
-static void speck128_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
+void crypto_speck128_encrypt(const struct speck128_tfm_ctx *ctx,
+ u8 *out, const u8 *in)
{
- const struct speck128_tfm_ctx *ctx = crypto_tfm_ctx(tfm);
u64 y = get_unaligned_le64(in);
u64 x = get_unaligned_le64(in + 8);
int i;
@@ -78,10 +63,16 @@ static void speck128_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
put_unaligned_le64(y, out);
put_unaligned_le64(x, out + 8);
}
+EXPORT_SYMBOL_GPL(crypto_speck128_encrypt);
-static void speck128_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
+static void speck128_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
+{
+ crypto_speck128_encrypt(crypto_tfm_ctx(tfm), out, in);
+}
+
+void crypto_speck128_decrypt(const struct speck128_tfm_ctx *ctx,
+ u8 *out, const u8 *in)
{
- const struct speck128_tfm_ctx *ctx = crypto_tfm_ctx(tfm);
u64 y = get_unaligned_le64(in);
u64 x = get_unaligned_le64(in + 8);
int i;
@@ -92,11 +83,16 @@ static void speck128_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
put_unaligned_le64(y, out);
put_unaligned_le64(x, out + 8);
}
+EXPORT_SYMBOL_GPL(crypto_speck128_decrypt);
-static int speck128_setkey(struct crypto_tfm *tfm, const u8 *key,
+static void speck128_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
+{
+ crypto_speck128_decrypt(crypto_tfm_ctx(tfm), out, in);
+}
+
+int crypto_speck128_setkey(struct speck128_tfm_ctx *ctx, const u8 *key,
unsigned int keylen)
{
- struct speck128_tfm_ctx *ctx = crypto_tfm_ctx(tfm);
u64 l[3];
u64 k;
int i;
@@ -138,21 +134,15 @@ static int speck128_setkey(struct crypto_tfm *tfm, const u8 *key,
return 0;
}
+EXPORT_SYMBOL_GPL(crypto_speck128_setkey);
-/* Speck64 */
-
-#define SPECK64_BLOCK_SIZE 8
-
-#define SPECK64_96_KEY_SIZE 12
-#define SPECK64_96_NROUNDS 26
-
-#define SPECK64_128_KEY_SIZE 16
-#define SPECK64_128_NROUNDS 27
+static int speck128_setkey(struct crypto_tfm *tfm, const u8 *key,
+ unsigned int keylen)
+{
+ return crypto_speck128_setkey(crypto_tfm_ctx(tfm), key, keylen);
+}
-struct speck64_tfm_ctx {
- u32 round_keys[SPECK64_128_NROUNDS];
- int nrounds;
-};
+/* Speck64 */
static __always_inline void speck64_round(u32 *x, u32 *y, u32 k)
{
@@ -172,9 +162,9 @@ static __always_inline void speck64_unround(u32 *x, u32 *y, u32 k)
*x = rol32(*x, 8);
}
-static void speck64_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
+void crypto_speck64_encrypt(const struct speck64_tfm_ctx *ctx,
+ u8 *out, const u8 *in)
{
- const struct speck64_tfm_ctx *ctx = crypto_tfm_ctx(tfm);
u32 y = get_unaligned_le32(in);
u32 x = get_unaligned_le32(in + 4);
int i;
@@ -185,10 +175,16 @@ static void speck64_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
put_unaligned_le32(y, out);
put_unaligned_le32(x, out + 4);
}
+EXPORT_SYMBOL_GPL(crypto_speck64_encrypt);
-static void speck64_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
+static void speck64_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
+{
+ crypto_speck64_encrypt(crypto_tfm_ctx(tfm), out, in);
+}
+
+void crypto_speck64_decrypt(const struct speck64_tfm_ctx *ctx,
+ u8 *out, const u8 *in)
{
- const struct speck64_tfm_ctx *ctx = crypto_tfm_ctx(tfm);
u32 y = get_unaligned_le32(in);
u32 x = get_unaligned_le32(in + 4);
int i;
@@ -199,11 +195,16 @@ static void speck64_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
put_unaligned_le32(y, out);
put_unaligned_le32(x, out + 4);
}
+EXPORT_SYMBOL_GPL(crypto_speck64_decrypt);
-static int speck64_setkey(struct crypto_tfm *tfm, const u8 *key,
+static void speck64_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
+{
+ crypto_speck64_decrypt(crypto_tfm_ctx(tfm), out, in);
+}
+
+int crypto_speck64_setkey(struct speck64_tfm_ctx *ctx, const u8 *key,
unsigned int keylen)
{
- struct speck64_tfm_ctx *ctx = crypto_tfm_ctx(tfm);
u32 l[3];
u32 k;
int i;
@@ -236,6 +237,13 @@ static int speck64_setkey(struct crypto_tfm *tfm, const u8 *key,
return 0;
}
+EXPORT_SYMBOL_GPL(crypto_speck64_setkey);
+
+static int speck64_setkey(struct crypto_tfm *tfm, const u8 *key,
+ unsigned int keylen)
+{
+ return crypto_speck64_setkey(crypto_tfm_ctx(tfm), key, keylen);
+}
/* Algorithm definitions */
diff --git a/include/crypto/speck.h b/include/crypto/speck.h
new file mode 100644
index 000000000000..73cfc952d405
--- /dev/null
+++ b/include/crypto/speck.h
@@ -0,0 +1,62 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Common values for the Speck algorithm
+ */
+
+#ifndef _CRYPTO_SPECK_H
+#define _CRYPTO_SPECK_H
+
+#include <linux/types.h>
+
+/* Speck128 */
+
+#define SPECK128_BLOCK_SIZE 16
+
+#define SPECK128_128_KEY_SIZE 16
+#define SPECK128_128_NROUNDS 32
+
+#define SPECK128_192_KEY_SIZE 24
+#define SPECK128_192_NROUNDS 33
+
+#define SPECK128_256_KEY_SIZE 32
+#define SPECK128_256_NROUNDS 34
+
+struct speck128_tfm_ctx {
+ u64 round_keys[SPECK128_256_NROUNDS];
+ int nrounds;
+};
+
+void crypto_speck128_encrypt(const struct speck128_tfm_ctx *ctx,
+ u8 *out, const u8 *in);
+
+void crypto_speck128_decrypt(const struct speck128_tfm_ctx *ctx,
+ u8 *out, const u8 *in);
+
+int crypto_speck128_setkey(struct speck128_tfm_ctx *ctx, const u8 *key,
+ unsigned int keysize);
+
+/* Speck64 */
+
+#define SPECK64_BLOCK_SIZE 8
+
+#define SPECK64_96_KEY_SIZE 12
+#define SPECK64_96_NROUNDS 26
+
+#define SPECK64_128_KEY_SIZE 16
+#define SPECK64_128_NROUNDS 27
+
+struct speck64_tfm_ctx {
+ u32 round_keys[SPECK64_128_NROUNDS];
+ int nrounds;
+};
+
+void crypto_speck64_encrypt(const struct speck64_tfm_ctx *ctx,
+ u8 *out, const u8 *in);
+
+void crypto_speck64_decrypt(const struct speck64_tfm_ctx *ctx,
+ u8 *out, const u8 *in);
+
+int crypto_speck64_setkey(struct speck64_tfm_ctx *ctx, const u8 *key,
+ unsigned int keysize);
+
+#endif /* _CRYPTO_SPECK_H */
--
2.16.1.291.g4437f3f132-goog
Add test vectors for Speck64-XTS, generated in userspace using C code.
The inputs were borrowed from the AES-XTS test vectors, with key lengths
adjusted.
xts-speck64-neon passes these tests. However, they aren't currently
applicable for the generic XTS template, as that only supports a 128-bit
block size.
Signed-off-by: Eric Biggers <[email protected]>
---
crypto/testmgr.c | 9 +
crypto/testmgr.h | 671 +++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 680 insertions(+)
diff --git a/crypto/testmgr.c b/crypto/testmgr.c
index e011a347d51b..9f82e7bc9c56 100644
--- a/crypto/testmgr.c
+++ b/crypto/testmgr.c
@@ -3584,6 +3584,15 @@ static const struct alg_test_desc alg_test_descs[] = {
.dec = __VECS(speck128_xts_dec_tv_template)
}
}
+ }, {
+ .alg = "xts(speck64)",
+ .test = alg_test_skcipher,
+ .suite = {
+ .cipher = {
+ .enc = __VECS(speck64_xts_enc_tv_template),
+ .dec = __VECS(speck64_xts_dec_tv_template)
+ }
+ }
}, {
.alg = "xts(twofish)",
.test = alg_test_skcipher,
diff --git a/crypto/testmgr.h b/crypto/testmgr.h
index 0212e0ebcd0c..da72fd394f35 100644
--- a/crypto/testmgr.h
+++ b/crypto/testmgr.h
@@ -15138,6 +15138,677 @@ static const struct cipher_testvec speck64_dec_tv_template[] = {
},
};
+/*
+ * Speck64-XTS test vectors, taken from the AES-XTS test vectors with the result
+ * recomputed with Speck64 as the cipher, and key lengths adjusted
+ */
+
+static const struct cipher_testvec speck64_xts_enc_tv_template[] = {
+ {
+ .key = "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .klen = 24,
+ .iv = "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .ilen = 32,
+ .result = "\x84\xaf\x54\x07\x19\xd4\x7c\xa6"
+ "\xe4\xfe\xdf\xc4\x1f\x34\xc3\xc2"
+ "\x80\xf5\x72\xe7\xcd\xf0\x99\x22"
+ "\x35\xa7\x2f\x06\xef\xdc\x51\xaa",
+ .rlen = 32,
+ }, {
+ .key = "\x11\x11\x11\x11\x11\x11\x11\x11"
+ "\x11\x11\x11\x11\x11\x11\x11\x11"
+ "\x22\x22\x22\x22\x22\x22\x22\x22",
+ .klen = 24,
+ .iv = "\x33\x33\x33\x33\x33\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44",
+ .ilen = 32,
+ .result = "\x12\x56\x73\xcd\x15\x87\xa8\x59"
+ "\xcf\x84\xae\xd9\x1c\x66\xd6\x9f"
+ "\xb3\x12\x69\x7e\x36\xeb\x52\xff"
+ "\x62\xdd\xba\x90\xb3\xe1\xee\x99",
+ .rlen = 32,
+ }, {
+ .key = "\xff\xfe\xfd\xfc\xfb\xfa\xf9\xf8"
+ "\xf7\xf6\xf5\xf4\xf3\xf2\xf1\xf0"
+ "\x22\x22\x22\x22\x22\x22\x22\x22",
+ .klen = 24,
+ .iv = "\x33\x33\x33\x33\x33\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44",
+ .ilen = 32,
+ .result = "\x15\x1b\xe4\x2c\xa2\x5a\x2d\x2c"
+ "\x27\x36\xc0\xbf\x5d\xea\x36\x37"
+ "\x2d\x1a\x88\xbc\x66\xb5\xd0\x0b"
+ "\xa1\xbc\x19\xb2\x0f\x3b\x75\x34",
+ .rlen = 32,
+ }, {
+ .key = "\x27\x18\x28\x18\x28\x45\x90\x45"
+ "\x23\x53\x60\x28\x74\x71\x35\x26"
+ "\x31\x41\x59\x26\x53\x58\x97\x93",
+ .klen = 24,
+ .iv = "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17"
+ "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f"
+ "\x20\x21\x22\x23\x24\x25\x26\x27"
+ "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f"
+ "\x30\x31\x32\x33\x34\x35\x36\x37"
+ "\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f"
+ "\x40\x41\x42\x43\x44\x45\x46\x47"
+ "\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f"
+ "\x50\x51\x52\x53\x54\x55\x56\x57"
+ "\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f"
+ "\x60\x61\x62\x63\x64\x65\x66\x67"
+ "\x68\x69\x6a\x6b\x6c\x6d\x6e\x6f"
+ "\x70\x71\x72\x73\x74\x75\x76\x77"
+ "\x78\x79\x7a\x7b\x7c\x7d\x7e\x7f"
+ "\x80\x81\x82\x83\x84\x85\x86\x87"
+ "\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
+ "\x90\x91\x92\x93\x94\x95\x96\x97"
+ "\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
+ "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7"
+ "\xa8\xa9\xaa\xab\xac\xad\xae\xaf"
+ "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7"
+ "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf"
+ "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7"
+ "\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
+ "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7"
+ "\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf"
+ "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7"
+ "\xe8\xe9\xea\xeb\xec\xed\xee\xef"
+ "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7"
+ "\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff"
+ "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17"
+ "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f"
+ "\x20\x21\x22\x23\x24\x25\x26\x27"
+ "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f"
+ "\x30\x31\x32\x33\x34\x35\x36\x37"
+ "\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f"
+ "\x40\x41\x42\x43\x44\x45\x46\x47"
+ "\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f"
+ "\x50\x51\x52\x53\x54\x55\x56\x57"
+ "\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f"
+ "\x60\x61\x62\x63\x64\x65\x66\x67"
+ "\x68\x69\x6a\x6b\x6c\x6d\x6e\x6f"
+ "\x70\x71\x72\x73\x74\x75\x76\x77"
+ "\x78\x79\x7a\x7b\x7c\x7d\x7e\x7f"
+ "\x80\x81\x82\x83\x84\x85\x86\x87"
+ "\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
+ "\x90\x91\x92\x93\x94\x95\x96\x97"
+ "\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
+ "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7"
+ "\xa8\xa9\xaa\xab\xac\xad\xae\xaf"
+ "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7"
+ "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf"
+ "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7"
+ "\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
+ "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7"
+ "\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf"
+ "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7"
+ "\xe8\xe9\xea\xeb\xec\xed\xee\xef"
+ "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7"
+ "\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff",
+ .ilen = 512,
+ .result = "\xaf\xa1\x81\xa6\x32\xbb\x15\x8e"
+ "\xf8\x95\x2e\xd3\xe6\xee\x7e\x09"
+ "\x0c\x1a\xf5\x02\x97\x8b\xe3\xb3"
+ "\x11\xc7\x39\x96\xd0\x95\xf4\x56"
+ "\xf4\xdd\x03\x38\x01\x44\x2c\xcf"
+ "\x88\xae\x8e\x3c\xcd\xe7\xaa\x66"
+ "\xfe\x3d\xc6\xfb\x01\x23\x51\x43"
+ "\xd5\xd2\x13\x86\x94\x34\xe9\x62"
+ "\xf9\x89\xe3\xd1\x7b\xbe\xf8\xef"
+ "\x76\x35\x04\x3f\xdb\x23\x9d\x0b"
+ "\x85\x42\xb9\x02\xd6\xcc\xdb\x96"
+ "\xa7\x6b\x27\xb6\xd4\x45\x8f\x7d"
+ "\xae\xd2\x04\xd5\xda\xc1\x7e\x24"
+ "\x8c\x73\xbe\x48\x7e\xcf\x65\x28"
+ "\x29\xe5\xbe\x54\x30\xcb\x46\x95"
+ "\x4f\x2e\x8a\x36\xc8\x27\xc5\xbe"
+ "\xd0\x1a\xaf\xab\x26\xcd\x9e\x69"
+ "\xa1\x09\x95\x71\x26\xe9\xc4\xdf"
+ "\xe6\x31\xc3\x46\xda\xaf\x0b\x41"
+ "\x1f\xab\xb1\x8e\xd6\xfc\x0b\xb3"
+ "\x82\xc0\x37\x27\xfc\x91\xa7\x05"
+ "\xfb\xc5\xdc\x2b\x74\x96\x48\x43"
+ "\x5d\x9c\x19\x0f\x60\x63\x3a\x1f"
+ "\x6f\xf0\x03\xbe\x4d\xfd\xc8\x4a"
+ "\xc6\xa4\x81\x6d\xc3\x12\x2a\x5c"
+ "\x07\xff\xf3\x72\x74\x48\xb5\x40"
+ "\x50\xb5\xdd\x90\x43\x31\x18\x15"
+ "\x7b\xf2\xa6\xdb\x83\xc8\x4b\x4a"
+ "\x29\x93\x90\x8b\xda\x07\xf0\x35"
+ "\x6d\x90\x88\x09\x4e\x83\xf5\x5b"
+ "\x94\x12\xbb\x33\x27\x1d\x3f\x23"
+ "\x51\xa8\x7c\x07\xa2\xae\x77\xa6"
+ "\x50\xfd\xcc\xc0\x4f\x80\x7a\x9f"
+ "\x66\xdd\xcd\x75\x24\x8b\x33\xf7"
+ "\x20\xdb\x83\x9b\x4f\x11\x63\x6e"
+ "\xcf\x37\xef\xc9\x11\x01\x5c\x45"
+ "\x32\x99\x7c\x3c\x9e\x42\x89\xe3"
+ "\x70\x6d\x15\x9f\xb1\xe6\xb6\x05"
+ "\xfe\x0c\xb9\x49\x2d\x90\x6d\xcc"
+ "\x5d\x3f\xc1\xfe\x89\x0a\x2e\x2d"
+ "\xa0\xa8\x89\x3b\x73\x39\xa5\x94"
+ "\x4c\xa4\xa6\xbb\xa7\x14\x46\x89"
+ "\x10\xff\xaf\xef\xca\xdd\x4f\x80"
+ "\xb3\xdf\x3b\xab\xd4\xe5\x5a\xc7"
+ "\x33\xca\x00\x8b\x8b\x3f\xea\xec"
+ "\x68\x8a\xc2\x6d\xfd\xd4\x67\x0f"
+ "\x22\x31\xe1\x0e\xfe\x5a\x04\xd5"
+ "\x64\xa3\xf1\x1a\x76\x28\xcc\x35"
+ "\x36\xa7\x0a\x74\xf7\x1c\x44\x9b"
+ "\xc7\x1b\x53\x17\x02\xea\xd1\xad"
+ "\x13\x51\x73\xc0\xa0\xb2\x05\x32"
+ "\xa8\xa2\x37\x2e\xe1\x7a\x3a\x19"
+ "\x26\xb4\x6c\x62\x5d\xb3\x1a\x1d"
+ "\x59\xda\xee\x1a\x22\x18\xda\x0d"
+ "\x88\x0f\x55\x8b\x72\x62\xfd\xc1"
+ "\x69\x13\xcd\x0d\x5f\xc1\x09\x52"
+ "\xee\xd6\xe3\x84\x4d\xee\xf6\x88"
+ "\xaf\x83\xdc\x76\xf4\xc0\x93\x3f"
+ "\x4a\x75\x2f\xb0\x0b\x3e\xc4\x54"
+ "\x7d\x69\x8d\x00\x62\x77\x0d\x14"
+ "\xbe\x7c\xa6\x7d\xc5\x24\x4f\xf3"
+ "\x50\xf7\x5f\xf4\xc2\xca\x41\x97"
+ "\x37\xbe\x75\x74\xcd\xf0\x75\x6e"
+ "\x25\x23\x94\xbd\xda\x8d\xb0\xd4",
+ .rlen = 512,
+ }, {
+ .key = "\x27\x18\x28\x18\x28\x45\x90\x45"
+ "\x23\x53\x60\x28\x74\x71\x35\x26"
+ "\x62\x49\x77\x57\x24\x70\x93\x69"
+ "\x99\x59\x57\x49\x66\x96\x76\x27",
+ .klen = 32,
+ .iv = "\xff\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17"
+ "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f"
+ "\x20\x21\x22\x23\x24\x25\x26\x27"
+ "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f"
+ "\x30\x31\x32\x33\x34\x35\x36\x37"
+ "\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f"
+ "\x40\x41\x42\x43\x44\x45\x46\x47"
+ "\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f"
+ "\x50\x51\x52\x53\x54\x55\x56\x57"
+ "\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f"
+ "\x60\x61\x62\x63\x64\x65\x66\x67"
+ "\x68\x69\x6a\x6b\x6c\x6d\x6e\x6f"
+ "\x70\x71\x72\x73\x74\x75\x76\x77"
+ "\x78\x79\x7a\x7b\x7c\x7d\x7e\x7f"
+ "\x80\x81\x82\x83\x84\x85\x86\x87"
+ "\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
+ "\x90\x91\x92\x93\x94\x95\x96\x97"
+ "\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
+ "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7"
+ "\xa8\xa9\xaa\xab\xac\xad\xae\xaf"
+ "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7"
+ "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf"
+ "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7"
+ "\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
+ "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7"
+ "\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf"
+ "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7"
+ "\xe8\xe9\xea\xeb\xec\xed\xee\xef"
+ "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7"
+ "\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff"
+ "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17"
+ "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f"
+ "\x20\x21\x22\x23\x24\x25\x26\x27"
+ "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f"
+ "\x30\x31\x32\x33\x34\x35\x36\x37"
+ "\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f"
+ "\x40\x41\x42\x43\x44\x45\x46\x47"
+ "\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f"
+ "\x50\x51\x52\x53\x54\x55\x56\x57"
+ "\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f"
+ "\x60\x61\x62\x63\x64\x65\x66\x67"
+ "\x68\x69\x6a\x6b\x6c\x6d\x6e\x6f"
+ "\x70\x71\x72\x73\x74\x75\x76\x77"
+ "\x78\x79\x7a\x7b\x7c\x7d\x7e\x7f"
+ "\x80\x81\x82\x83\x84\x85\x86\x87"
+ "\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
+ "\x90\x91\x92\x93\x94\x95\x96\x97"
+ "\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
+ "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7"
+ "\xa8\xa9\xaa\xab\xac\xad\xae\xaf"
+ "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7"
+ "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf"
+ "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7"
+ "\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
+ "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7"
+ "\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf"
+ "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7"
+ "\xe8\xe9\xea\xeb\xec\xed\xee\xef"
+ "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7"
+ "\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff",
+ .ilen = 512,
+ .result = "\x55\xed\x71\xd3\x02\x8e\x15\x3b"
+ "\xc6\x71\x29\x2d\x3e\x89\x9f\x59"
+ "\x68\x6a\xcc\x8a\x56\x97\xf3\x95"
+ "\x4e\x51\x08\xda\x2a\xf8\x6f\x3c"
+ "\x78\x16\xea\x80\xdb\x33\x75\x94"
+ "\xf9\x29\xc4\x2b\x76\x75\x97\xc7"
+ "\xf2\x98\x2c\xf9\xff\xc8\xd5\x2b"
+ "\x18\xf1\xaf\xcf\x7c\xc5\x0b\xee"
+ "\xad\x3c\x76\x7c\xe6\x27\xa2\x2a"
+ "\xe4\x66\xe1\xab\xa2\x39\xfc\x7c"
+ "\xf5\xec\x32\x74\xa3\xb8\x03\x88"
+ "\x52\xfc\x2e\x56\x3f\xa1\xf0\x9f"
+ "\x84\x5e\x46\xed\x20\x89\xb6\x44"
+ "\x8d\xd0\xed\x54\x47\x16\xbe\x95"
+ "\x8a\xb3\x6b\x72\xc4\x32\x52\x13"
+ "\x1b\xb0\x82\xbe\xac\xf9\x70\xa6"
+ "\x44\x18\xdd\x8c\x6e\xca\x6e\x45"
+ "\x8f\x1e\x10\x07\x57\x25\x98\x7b"
+ "\x17\x8c\x78\xdd\x80\xa7\xd9\xd8"
+ "\x63\xaf\xb9\x67\x57\xfd\xbc\xdb"
+ "\x44\xe9\xc5\x65\xd1\xc7\x3b\xff"
+ "\x20\xa0\x80\x1a\xc3\x9a\xad\x5e"
+ "\x5d\x3b\xd3\x07\xd9\xf5\xfd\x3d"
+ "\x4a\x8b\xa8\xd2\x6e\x7a\x51\x65"
+ "\x6c\x8e\x95\xe0\x45\xc9\x5f\x4a"
+ "\x09\x3c\x3d\x71\x7f\x0c\x84\x2a"
+ "\xc8\x48\x52\x1a\xc2\xd5\xd6\x78"
+ "\x92\x1e\xa0\x90\x2e\xea\xf0\xf3"
+ "\xdc\x0f\xb1\xaf\x0d\x9b\x06\x2e"
+ "\x35\x10\x30\x82\x0d\xe7\xc5\x9b"
+ "\xde\x44\x18\xbd\x9f\xd1\x45\xa9"
+ "\x7b\x7a\x4a\xad\x35\x65\x27\xca"
+ "\xb2\xc3\xd4\x9b\x71\x86\x70\xee"
+ "\xf1\x89\x3b\x85\x4b\x5b\xaa\xaf"
+ "\xfc\x42\xc8\x31\x59\xbe\x16\x60"
+ "\x4f\xf9\xfa\x12\xea\xd0\xa7\x14"
+ "\xf0\x7a\xf3\xd5\x8d\xbd\x81\xef"
+ "\x52\x7f\x29\x51\x94\x20\x67\x3c"
+ "\xd1\xaf\x77\x9f\x22\x5a\x4e\x63"
+ "\xe7\xff\x73\x25\xd1\xdd\x96\x8a"
+ "\x98\x52\x6d\xf3\xac\x3e\xf2\x18"
+ "\x6d\xf6\x0a\x29\xa6\x34\x3d\xed"
+ "\xe3\x27\x0d\x9d\x0a\x02\x44\x7e"
+ "\x5a\x7e\x67\x0f\x0a\x9e\xd6\xad"
+ "\x91\xe6\x4d\x81\x8c\x5c\x59\xaa"
+ "\xfb\xeb\x56\x53\xd2\x7d\x4c\x81"
+ "\x65\x53\x0f\x41\x11\xbd\x98\x99"
+ "\xf9\xc6\xfa\x51\x2e\xa3\xdd\x8d"
+ "\x84\x98\xf9\x34\xed\x33\x2a\x1f"
+ "\x82\xed\xc1\x73\x98\xd3\x02\xdc"
+ "\xe6\xc2\x33\x1d\xa2\xb4\xca\x76"
+ "\x63\x51\x34\x9d\x96\x12\xae\xce"
+ "\x83\xc9\x76\x5e\xa4\x1b\x53\x37"
+ "\x17\xd5\xc0\x80\x1d\x62\xf8\x3d"
+ "\x54\x27\x74\xbb\x10\x86\x57\x46"
+ "\x68\xe1\xed\x14\xe7\x9d\xfc\x84"
+ "\x47\xbc\xc2\xf8\x19\x4b\x99\xcf"
+ "\x7a\xe9\xc4\xb8\x8c\x82\x72\x4d"
+ "\x7b\x4f\x38\x55\x36\x71\x64\xc1"
+ "\xfc\x5c\x75\x52\x33\x02\x18\xf8"
+ "\x17\xe1\x2b\xc2\x43\x39\xbd\x76"
+ "\x9b\x63\x76\x32\x2f\x19\x72\x10"
+ "\x9f\x21\x0c\xf1\x66\x50\x7f\xa5"
+ "\x0d\x1f\x46\xe0\xba\xd3\x2f\x3c",
+ .rlen = 512,
+ .also_non_np = 1,
+ .np = 3,
+ .tap = { 512 - 20, 4, 16 },
+ }
+};
+
+static const struct cipher_testvec speck64_xts_dec_tv_template[] = {
+ {
+ .key = "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .klen = 24,
+ .iv = "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\x84\xaf\x54\x07\x19\xd4\x7c\xa6"
+ "\xe4\xfe\xdf\xc4\x1f\x34\xc3\xc2"
+ "\x80\xf5\x72\xe7\xcd\xf0\x99\x22"
+ "\x35\xa7\x2f\x06\xef\xdc\x51\xaa",
+ .ilen = 32,
+ .result = "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .rlen = 32,
+ }, {
+ .key = "\x11\x11\x11\x11\x11\x11\x11\x11"
+ "\x11\x11\x11\x11\x11\x11\x11\x11"
+ "\x22\x22\x22\x22\x22\x22\x22\x22",
+ .klen = 24,
+ .iv = "\x33\x33\x33\x33\x33\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\x12\x56\x73\xcd\x15\x87\xa8\x59"
+ "\xcf\x84\xae\xd9\x1c\x66\xd6\x9f"
+ "\xb3\x12\x69\x7e\x36\xeb\x52\xff"
+ "\x62\xdd\xba\x90\xb3\xe1\xee\x99",
+ .ilen = 32,
+ .result = "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44",
+ .rlen = 32,
+ }, {
+ .key = "\xff\xfe\xfd\xfc\xfb\xfa\xf9\xf8"
+ "\xf7\xf6\xf5\xf4\xf3\xf2\xf1\xf0"
+ "\x22\x22\x22\x22\x22\x22\x22\x22",
+ .klen = 24,
+ .iv = "\x33\x33\x33\x33\x33\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\x15\x1b\xe4\x2c\xa2\x5a\x2d\x2c"
+ "\x27\x36\xc0\xbf\x5d\xea\x36\x37"
+ "\x2d\x1a\x88\xbc\x66\xb5\xd0\x0b"
+ "\xa1\xbc\x19\xb2\x0f\x3b\x75\x34",
+ .ilen = 32,
+ .result = "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44",
+ .rlen = 32,
+ }, {
+ .key = "\x27\x18\x28\x18\x28\x45\x90\x45"
+ "\x23\x53\x60\x28\x74\x71\x35\x26"
+ "\x31\x41\x59\x26\x53\x58\x97\x93",
+ .klen = 24,
+ .iv = "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\xaf\xa1\x81\xa6\x32\xbb\x15\x8e"
+ "\xf8\x95\x2e\xd3\xe6\xee\x7e\x09"
+ "\x0c\x1a\xf5\x02\x97\x8b\xe3\xb3"
+ "\x11\xc7\x39\x96\xd0\x95\xf4\x56"
+ "\xf4\xdd\x03\x38\x01\x44\x2c\xcf"
+ "\x88\xae\x8e\x3c\xcd\xe7\xaa\x66"
+ "\xfe\x3d\xc6\xfb\x01\x23\x51\x43"
+ "\xd5\xd2\x13\x86\x94\x34\xe9\x62"
+ "\xf9\x89\xe3\xd1\x7b\xbe\xf8\xef"
+ "\x76\x35\x04\x3f\xdb\x23\x9d\x0b"
+ "\x85\x42\xb9\x02\xd6\xcc\xdb\x96"
+ "\xa7\x6b\x27\xb6\xd4\x45\x8f\x7d"
+ "\xae\xd2\x04\xd5\xda\xc1\x7e\x24"
+ "\x8c\x73\xbe\x48\x7e\xcf\x65\x28"
+ "\x29\xe5\xbe\x54\x30\xcb\x46\x95"
+ "\x4f\x2e\x8a\x36\xc8\x27\xc5\xbe"
+ "\xd0\x1a\xaf\xab\x26\xcd\x9e\x69"
+ "\xa1\x09\x95\x71\x26\xe9\xc4\xdf"
+ "\xe6\x31\xc3\x46\xda\xaf\x0b\x41"
+ "\x1f\xab\xb1\x8e\xd6\xfc\x0b\xb3"
+ "\x82\xc0\x37\x27\xfc\x91\xa7\x05"
+ "\xfb\xc5\xdc\x2b\x74\x96\x48\x43"
+ "\x5d\x9c\x19\x0f\x60\x63\x3a\x1f"
+ "\x6f\xf0\x03\xbe\x4d\xfd\xc8\x4a"
+ "\xc6\xa4\x81\x6d\xc3\x12\x2a\x5c"
+ "\x07\xff\xf3\x72\x74\x48\xb5\x40"
+ "\x50\xb5\xdd\x90\x43\x31\x18\x15"
+ "\x7b\xf2\xa6\xdb\x83\xc8\x4b\x4a"
+ "\x29\x93\x90\x8b\xda\x07\xf0\x35"
+ "\x6d\x90\x88\x09\x4e\x83\xf5\x5b"
+ "\x94\x12\xbb\x33\x27\x1d\x3f\x23"
+ "\x51\xa8\x7c\x07\xa2\xae\x77\xa6"
+ "\x50\xfd\xcc\xc0\x4f\x80\x7a\x9f"
+ "\x66\xdd\xcd\x75\x24\x8b\x33\xf7"
+ "\x20\xdb\x83\x9b\x4f\x11\x63\x6e"
+ "\xcf\x37\xef\xc9\x11\x01\x5c\x45"
+ "\x32\x99\x7c\x3c\x9e\x42\x89\xe3"
+ "\x70\x6d\x15\x9f\xb1\xe6\xb6\x05"
+ "\xfe\x0c\xb9\x49\x2d\x90\x6d\xcc"
+ "\x5d\x3f\xc1\xfe\x89\x0a\x2e\x2d"
+ "\xa0\xa8\x89\x3b\x73\x39\xa5\x94"
+ "\x4c\xa4\xa6\xbb\xa7\x14\x46\x89"
+ "\x10\xff\xaf\xef\xca\xdd\x4f\x80"
+ "\xb3\xdf\x3b\xab\xd4\xe5\x5a\xc7"
+ "\x33\xca\x00\x8b\x8b\x3f\xea\xec"
+ "\x68\x8a\xc2\x6d\xfd\xd4\x67\x0f"
+ "\x22\x31\xe1\x0e\xfe\x5a\x04\xd5"
+ "\x64\xa3\xf1\x1a\x76\x28\xcc\x35"
+ "\x36\xa7\x0a\x74\xf7\x1c\x44\x9b"
+ "\xc7\x1b\x53\x17\x02\xea\xd1\xad"
+ "\x13\x51\x73\xc0\xa0\xb2\x05\x32"
+ "\xa8\xa2\x37\x2e\xe1\x7a\x3a\x19"
+ "\x26\xb4\x6c\x62\x5d\xb3\x1a\x1d"
+ "\x59\xda\xee\x1a\x22\x18\xda\x0d"
+ "\x88\x0f\x55\x8b\x72\x62\xfd\xc1"
+ "\x69\x13\xcd\x0d\x5f\xc1\x09\x52"
+ "\xee\xd6\xe3\x84\x4d\xee\xf6\x88"
+ "\xaf\x83\xdc\x76\xf4\xc0\x93\x3f"
+ "\x4a\x75\x2f\xb0\x0b\x3e\xc4\x54"
+ "\x7d\x69\x8d\x00\x62\x77\x0d\x14"
+ "\xbe\x7c\xa6\x7d\xc5\x24\x4f\xf3"
+ "\x50\xf7\x5f\xf4\xc2\xca\x41\x97"
+ "\x37\xbe\x75\x74\xcd\xf0\x75\x6e"
+ "\x25\x23\x94\xbd\xda\x8d\xb0\xd4",
+ .ilen = 512,
+ .result = "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17"
+ "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f"
+ "\x20\x21\x22\x23\x24\x25\x26\x27"
+ "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f"
+ "\x30\x31\x32\x33\x34\x35\x36\x37"
+ "\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f"
+ "\x40\x41\x42\x43\x44\x45\x46\x47"
+ "\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f"
+ "\x50\x51\x52\x53\x54\x55\x56\x57"
+ "\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f"
+ "\x60\x61\x62\x63\x64\x65\x66\x67"
+ "\x68\x69\x6a\x6b\x6c\x6d\x6e\x6f"
+ "\x70\x71\x72\x73\x74\x75\x76\x77"
+ "\x78\x79\x7a\x7b\x7c\x7d\x7e\x7f"
+ "\x80\x81\x82\x83\x84\x85\x86\x87"
+ "\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
+ "\x90\x91\x92\x93\x94\x95\x96\x97"
+ "\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
+ "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7"
+ "\xa8\xa9\xaa\xab\xac\xad\xae\xaf"
+ "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7"
+ "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf"
+ "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7"
+ "\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
+ "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7"
+ "\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf"
+ "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7"
+ "\xe8\xe9\xea\xeb\xec\xed\xee\xef"
+ "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7"
+ "\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff"
+ "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17"
+ "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f"
+ "\x20\x21\x22\x23\x24\x25\x26\x27"
+ "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f"
+ "\x30\x31\x32\x33\x34\x35\x36\x37"
+ "\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f"
+ "\x40\x41\x42\x43\x44\x45\x46\x47"
+ "\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f"
+ "\x50\x51\x52\x53\x54\x55\x56\x57"
+ "\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f"
+ "\x60\x61\x62\x63\x64\x65\x66\x67"
+ "\x68\x69\x6a\x6b\x6c\x6d\x6e\x6f"
+ "\x70\x71\x72\x73\x74\x75\x76\x77"
+ "\x78\x79\x7a\x7b\x7c\x7d\x7e\x7f"
+ "\x80\x81\x82\x83\x84\x85\x86\x87"
+ "\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
+ "\x90\x91\x92\x93\x94\x95\x96\x97"
+ "\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
+ "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7"
+ "\xa8\xa9\xaa\xab\xac\xad\xae\xaf"
+ "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7"
+ "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf"
+ "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7"
+ "\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
+ "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7"
+ "\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf"
+ "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7"
+ "\xe8\xe9\xea\xeb\xec\xed\xee\xef"
+ "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7"
+ "\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff",
+ .rlen = 512,
+ }, {
+ .key = "\x27\x18\x28\x18\x28\x45\x90\x45"
+ "\x23\x53\x60\x28\x74\x71\x35\x26"
+ "\x62\x49\x77\x57\x24\x70\x93\x69"
+ "\x99\x59\x57\x49\x66\x96\x76\x27",
+ .klen = 32,
+ .iv = "\xff\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\x55\xed\x71\xd3\x02\x8e\x15\x3b"
+ "\xc6\x71\x29\x2d\x3e\x89\x9f\x59"
+ "\x68\x6a\xcc\x8a\x56\x97\xf3\x95"
+ "\x4e\x51\x08\xda\x2a\xf8\x6f\x3c"
+ "\x78\x16\xea\x80\xdb\x33\x75\x94"
+ "\xf9\x29\xc4\x2b\x76\x75\x97\xc7"
+ "\xf2\x98\x2c\xf9\xff\xc8\xd5\x2b"
+ "\x18\xf1\xaf\xcf\x7c\xc5\x0b\xee"
+ "\xad\x3c\x76\x7c\xe6\x27\xa2\x2a"
+ "\xe4\x66\xe1\xab\xa2\x39\xfc\x7c"
+ "\xf5\xec\x32\x74\xa3\xb8\x03\x88"
+ "\x52\xfc\x2e\x56\x3f\xa1\xf0\x9f"
+ "\x84\x5e\x46\xed\x20\x89\xb6\x44"
+ "\x8d\xd0\xed\x54\x47\x16\xbe\x95"
+ "\x8a\xb3\x6b\x72\xc4\x32\x52\x13"
+ "\x1b\xb0\x82\xbe\xac\xf9\x70\xa6"
+ "\x44\x18\xdd\x8c\x6e\xca\x6e\x45"
+ "\x8f\x1e\x10\x07\x57\x25\x98\x7b"
+ "\x17\x8c\x78\xdd\x80\xa7\xd9\xd8"
+ "\x63\xaf\xb9\x67\x57\xfd\xbc\xdb"
+ "\x44\xe9\xc5\x65\xd1\xc7\x3b\xff"
+ "\x20\xa0\x80\x1a\xc3\x9a\xad\x5e"
+ "\x5d\x3b\xd3\x07\xd9\xf5\xfd\x3d"
+ "\x4a\x8b\xa8\xd2\x6e\x7a\x51\x65"
+ "\x6c\x8e\x95\xe0\x45\xc9\x5f\x4a"
+ "\x09\x3c\x3d\x71\x7f\x0c\x84\x2a"
+ "\xc8\x48\x52\x1a\xc2\xd5\xd6\x78"
+ "\x92\x1e\xa0\x90\x2e\xea\xf0\xf3"
+ "\xdc\x0f\xb1\xaf\x0d\x9b\x06\x2e"
+ "\x35\x10\x30\x82\x0d\xe7\xc5\x9b"
+ "\xde\x44\x18\xbd\x9f\xd1\x45\xa9"
+ "\x7b\x7a\x4a\xad\x35\x65\x27\xca"
+ "\xb2\xc3\xd4\x9b\x71\x86\x70\xee"
+ "\xf1\x89\x3b\x85\x4b\x5b\xaa\xaf"
+ "\xfc\x42\xc8\x31\x59\xbe\x16\x60"
+ "\x4f\xf9\xfa\x12\xea\xd0\xa7\x14"
+ "\xf0\x7a\xf3\xd5\x8d\xbd\x81\xef"
+ "\x52\x7f\x29\x51\x94\x20\x67\x3c"
+ "\xd1\xaf\x77\x9f\x22\x5a\x4e\x63"
+ "\xe7\xff\x73\x25\xd1\xdd\x96\x8a"
+ "\x98\x52\x6d\xf3\xac\x3e\xf2\x18"
+ "\x6d\xf6\x0a\x29\xa6\x34\x3d\xed"
+ "\xe3\x27\x0d\x9d\x0a\x02\x44\x7e"
+ "\x5a\x7e\x67\x0f\x0a\x9e\xd6\xad"
+ "\x91\xe6\x4d\x81\x8c\x5c\x59\xaa"
+ "\xfb\xeb\x56\x53\xd2\x7d\x4c\x81"
+ "\x65\x53\x0f\x41\x11\xbd\x98\x99"
+ "\xf9\xc6\xfa\x51\x2e\xa3\xdd\x8d"
+ "\x84\x98\xf9\x34\xed\x33\x2a\x1f"
+ "\x82\xed\xc1\x73\x98\xd3\x02\xdc"
+ "\xe6\xc2\x33\x1d\xa2\xb4\xca\x76"
+ "\x63\x51\x34\x9d\x96\x12\xae\xce"
+ "\x83\xc9\x76\x5e\xa4\x1b\x53\x37"
+ "\x17\xd5\xc0\x80\x1d\x62\xf8\x3d"
+ "\x54\x27\x74\xbb\x10\x86\x57\x46"
+ "\x68\xe1\xed\x14\xe7\x9d\xfc\x84"
+ "\x47\xbc\xc2\xf8\x19\x4b\x99\xcf"
+ "\x7a\xe9\xc4\xb8\x8c\x82\x72\x4d"
+ "\x7b\x4f\x38\x55\x36\x71\x64\xc1"
+ "\xfc\x5c\x75\x52\x33\x02\x18\xf8"
+ "\x17\xe1\x2b\xc2\x43\x39\xbd\x76"
+ "\x9b\x63\x76\x32\x2f\x19\x72\x10"
+ "\x9f\x21\x0c\xf1\x66\x50\x7f\xa5"
+ "\x0d\x1f\x46\xe0\xba\xd3\x2f\x3c",
+ .ilen = 512,
+ .result = "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17"
+ "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f"
+ "\x20\x21\x22\x23\x24\x25\x26\x27"
+ "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f"
+ "\x30\x31\x32\x33\x34\x35\x36\x37"
+ "\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f"
+ "\x40\x41\x42\x43\x44\x45\x46\x47"
+ "\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f"
+ "\x50\x51\x52\x53\x54\x55\x56\x57"
+ "\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f"
+ "\x60\x61\x62\x63\x64\x65\x66\x67"
+ "\x68\x69\x6a\x6b\x6c\x6d\x6e\x6f"
+ "\x70\x71\x72\x73\x74\x75\x76\x77"
+ "\x78\x79\x7a\x7b\x7c\x7d\x7e\x7f"
+ "\x80\x81\x82\x83\x84\x85\x86\x87"
+ "\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
+ "\x90\x91\x92\x93\x94\x95\x96\x97"
+ "\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
+ "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7"
+ "\xa8\xa9\xaa\xab\xac\xad\xae\xaf"
+ "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7"
+ "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf"
+ "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7"
+ "\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
+ "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7"
+ "\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf"
+ "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7"
+ "\xe8\xe9\xea\xeb\xec\xed\xee\xef"
+ "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7"
+ "\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff"
+ "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17"
+ "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f"
+ "\x20\x21\x22\x23\x24\x25\x26\x27"
+ "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f"
+ "\x30\x31\x32\x33\x34\x35\x36\x37"
+ "\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f"
+ "\x40\x41\x42\x43\x44\x45\x46\x47"
+ "\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f"
+ "\x50\x51\x52\x53\x54\x55\x56\x57"
+ "\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f"
+ "\x60\x61\x62\x63\x64\x65\x66\x67"
+ "\x68\x69\x6a\x6b\x6c\x6d\x6e\x6f"
+ "\x70\x71\x72\x73\x74\x75\x76\x77"
+ "\x78\x79\x7a\x7b\x7c\x7d\x7e\x7f"
+ "\x80\x81\x82\x83\x84\x85\x86\x87"
+ "\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
+ "\x90\x91\x92\x93\x94\x95\x96\x97"
+ "\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
+ "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7"
+ "\xa8\xa9\xaa\xab\xac\xad\xae\xaf"
+ "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7"
+ "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf"
+ "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7"
+ "\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
+ "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7"
+ "\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf"
+ "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7"
+ "\xe8\xe9\xea\xeb\xec\xed\xee\xef"
+ "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7"
+ "\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff",
+ .rlen = 512,
+ .also_non_np = 1,
+ .np = 3,
+ .tap = { 512 - 20, 4, 16 },
+ }
+};
+
/* Cast6 test vectors from RFC 2612 */
static const struct cipher_testvec cast6_enc_tv_template[] = {
{
--
2.16.1.291.g4437f3f132-goog
Add an ARM NEON-accelerated implementation of Speck-XTS. It operates on
128-byte chunks at a time, i.e. 8 blocks for Speck128 or 16 blocks for
Speck64. Each 128-byte chunk goes through XTS preprocessing, then is
encrypted/decrypted (doing one cipher round for all the blocks, then the
next round, etc.), then goes through XTS postprocessing.
The performance depends on the processor but can be about 3 times faster
than the generic code. For example, on an ARMv7 processor we observe
the following performance with Speck128/256-XTS:
xts-speck128-neon: Encryption 107.9 MB/s, Decryption 108.1 MB/s
xts(speck128-generic): Encryption 32.1 MB/s, Decryption 36.6 MB/s
In comparison to AES-256-XTS without the Cryptography Extensions:
xts-aes-neonbs: Encryption 41.2 MB/s, Decryption 36.7 MB/s
xts(aes-asm): Encryption 31.7 MB/s, Decryption 30.8 MB/s
xts(aes-generic): Encryption 21.2 MB/s, Decryption 20.9 MB/s
Speck64/128-XTS is even faster:
xts-speck64-neon: Encryption 138.6 MB/s, Decryption 139.1 MB/s
Note that as with the generic code, only the Speck128 and Speck64
variants are supported. Also, for now only the XTS mode of operation is
supported, to target the disk and file encryption use cases. The NEON
code also only handles the portion of the data that is evenly divisible
into 128-byte chunks, with any remainder handled by a C fallback. Of
course, other modes of operation could be added later if needed, and/or
the NEON code could be updated to handle other buffer sizes.
The XTS specification is only defined for AES which has a 128-bit block
size, so for the GF(2^64) math needed for Speck64-XTS we use the
reducing polynomial 'x^64 + x^4 + x^3 + x + 1' given by the original XEX
paper. Of course, when possible users should use Speck128-XTS, but even
that may be too slow on some processors; Speck64-XTS can be faster.
Signed-off-by: Eric Biggers <[email protected]>
---
arch/arm/crypto/Kconfig | 6 +
arch/arm/crypto/Makefile | 2 +
arch/arm/crypto/speck-neon-core.S | 432 ++++++++++++++++++++++++++++++
arch/arm/crypto/speck-neon-glue.c | 288 ++++++++++++++++++++
4 files changed, 728 insertions(+)
create mode 100644 arch/arm/crypto/speck-neon-core.S
create mode 100644 arch/arm/crypto/speck-neon-glue.c
diff --git a/arch/arm/crypto/Kconfig b/arch/arm/crypto/Kconfig
index b8e69fe282b8..925d1364727a 100644
--- a/arch/arm/crypto/Kconfig
+++ b/arch/arm/crypto/Kconfig
@@ -121,4 +121,10 @@ config CRYPTO_CHACHA20_NEON
select CRYPTO_BLKCIPHER
select CRYPTO_CHACHA20
+config CRYPTO_SPECK_NEON
+ tristate "NEON accelerated Speck cipher algorithms"
+ depends on KERNEL_MODE_NEON
+ select CRYPTO_BLKCIPHER
+ select CRYPTO_SPECK
+
endif
diff --git a/arch/arm/crypto/Makefile b/arch/arm/crypto/Makefile
index 30ef8e291271..a758107c5525 100644
--- a/arch/arm/crypto/Makefile
+++ b/arch/arm/crypto/Makefile
@@ -10,6 +10,7 @@ obj-$(CONFIG_CRYPTO_SHA1_ARM_NEON) += sha1-arm-neon.o
obj-$(CONFIG_CRYPTO_SHA256_ARM) += sha256-arm.o
obj-$(CONFIG_CRYPTO_SHA512_ARM) += sha512-arm.o
obj-$(CONFIG_CRYPTO_CHACHA20_NEON) += chacha20-neon.o
+obj-$(CONFIG_CRYPTO_SPECK_NEON) += speck-neon.o
ce-obj-$(CONFIG_CRYPTO_AES_ARM_CE) += aes-arm-ce.o
ce-obj-$(CONFIG_CRYPTO_SHA1_ARM_CE) += sha1-arm-ce.o
@@ -53,6 +54,7 @@ ghash-arm-ce-y := ghash-ce-core.o ghash-ce-glue.o
crct10dif-arm-ce-y := crct10dif-ce-core.o crct10dif-ce-glue.o
crc32-arm-ce-y:= crc32-ce-core.o crc32-ce-glue.o
chacha20-neon-y := chacha20-neon-core.o chacha20-neon-glue.o
+speck-neon-y := speck-neon-core.o speck-neon-glue.o
quiet_cmd_perl = PERL $@
cmd_perl = $(PERL) $(<) > $(@)
diff --git a/arch/arm/crypto/speck-neon-core.S b/arch/arm/crypto/speck-neon-core.S
new file mode 100644
index 000000000000..3c1e203e53b9
--- /dev/null
+++ b/arch/arm/crypto/speck-neon-core.S
@@ -0,0 +1,432 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * NEON-accelerated implementation of Speck128-XTS and Speck64-XTS
+ *
+ * Copyright (c) 2018 Google, Inc
+ *
+ * Author: Eric Biggers <[email protected]>
+ */
+
+#include <linux/linkage.h>
+
+ .text
+ .fpu neon
+
+ // arguments
+ ROUND_KEYS .req r0 // const {u64,u32} *round_keys
+ NROUNDS .req r1 // int nrounds
+ DST .req r2 // void *dst
+ SRC .req r3 // const void *src
+ NBYTES .req r4 // unsigned int nbytes
+ TWEAK .req r5 // void *tweak
+
+ // registers which hold the data being encrypted/decrypted
+ X0 .req q0
+ X0_L .req d0
+ X0_H .req d1
+ Y0 .req q1
+ Y0_H .req d3
+ X1 .req q2
+ X1_L .req d4
+ X1_H .req d5
+ Y1 .req q3
+ Y1_H .req d7
+ X2 .req q4
+ X2_L .req d8
+ X2_H .req d9
+ Y2 .req q5
+ Y2_H .req d11
+ X3 .req q6
+ X3_L .req d12
+ X3_H .req d13
+ Y3 .req q7
+ Y3_H .req d15
+
+ // the round key, duplicated in all lanes
+ ROUND_KEY .req q8
+ ROUND_KEY_L .req d16
+ ROUND_KEY_H .req d17
+
+ // index vector for vtbl-based 8-bit rotates
+ ROTATE_TABLE .req d18
+
+ // multiplication table for updating XTS tweaks
+ GF128MUL_TABLE .req d19
+ GF64MUL_TABLE .req d19
+
+ // current XTS tweak value(s)
+ TWEAKV .req q10
+ TWEAKV_L .req d20
+ TWEAKV_H .req d21
+
+ TMP0 .req q12
+ TMP0_L .req d24
+ TMP0_H .req d25
+ TMP1 .req q13
+ TMP2 .req q14
+ TMP3 .req q15
+
+ .align 4
+.Lror64_8_table:
+ .byte 1, 2, 3, 4, 5, 6, 7, 0
+.Lror32_8_table:
+ .byte 1, 2, 3, 0, 5, 6, 7, 4
+.Lrol64_8_table:
+ .byte 7, 0, 1, 2, 3, 4, 5, 6
+.Lrol32_8_table:
+ .byte 3, 0, 1, 2, 7, 4, 5, 6
+.Lgf128mul_table:
+ .byte 0, 0x87
+ .fill 14
+.Lgf64mul_table:
+ .byte 0, 0x1b, (0x1b << 1), (0x1b << 1) ^ 0x1b
+ .fill 12
+
+/*
+ * _speck_round_128bytes() - Speck encryption round on 128 bytes at a time
+ *
+ * Do one Speck encryption round on the 128 bytes (8 blocks for Speck128, 16 for
+ * Speck64) stored in X0-X3 and Y0-Y3, using the round key stored in all lanes
+ * of ROUND_KEY. 'n' is the lane size: 64 for Speck128, or 32 for Speck64.
+ *
+ * The 8-bit rotates are implemented using vtbl instead of vshr + vsli because
+ * the vtbl approach is faster on some processors and the same speed on others.
+ */
+.macro _speck_round_128bytes n
+
+ // x = ror(x, 8)
+ vtbl.8 X0_L, {X0_L}, ROTATE_TABLE
+ vtbl.8 X0_H, {X0_H}, ROTATE_TABLE
+ vtbl.8 X1_L, {X1_L}, ROTATE_TABLE
+ vtbl.8 X1_H, {X1_H}, ROTATE_TABLE
+ vtbl.8 X2_L, {X2_L}, ROTATE_TABLE
+ vtbl.8 X2_H, {X2_H}, ROTATE_TABLE
+ vtbl.8 X3_L, {X3_L}, ROTATE_TABLE
+ vtbl.8 X3_H, {X3_H}, ROTATE_TABLE
+
+ // x += y
+ vadd.u\n X0, Y0
+ vadd.u\n X1, Y1
+ vadd.u\n X2, Y2
+ vadd.u\n X3, Y3
+
+ // x ^= k
+ veor X0, ROUND_KEY
+ veor X1, ROUND_KEY
+ veor X2, ROUND_KEY
+ veor X3, ROUND_KEY
+
+ // y = rol(y, 3)
+ vshl.u\n TMP0, Y0, #3
+ vshl.u\n TMP1, Y1, #3
+ vshl.u\n TMP2, Y2, #3
+ vshl.u\n TMP3, Y3, #3
+ vsri.u\n TMP0, Y0, #(\n - 3)
+ vsri.u\n TMP1, Y1, #(\n - 3)
+ vsri.u\n TMP2, Y2, #(\n - 3)
+ vsri.u\n TMP3, Y3, #(\n - 3)
+
+ // y ^= x
+ veor Y0, TMP0, X0
+ veor Y1, TMP1, X1
+ veor Y2, TMP2, X2
+ veor Y3, TMP3, X3
+.endm
+
+/*
+ * _speck_unround_128bytes() - Speck decryption round on 128 bytes at a time
+ *
+ * This is the inverse of _speck_round_128bytes().
+ */
+.macro _speck_unround_128bytes n
+
+ // y ^= x
+ veor TMP0, Y0, X0
+ veor TMP1, Y1, X1
+ veor TMP2, Y2, X2
+ veor TMP3, Y3, X3
+
+ // y = ror(y, 3)
+ vshr.u\n Y0, TMP0, #3
+ vshr.u\n Y1, TMP1, #3
+ vshr.u\n Y2, TMP2, #3
+ vshr.u\n Y3, TMP3, #3
+ vsli.u\n Y0, TMP0, #(\n - 3)
+ vsli.u\n Y1, TMP1, #(\n - 3)
+ vsli.u\n Y2, TMP2, #(\n - 3)
+ vsli.u\n Y3, TMP3, #(\n - 3)
+
+ // x ^= k
+ veor X0, ROUND_KEY
+ veor X1, ROUND_KEY
+ veor X2, ROUND_KEY
+ veor X3, ROUND_KEY
+
+ // x -= y
+ vsub.u\n X0, Y0
+ vsub.u\n X1, Y1
+ vsub.u\n X2, Y2
+ vsub.u\n X3, Y3
+
+ // x = rol(x, 8);
+ vtbl.8 X0_L, {X0_L}, ROTATE_TABLE
+ vtbl.8 X0_H, {X0_H}, ROTATE_TABLE
+ vtbl.8 X1_L, {X1_L}, ROTATE_TABLE
+ vtbl.8 X1_H, {X1_H}, ROTATE_TABLE
+ vtbl.8 X2_L, {X2_L}, ROTATE_TABLE
+ vtbl.8 X2_H, {X2_H}, ROTATE_TABLE
+ vtbl.8 X3_L, {X3_L}, ROTATE_TABLE
+ vtbl.8 X3_H, {X3_H}, ROTATE_TABLE
+.endm
+
+.macro _xts128_precrypt_one dst_reg, tweak_buf, tmp
+
+ // Load the next source block
+ vld1.8 {\dst_reg}, [SRC]!
+
+ // Save the current tweak in the tweak buffer
+ vst1.8 {TWEAKV}, [\tweak_buf:128]!
+
+ // XOR the next source block with the current tweak
+ veor \dst_reg, TWEAKV
+
+ /*
+ * Calculate the next tweak by multiplying the current one by x,
+ * modulo p(x) = x^128 + x^7 + x^2 + x + 1.
+ */
+ vshr.u64 \tmp, TWEAKV, #63
+ vshl.u64 TWEAKV, #1
+ veor TWEAKV_H, \tmp\()_L
+ vtbl.8 \tmp\()_H, {GF128MUL_TABLE}, \tmp\()_H
+ veor TWEAKV_L, \tmp\()_H
+.endm
+
+.macro _xts64_precrypt_two dst_reg, tweak_buf, tmp
+
+ // Load the next two source blocks
+ vld1.8 {\dst_reg}, [SRC]!
+
+ // Save the current two tweaks in the tweak buffer
+ vst1.8 {TWEAKV}, [\tweak_buf:128]!
+
+ // XOR the next two source blocks with the current two tweaks
+ veor \dst_reg, TWEAKV
+
+ /*
+ * Calculate the next two tweaks by multiplying the current ones by x^2,
+ * modulo p(x) = x^64 + x^4 + x^3 + x + 1.
+ */
+ vshr.u64 \tmp, TWEAKV, #62
+ vshl.u64 TWEAKV, #2
+ vtbl.8 \tmp\()_L, {GF64MUL_TABLE}, \tmp\()_L
+ vtbl.8 \tmp\()_H, {GF64MUL_TABLE}, \tmp\()_H
+ veor TWEAKV, \tmp
+.endm
+
+/*
+ * _speck_xts_crypt() - Speck-XTS encryption/decryption
+ *
+ * Encrypt or decrypt NBYTES bytes of data from the SRC buffer to the DST buffer
+ * using Speck-XTS, specifically the variant with a block size of '2n' and round
+ * count given by NROUNDS. The expanded round keys are given in ROUND_KEYS, and
+ * the current XTS tweak value is given in TWEAK. It's assumed that NBYTES is a
+ * nonzero multiple of 128.
+ */
+.macro _speck_xts_crypt n, decrypting
+ push {r4-r7}
+ mov r7, sp
+
+ /*
+ * The first four parameters were passed in registers r0-r3. Load the
+ * additional parameters, which were passed on the stack.
+ */
+ ldr NBYTES, [sp, #16]
+ ldr TWEAK, [sp, #20]
+
+ /*
+ * If decrypting, modify the ROUND_KEYS parameter to point to the last
+ * round key rather than the first, since for decryption the round keys
+ * are used in reverse order.
+ */
+.if \decrypting
+.if \n == 64
+ add ROUND_KEYS, ROUND_KEYS, NROUNDS, lsl #3
+ sub ROUND_KEYS, #8
+.else
+ add ROUND_KEYS, ROUND_KEYS, NROUNDS, lsl #2
+ sub ROUND_KEYS, #4
+.endif
+.endif
+
+ // Load the index vector for vtbl-based 8-bit rotates
+.if \decrypting
+ ldr r12, =.Lrol\n\()_8_table
+.else
+ ldr r12, =.Lror\n\()_8_table
+.endif
+ vld1.8 {ROTATE_TABLE}, [r12:64]
+
+ // One-time XTS preparation
+
+ /*
+ * Allocate stack space to store 128 bytes worth of tweaks. For
+ * performance, this space is aligned to a 16-byte boundary so that we
+ * can use the load/store instructions that declare 16-byte alignment.
+ */
+ sub sp, #128
+ bic sp, #0xf
+
+.if \n == 64
+ // Load first tweak
+ vld1.8 {TWEAKV}, [TWEAK]
+
+ // Load GF(2^128) multiplication table
+ ldr r12, =.Lgf128mul_table
+ vld1.8 {GF128MUL_TABLE}, [r12:64]
+.else
+ // Load first tweak
+ vld1.8 {TWEAKV_L}, [TWEAK]
+
+ // Load GF(2^64) multiplication table
+ ldr r12, =.Lgf64mul_table
+ vld1.8 {GF64MUL_TABLE}, [r12:64]
+
+ // Calculate second tweak, packing it together with the first
+ vshr.u64 TMP0_L, TWEAKV_L, #63
+ vtbl.u8 TMP0_L, {GF64MUL_TABLE}, TMP0_L
+ vshl.u64 TWEAKV_H, TWEAKV_L, #1
+ veor TWEAKV_H, TMP0_L
+.endif
+
+.Lnext_128bytes_\@:
+
+ /*
+ * Load the source blocks into {X,Y}[0-3], XOR them with their XTS tweak
+ * values, and save the tweaks on the stack for later. Then
+ * de-interleave the 'x' and 'y' elements of each block, i.e. make it so
+ * that the X[0-3] registers contain only the second halves of blocks,
+ * and the Y[0-3] registers contain only the first halves of blocks.
+ * (Speck uses the order (y, x) rather than the more intuitive (x, y).)
+ */
+ mov r12, sp
+.if \n == 64
+ _xts128_precrypt_one X0, r12, TMP0
+ _xts128_precrypt_one Y0, r12, TMP0
+ _xts128_precrypt_one X1, r12, TMP0
+ _xts128_precrypt_one Y1, r12, TMP0
+ _xts128_precrypt_one X2, r12, TMP0
+ _xts128_precrypt_one Y2, r12, TMP0
+ _xts128_precrypt_one X3, r12, TMP0
+ _xts128_precrypt_one Y3, r12, TMP0
+ vswp X0_L, Y0_H
+ vswp X1_L, Y1_H
+ vswp X2_L, Y2_H
+ vswp X3_L, Y3_H
+.else
+ _xts64_precrypt_two X0, r12, TMP0
+ _xts64_precrypt_two Y0, r12, TMP0
+ _xts64_precrypt_two X1, r12, TMP0
+ _xts64_precrypt_two Y1, r12, TMP0
+ _xts64_precrypt_two X2, r12, TMP0
+ _xts64_precrypt_two Y2, r12, TMP0
+ _xts64_precrypt_two X3, r12, TMP0
+ _xts64_precrypt_two Y3, r12, TMP0
+ vuzp.32 Y0, X0
+ vuzp.32 Y1, X1
+ vuzp.32 Y2, X2
+ vuzp.32 Y3, X3
+.endif
+
+ // Do the cipher rounds
+
+ mov r12, ROUND_KEYS
+ mov r6, NROUNDS
+
+.Lnext_round_\@:
+.if \decrypting
+.if \n == 64
+ vld1.64 ROUND_KEY_L, [r12]
+ sub r12, #8
+ vmov ROUND_KEY_H, ROUND_KEY_L
+.else
+ vld1.32 {ROUND_KEY_L[],ROUND_KEY_H[]}, [r12]
+ sub r12, #4
+.endif
+ _speck_unround_128bytes \n
+.else
+.if \n == 64
+ vld1.64 ROUND_KEY_L, [r12]!
+ vmov ROUND_KEY_H, ROUND_KEY_L
+.else
+ vld1.32 {ROUND_KEY_L[],ROUND_KEY_H[]}, [r12]!
+.endif
+ _speck_round_128bytes \n
+.endif
+ subs r6, r6, #1
+ bne .Lnext_round_\@
+
+ // Re-interleave the 'x' and 'y' elements of each block
+.if \n == 64
+ vswp X0_L, Y0_H
+ vswp X1_L, Y1_H
+ vswp X2_L, Y2_H
+ vswp X3_L, Y3_H
+.else
+ vzip.32 Y0, X0
+ vzip.32 Y1, X1
+ vzip.32 Y2, X2
+ vzip.32 Y3, X3
+.endif
+
+ // XOR the encrypted/decrypted blocks with the tweaks we saved earlier
+ mov r12, sp
+ vld1.8 {TMP0, TMP1}, [r12:128]!
+ vld1.8 {TMP2, TMP3}, [r12:128]!
+ veor X0, TMP0
+ veor Y0, TMP1
+ veor X1, TMP2
+ veor Y1, TMP3
+ vld1.8 {TMP0, TMP1}, [r12:128]!
+ vld1.8 {TMP2, TMP3}, [r12:128]!
+ veor X2, TMP0
+ veor Y2, TMP1
+ veor X3, TMP2
+ veor Y3, TMP3
+
+ // Store the ciphertext in the destination buffer
+ vst1.8 {X0, Y0}, [DST]!
+ vst1.8 {X1, Y1}, [DST]!
+ vst1.8 {X2, Y2}, [DST]!
+ vst1.8 {X3, Y3}, [DST]!
+
+ // Continue if there are more 128-byte chunks remaining, else return
+ subs NBYTES, #128
+ bne .Lnext_128bytes_\@
+
+ // Store the next tweak
+.if \n == 64
+ vst1.8 {TWEAKV}, [TWEAK]
+.else
+ vst1.8 {TWEAKV_L}, [TWEAK]
+.endif
+
+ mov sp, r7
+ pop {r4-r7}
+ bx lr
+.endm
+
+ENTRY(speck128_xts_encrypt_neon)
+ _speck_xts_crypt n=64, decrypting=0
+ENDPROC(speck128_xts_encrypt_neon)
+
+ENTRY(speck128_xts_decrypt_neon)
+ _speck_xts_crypt n=64, decrypting=1
+ENDPROC(speck128_xts_decrypt_neon)
+
+ENTRY(speck64_xts_encrypt_neon)
+ _speck_xts_crypt n=32, decrypting=0
+ENDPROC(speck64_xts_encrypt_neon)
+
+ENTRY(speck64_xts_decrypt_neon)
+ _speck_xts_crypt n=32, decrypting=1
+ENDPROC(speck64_xts_decrypt_neon)
diff --git a/arch/arm/crypto/speck-neon-glue.c b/arch/arm/crypto/speck-neon-glue.c
new file mode 100644
index 000000000000..f012c3ea998f
--- /dev/null
+++ b/arch/arm/crypto/speck-neon-glue.c
@@ -0,0 +1,288 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * NEON-accelerated implementation of Speck128-XTS and Speck64-XTS
+ *
+ * Copyright (c) 2018 Google, Inc
+ *
+ * Note: the NIST recommendation for XTS only specifies a 128-bit block size,
+ * but a 64-bit version (needed for Speck64) is fairly straightforward; the math
+ * is just done in GF(2^64) instead of GF(2^128), with the reducing polynomial
+ * x^64 + x^4 + x^3 + x + 1 from the original XEX paper (Rogaway, 2004:
+ * "Efficient Instantiations of Tweakable Blockciphers and Refinements to Modes
+ * OCB and PMAC"), represented as 0x1B.
+ */
+
+#include <asm/hwcap.h>
+#include <asm/neon.h>
+#include <asm/simd.h>
+#include <crypto/algapi.h>
+#include <crypto/gf128mul.h>
+#include <crypto/internal/skcipher.h>
+#include <crypto/speck.h>
+#include <crypto/xts.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+
+/* The assembly functions only handle multiples of 128 bytes */
+#define SPECK_NEON_CHUNK_SIZE 128
+
+/* Speck128 */
+
+struct speck128_xts_tfm_ctx {
+ struct speck128_tfm_ctx main_key;
+ struct speck128_tfm_ctx tweak_key;
+};
+
+asmlinkage void speck128_xts_encrypt_neon(const u64 *round_keys, int nrounds,
+ void *dst, const void *src,
+ unsigned int nbytes, void *tweak);
+
+asmlinkage void speck128_xts_decrypt_neon(const u64 *round_keys, int nrounds,
+ void *dst, const void *src,
+ unsigned int nbytes, void *tweak);
+
+typedef void (*speck128_crypt_one_t)(const struct speck128_tfm_ctx *,
+ u8 *, const u8 *);
+typedef void (*speck128_xts_crypt_many_t)(const u64 *, int, void *,
+ const void *, unsigned int, void *);
+
+static __always_inline int
+__speck128_xts_crypt(struct skcipher_request *req,
+ speck128_crypt_one_t crypt_one,
+ speck128_xts_crypt_many_t crypt_many)
+{
+ struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
+ const struct speck128_xts_tfm_ctx *ctx = crypto_skcipher_ctx(tfm);
+ struct skcipher_walk walk;
+ le128 tweak;
+ int err;
+
+ err = skcipher_walk_virt(&walk, req, true);
+
+ crypto_speck128_encrypt(&ctx->tweak_key, (u8 *)&tweak, walk.iv);
+
+ while (walk.nbytes > 0) {
+ unsigned int nbytes = walk.nbytes;
+ u8 *dst = walk.dst.virt.addr;
+ const u8 *src = walk.src.virt.addr;
+
+ if (nbytes >= SPECK_NEON_CHUNK_SIZE && may_use_simd()) {
+ unsigned int count;
+
+ count = round_down(nbytes, SPECK_NEON_CHUNK_SIZE);
+ kernel_neon_begin();
+ (*crypt_many)(ctx->main_key.round_keys,
+ ctx->main_key.nrounds,
+ dst, src, count, &tweak);
+ kernel_neon_end();
+ dst += count;
+ src += count;
+ nbytes -= count;
+ }
+
+ /* Handle any remainder with generic code */
+ while (nbytes >= sizeof(tweak)) {
+ le128_xor((le128 *)dst, (const le128 *)src, &tweak);
+ (*crypt_one)(&ctx->main_key, dst, dst);
+ le128_xor((le128 *)dst, (const le128 *)dst, &tweak);
+ gf128mul_x_ble(&tweak, &tweak);
+
+ dst += sizeof(tweak);
+ src += sizeof(tweak);
+ nbytes -= sizeof(tweak);
+ }
+ err = skcipher_walk_done(&walk, nbytes);
+ }
+
+ return err;
+}
+
+static int speck128_xts_encrypt(struct skcipher_request *req)
+{
+ return __speck128_xts_crypt(req, crypto_speck128_encrypt,
+ speck128_xts_encrypt_neon);
+}
+
+static int speck128_xts_decrypt(struct skcipher_request *req)
+{
+ return __speck128_xts_crypt(req, crypto_speck128_decrypt,
+ speck128_xts_decrypt_neon);
+}
+
+static int speck128_xts_setkey(struct crypto_skcipher *tfm, const u8 *key,
+ unsigned int keylen)
+{
+ struct speck128_xts_tfm_ctx *ctx = crypto_skcipher_ctx(tfm);
+ int err;
+
+ err = xts_verify_key(tfm, key, keylen);
+ if (err)
+ return err;
+
+ keylen /= 2;
+
+ err = crypto_speck128_setkey(&ctx->main_key, key, keylen);
+ if (err)
+ return err;
+
+ return crypto_speck128_setkey(&ctx->tweak_key, key + keylen, keylen);
+}
+
+/* Speck64 */
+
+struct speck64_xts_tfm_ctx {
+ struct speck64_tfm_ctx main_key;
+ struct speck64_tfm_ctx tweak_key;
+};
+
+asmlinkage void speck64_xts_encrypt_neon(const u32 *round_keys, int nrounds,
+ void *dst, const void *src,
+ unsigned int nbytes, void *tweak);
+
+asmlinkage void speck64_xts_decrypt_neon(const u32 *round_keys, int nrounds,
+ void *dst, const void *src,
+ unsigned int nbytes, void *tweak);
+
+typedef void (*speck64_crypt_one_t)(const struct speck64_tfm_ctx *,
+ u8 *, const u8 *);
+typedef void (*speck64_xts_crypt_many_t)(const u32 *, int, void *,
+ const void *, unsigned int, void *);
+
+static __always_inline int
+__speck64_xts_crypt(struct skcipher_request *req, speck64_crypt_one_t crypt_one,
+ speck64_xts_crypt_many_t crypt_many)
+{
+ struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
+ const struct speck64_xts_tfm_ctx *ctx = crypto_skcipher_ctx(tfm);
+ struct skcipher_walk walk;
+ __le64 tweak;
+ int err;
+
+ err = skcipher_walk_virt(&walk, req, true);
+
+ crypto_speck64_encrypt(&ctx->tweak_key, (u8 *)&tweak, walk.iv);
+
+ while (walk.nbytes > 0) {
+ unsigned int nbytes = walk.nbytes;
+ u8 *dst = walk.dst.virt.addr;
+ const u8 *src = walk.src.virt.addr;
+
+ if (nbytes >= SPECK_NEON_CHUNK_SIZE && may_use_simd()) {
+ unsigned int count;
+
+ count = round_down(nbytes, SPECK_NEON_CHUNK_SIZE);
+ kernel_neon_begin();
+ (*crypt_many)(ctx->main_key.round_keys,
+ ctx->main_key.nrounds,
+ dst, src, count, &tweak);
+ kernel_neon_end();
+ dst += count;
+ src += count;
+ nbytes -= count;
+ }
+
+ /* Handle any remainder with generic code */
+ while (nbytes >= sizeof(tweak)) {
+ *(__le64 *)dst = *(__le64 *)src ^ tweak;
+ (*crypt_one)(&ctx->main_key, dst, dst);
+ *(__le64 *)dst ^= tweak;
+ tweak = cpu_to_le64((le64_to_cpu(tweak) << 1) ^
+ ((tweak & cpu_to_le64(1ULL << 63)) ?
+ 0x1B : 0));
+ dst += sizeof(tweak);
+ src += sizeof(tweak);
+ nbytes -= sizeof(tweak);
+ }
+ err = skcipher_walk_done(&walk, nbytes);
+ }
+
+ return err;
+}
+
+static int speck64_xts_encrypt(struct skcipher_request *req)
+{
+ return __speck64_xts_crypt(req, crypto_speck64_encrypt,
+ speck64_xts_encrypt_neon);
+}
+
+static int speck64_xts_decrypt(struct skcipher_request *req)
+{
+ return __speck64_xts_crypt(req, crypto_speck64_decrypt,
+ speck64_xts_decrypt_neon);
+}
+
+static int speck64_xts_setkey(struct crypto_skcipher *tfm, const u8 *key,
+ unsigned int keylen)
+{
+ struct speck64_xts_tfm_ctx *ctx = crypto_skcipher_ctx(tfm);
+ int err;
+
+ err = xts_verify_key(tfm, key, keylen);
+ if (err)
+ return err;
+
+ keylen /= 2;
+
+ err = crypto_speck64_setkey(&ctx->main_key, key, keylen);
+ if (err)
+ return err;
+
+ return crypto_speck64_setkey(&ctx->tweak_key, key + keylen, keylen);
+}
+
+static struct skcipher_alg speck_algs[] = {
+ {
+ .base.cra_name = "xts(speck128)",
+ .base.cra_driver_name = "xts-speck128-neon",
+ .base.cra_priority = 300,
+ .base.cra_blocksize = SPECK128_BLOCK_SIZE,
+ .base.cra_ctxsize = sizeof(struct speck128_xts_tfm_ctx),
+ .base.cra_alignmask = 7,
+ .base.cra_module = THIS_MODULE,
+ .min_keysize = 2 * SPECK128_128_KEY_SIZE,
+ .max_keysize = 2 * SPECK128_256_KEY_SIZE,
+ .ivsize = SPECK128_BLOCK_SIZE,
+ .walksize = SPECK_NEON_CHUNK_SIZE,
+ .setkey = speck128_xts_setkey,
+ .encrypt = speck128_xts_encrypt,
+ .decrypt = speck128_xts_decrypt,
+ }, {
+ .base.cra_name = "xts(speck64)",
+ .base.cra_driver_name = "xts-speck64-neon",
+ .base.cra_priority = 300,
+ .base.cra_blocksize = SPECK64_BLOCK_SIZE,
+ .base.cra_ctxsize = sizeof(struct speck64_xts_tfm_ctx),
+ .base.cra_alignmask = 7,
+ .base.cra_module = THIS_MODULE,
+ .min_keysize = 2 * SPECK64_96_KEY_SIZE,
+ .max_keysize = 2 * SPECK64_128_KEY_SIZE,
+ .ivsize = SPECK64_BLOCK_SIZE,
+ .walksize = SPECK_NEON_CHUNK_SIZE,
+ .setkey = speck64_xts_setkey,
+ .encrypt = speck64_xts_encrypt,
+ .decrypt = speck64_xts_decrypt,
+ }
+};
+
+static int __init speck_neon_module_init(void)
+{
+ if (!(elf_hwcap & HWCAP_NEON))
+ return -ENODEV;
+ return crypto_register_skciphers(speck_algs, ARRAY_SIZE(speck_algs));
+}
+
+static void __exit speck_neon_module_exit(void)
+{
+ crypto_unregister_skciphers(speck_algs, ARRAY_SIZE(speck_algs));
+}
+
+module_init(speck_neon_module_init);
+module_exit(speck_neon_module_exit);
+
+MODULE_DESCRIPTION("Speck block cipher (NEON-accelerated)");
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Eric Biggers <[email protected]>");
+MODULE_ALIAS_CRYPTO("xts(speck128)");
+MODULE_ALIAS_CRYPTO("xts-speck128-neon");
+MODULE_ALIAS_CRYPTO("xts(speck64)");
+MODULE_ALIAS_CRYPTO("xts-speck64-neon");
--
2.16.1.291.g4437f3f132-goog
Add test vectors for Speck128-XTS, generated in userspace using C code.
The inputs were borrowed from the AES-XTS test vectors.
Both xts(speck128-generic) and xts-speck128-neon pass these tests.
Signed-off-by: Eric Biggers <[email protected]>
---
crypto/testmgr.c | 9 +
crypto/testmgr.h | 687 +++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 696 insertions(+)
diff --git a/crypto/testmgr.c b/crypto/testmgr.c
index 058ed5eb6620..e011a347d51b 100644
--- a/crypto/testmgr.c
+++ b/crypto/testmgr.c
@@ -3575,6 +3575,15 @@ static const struct alg_test_desc alg_test_descs[] = {
.dec = __VECS(serpent_xts_dec_tv_template)
}
}
+ }, {
+ .alg = "xts(speck128)",
+ .test = alg_test_skcipher,
+ .suite = {
+ .cipher = {
+ .enc = __VECS(speck128_xts_enc_tv_template),
+ .dec = __VECS(speck128_xts_dec_tv_template)
+ }
+ }
}, {
.alg = "xts(twofish)",
.test = alg_test_skcipher,
diff --git a/crypto/testmgr.h b/crypto/testmgr.h
index 3818210f77cf..0212e0ebcd0c 100644
--- a/crypto/testmgr.h
+++ b/crypto/testmgr.h
@@ -14411,6 +14411,693 @@ static const struct cipher_testvec speck128_dec_tv_template[] = {
},
};
+/*
+ * Speck128-XTS test vectors, taken from the AES-XTS test vectors with the
+ * result recomputed with Speck128 as the cipher
+ */
+
+static const struct cipher_testvec speck128_xts_enc_tv_template[] = {
+ {
+ .key = "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .klen = 32,
+ .iv = "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .ilen = 32,
+ .result = "\xbe\xa0\xe7\x03\xd7\xfe\xab\x62"
+ "\x3b\x99\x4a\x64\x74\x77\xac\xed"
+ "\xd8\xf4\xa6\xcf\xae\xb9\x07\x42"
+ "\x51\xd9\xb6\x1d\xe0\x5e\xbc\x54",
+ .rlen = 32,
+ }, {
+ .key = "\x11\x11\x11\x11\x11\x11\x11\x11"
+ "\x11\x11\x11\x11\x11\x11\x11\x11"
+ "\x22\x22\x22\x22\x22\x22\x22\x22"
+ "\x22\x22\x22\x22\x22\x22\x22\x22",
+ .klen = 32,
+ .iv = "\x33\x33\x33\x33\x33\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44",
+ .ilen = 32,
+ .result = "\xfb\x53\x81\x75\x6f\x9f\x34\xad"
+ "\x7e\x01\xed\x7b\xcc\xda\x4e\x4a"
+ "\xd4\x84\xa4\x53\xd5\x88\x73\x1b"
+ "\xfd\xcb\xae\x0d\xf3\x04\xee\xe6",
+ .rlen = 32,
+ }, {
+ .key = "\xff\xfe\xfd\xfc\xfb\xfa\xf9\xf8"
+ "\xf7\xf6\xf5\xf4\xf3\xf2\xf1\xf0"
+ "\x22\x22\x22\x22\x22\x22\x22\x22"
+ "\x22\x22\x22\x22\x22\x22\x22\x22",
+ .klen = 32,
+ .iv = "\x33\x33\x33\x33\x33\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44",
+ .ilen = 32,
+ .result = "\x21\x52\x84\x15\xd1\xf7\x21\x55"
+ "\xd9\x75\x4a\xd3\xc5\xdb\x9f\x7d"
+ "\xda\x63\xb2\xf1\x82\xb0\x89\x59"
+ "\x86\xd4\xaa\xaa\xdd\xff\x4f\x92",
+ .rlen = 32,
+ }, {
+ .key = "\x27\x18\x28\x18\x28\x45\x90\x45"
+ "\x23\x53\x60\x28\x74\x71\x35\x26"
+ "\x31\x41\x59\x26\x53\x58\x97\x93"
+ "\x23\x84\x62\x64\x33\x83\x27\x95",
+ .klen = 32,
+ .iv = "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17"
+ "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f"
+ "\x20\x21\x22\x23\x24\x25\x26\x27"
+ "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f"
+ "\x30\x31\x32\x33\x34\x35\x36\x37"
+ "\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f"
+ "\x40\x41\x42\x43\x44\x45\x46\x47"
+ "\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f"
+ "\x50\x51\x52\x53\x54\x55\x56\x57"
+ "\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f"
+ "\x60\x61\x62\x63\x64\x65\x66\x67"
+ "\x68\x69\x6a\x6b\x6c\x6d\x6e\x6f"
+ "\x70\x71\x72\x73\x74\x75\x76\x77"
+ "\x78\x79\x7a\x7b\x7c\x7d\x7e\x7f"
+ "\x80\x81\x82\x83\x84\x85\x86\x87"
+ "\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
+ "\x90\x91\x92\x93\x94\x95\x96\x97"
+ "\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
+ "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7"
+ "\xa8\xa9\xaa\xab\xac\xad\xae\xaf"
+ "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7"
+ "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf"
+ "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7"
+ "\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
+ "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7"
+ "\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf"
+ "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7"
+ "\xe8\xe9\xea\xeb\xec\xed\xee\xef"
+ "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7"
+ "\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff"
+ "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17"
+ "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f"
+ "\x20\x21\x22\x23\x24\x25\x26\x27"
+ "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f"
+ "\x30\x31\x32\x33\x34\x35\x36\x37"
+ "\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f"
+ "\x40\x41\x42\x43\x44\x45\x46\x47"
+ "\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f"
+ "\x50\x51\x52\x53\x54\x55\x56\x57"
+ "\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f"
+ "\x60\x61\x62\x63\x64\x65\x66\x67"
+ "\x68\x69\x6a\x6b\x6c\x6d\x6e\x6f"
+ "\x70\x71\x72\x73\x74\x75\x76\x77"
+ "\x78\x79\x7a\x7b\x7c\x7d\x7e\x7f"
+ "\x80\x81\x82\x83\x84\x85\x86\x87"
+ "\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
+ "\x90\x91\x92\x93\x94\x95\x96\x97"
+ "\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
+ "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7"
+ "\xa8\xa9\xaa\xab\xac\xad\xae\xaf"
+ "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7"
+ "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf"
+ "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7"
+ "\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
+ "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7"
+ "\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf"
+ "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7"
+ "\xe8\xe9\xea\xeb\xec\xed\xee\xef"
+ "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7"
+ "\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff",
+ .ilen = 512,
+ .result = "\x57\xb5\xf8\x71\x6e\x6d\xdd\x82"
+ "\x53\xd0\xed\x2d\x30\xc1\x20\xef"
+ "\x70\x67\x5e\xff\x09\x70\xbb\xc1"
+ "\x3a\x7b\x48\x26\xd9\x0b\xf4\x48"
+ "\xbe\xce\xb1\xc7\xb2\x67\xc4\xa7"
+ "\x76\xf8\x36\x30\xb7\xb4\x9a\xd9"
+ "\xf5\x9d\xd0\x7b\xc1\x06\x96\x44"
+ "\x19\xc5\x58\x84\x63\xb9\x12\x68"
+ "\x68\xc7\xaa\x18\x98\xf2\x1f\x5c"
+ "\x39\xa6\xd8\x32\x2b\xc3\x51\xfd"
+ "\x74\x79\x2e\xb4\x44\xd7\x69\xc4"
+ "\xfc\x29\xe6\xed\x26\x1e\xa6\x9d"
+ "\x1c\xbe\x00\x0e\x7f\x3a\xca\xfb"
+ "\x6d\x13\x65\xa0\xf9\x31\x12\xe2"
+ "\x26\xd1\xec\x2b\x0a\x8b\x59\x99"
+ "\xa7\x49\xa0\x0e\x09\x33\x85\x50"
+ "\xc3\x23\xca\x7a\xdd\x13\x45\x5f"
+ "\xde\x4c\xa7\xcb\x00\x8a\x66\x6f"
+ "\xa2\xb6\xb1\x2e\xe1\xa0\x18\xf6"
+ "\xad\xf3\xbd\xeb\xc7\xef\x55\x4f"
+ "\x79\x91\x8d\x36\x13\x7b\xd0\x4a"
+ "\x6c\x39\xfb\x53\xb8\x6f\x02\x51"
+ "\xa5\x20\xac\x24\x1c\x73\x59\x73"
+ "\x58\x61\x3a\x87\x58\xb3\x20\x56"
+ "\x39\x06\x2b\x4d\xd3\x20\x2b\x89"
+ "\x3f\xa2\xf0\x96\xeb\x7f\xa4\xcd"
+ "\x11\xae\xbd\xcb\x3a\xb4\xd9\x91"
+ "\x09\x35\x71\x50\x65\xac\x92\xe3"
+ "\x7b\x32\xc0\x7a\xdd\xd4\xc3\x92"
+ "\x6f\xeb\x79\xde\x6f\xd3\x25\xc9"
+ "\xcd\x63\xf5\x1e\x7a\x3b\x26\x9d"
+ "\x77\x04\x80\xa9\xbf\x38\xb5\xbd"
+ "\xb8\x05\x07\xbd\xfd\xab\x7b\xf8"
+ "\x2a\x26\xcc\x49\x14\x6d\x55\x01"
+ "\x06\x94\xd8\xb2\x2d\x53\x83\x1b"
+ "\x8f\xd4\xdd\x57\x12\x7e\x18\xba"
+ "\x8e\xe2\x4d\x80\xef\x7e\x6b\x9d"
+ "\x24\xa9\x60\xa4\x97\x85\x86\x2a"
+ "\x01\x00\x09\xf1\xcb\x4a\x24\x1c"
+ "\xd8\xf6\xe6\x5b\xe7\x5d\xf2\xc4"
+ "\x97\x1c\x10\xc6\x4d\x66\x4f\x98"
+ "\x87\x30\xac\xd5\xea\x73\x49\x10"
+ "\x80\xea\xe5\x5f\x4d\x5f\x03\x33"
+ "\x66\x02\x35\x3d\x60\x06\x36\x4f"
+ "\x14\x1c\xd8\x07\x1f\x78\xd0\xf8"
+ "\x4f\x6c\x62\x7c\x15\xa5\x7c\x28"
+ "\x7c\xcc\xeb\x1f\xd1\x07\x90\x93"
+ "\x7e\xc2\xa8\x3a\x80\xc0\xf5\x30"
+ "\xcc\x75\xcf\x16\x26\xa9\x26\x3b"
+ "\xe7\x68\x2f\x15\x21\x5b\xe4\x00"
+ "\xbd\x48\x50\xcd\x75\x70\xc4\x62"
+ "\xbb\x41\xfb\x89\x4a\x88\x3b\x3b"
+ "\x51\x66\x02\x69\x04\x97\x36\xd4"
+ "\x75\xae\x0b\xa3\x42\xf8\xca\x79"
+ "\x8f\x93\xe9\xcc\x38\xbd\xd6\xd2"
+ "\xf9\x70\x4e\xc3\x6a\x8e\x25\xbd"
+ "\xea\x15\x5a\xa0\x85\x7e\x81\x0d"
+ "\x03\xe7\x05\x39\xf5\x05\x26\xee"
+ "\xec\xaa\x1f\x3d\xc9\x98\x76\x01"
+ "\x2c\xf4\xfc\xa3\x88\x77\x38\xc4"
+ "\x50\x65\x50\x6d\x04\x1f\xdf\x5a"
+ "\xaa\xf2\x01\xa9\xc1\x8d\xee\xca"
+ "\x47\x26\xef\x39\xb8\xb4\xf2\xd1"
+ "\xd6\xbb\x1b\x2a\xc1\x34\x14\xcf",
+ .rlen = 512,
+ }, {
+ .key = "\x27\x18\x28\x18\x28\x45\x90\x45"
+ "\x23\x53\x60\x28\x74\x71\x35\x26"
+ "\x62\x49\x77\x57\x24\x70\x93\x69"
+ "\x99\x59\x57\x49\x66\x96\x76\x27"
+ "\x31\x41\x59\x26\x53\x58\x97\x93"
+ "\x23\x84\x62\x64\x33\x83\x27\x95"
+ "\x02\x88\x41\x97\x16\x93\x99\x37"
+ "\x51\x05\x82\x09\x74\x94\x45\x92",
+ .klen = 64,
+ .iv = "\xff\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17"
+ "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f"
+ "\x20\x21\x22\x23\x24\x25\x26\x27"
+ "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f"
+ "\x30\x31\x32\x33\x34\x35\x36\x37"
+ "\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f"
+ "\x40\x41\x42\x43\x44\x45\x46\x47"
+ "\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f"
+ "\x50\x51\x52\x53\x54\x55\x56\x57"
+ "\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f"
+ "\x60\x61\x62\x63\x64\x65\x66\x67"
+ "\x68\x69\x6a\x6b\x6c\x6d\x6e\x6f"
+ "\x70\x71\x72\x73\x74\x75\x76\x77"
+ "\x78\x79\x7a\x7b\x7c\x7d\x7e\x7f"
+ "\x80\x81\x82\x83\x84\x85\x86\x87"
+ "\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
+ "\x90\x91\x92\x93\x94\x95\x96\x97"
+ "\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
+ "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7"
+ "\xa8\xa9\xaa\xab\xac\xad\xae\xaf"
+ "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7"
+ "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf"
+ "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7"
+ "\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
+ "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7"
+ "\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf"
+ "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7"
+ "\xe8\xe9\xea\xeb\xec\xed\xee\xef"
+ "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7"
+ "\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff"
+ "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17"
+ "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f"
+ "\x20\x21\x22\x23\x24\x25\x26\x27"
+ "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f"
+ "\x30\x31\x32\x33\x34\x35\x36\x37"
+ "\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f"
+ "\x40\x41\x42\x43\x44\x45\x46\x47"
+ "\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f"
+ "\x50\x51\x52\x53\x54\x55\x56\x57"
+ "\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f"
+ "\x60\x61\x62\x63\x64\x65\x66\x67"
+ "\x68\x69\x6a\x6b\x6c\x6d\x6e\x6f"
+ "\x70\x71\x72\x73\x74\x75\x76\x77"
+ "\x78\x79\x7a\x7b\x7c\x7d\x7e\x7f"
+ "\x80\x81\x82\x83\x84\x85\x86\x87"
+ "\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
+ "\x90\x91\x92\x93\x94\x95\x96\x97"
+ "\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
+ "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7"
+ "\xa8\xa9\xaa\xab\xac\xad\xae\xaf"
+ "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7"
+ "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf"
+ "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7"
+ "\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
+ "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7"
+ "\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf"
+ "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7"
+ "\xe8\xe9\xea\xeb\xec\xed\xee\xef"
+ "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7"
+ "\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff",
+ .ilen = 512,
+ .result = "\xc5\x85\x2a\x4b\x73\xe4\xf6\xf1"
+ "\x7e\xf9\xf6\xe9\xa3\x73\x36\xcb"
+ "\xaa\xb6\x22\xb0\x24\x6e\x3d\x73"
+ "\x92\x99\xde\xd3\x76\xed\xcd\x63"
+ "\x64\x3a\x22\x57\xc1\x43\x49\xd4"
+ "\x79\x36\x31\x19\x62\xae\x10\x7e"
+ "\x7d\xcf\x7a\xe2\x6b\xce\x27\xfa"
+ "\xdc\x3d\xd9\x83\xd3\x42\x4c\xe0"
+ "\x1b\xd6\x1d\x1a\x6f\xd2\x03\x00"
+ "\xfc\x81\x99\x8a\x14\x62\xf5\x7e"
+ "\x0d\xe7\x12\xe8\x17\x9d\x0b\xec"
+ "\xe2\xf7\xc9\xa7\x63\xd1\x79\xb6"
+ "\x62\x62\x37\xfe\x0a\x4c\x4a\x37"
+ "\x70\xc7\x5e\x96\x5f\xbc\x8e\x9e"
+ "\x85\x3c\x4f\x26\x64\x85\xbc\x68"
+ "\xb0\xe0\x86\x5e\x26\x41\xce\x11"
+ "\x50\xda\x97\x14\xe9\x9e\xc7\x6d"
+ "\x3b\xdc\x43\xde\x2b\x27\x69\x7d"
+ "\xfc\xb0\x28\xbd\x8f\xb1\xc6\x31"
+ "\x14\x4d\xf0\x74\x37\xfd\x07\x25"
+ "\x96\x55\xe5\xfc\x9e\x27\x2a\x74"
+ "\x1b\x83\x4d\x15\x83\xac\x57\xa0"
+ "\xac\xa5\xd0\x38\xef\x19\x56\x53"
+ "\x25\x4b\xfc\xce\x04\x23\xe5\x6b"
+ "\xf6\xc6\x6c\x32\x0b\xb3\x12\xc5"
+ "\xed\x22\x34\x1c\x5d\xed\x17\x06"
+ "\x36\xa3\xe6\x77\xb9\x97\x46\xb8"
+ "\xe9\x3f\x7e\xc7\xbc\x13\x5c\xdc"
+ "\x6e\x3f\x04\x5e\xd1\x59\xa5\x82"
+ "\x35\x91\x3d\x1b\xe4\x97\x9f\x92"
+ "\x1c\x5e\x5f\x6f\x41\xd4\x62\xa1"
+ "\x8d\x39\xfc\x42\xfb\x38\x80\xb9"
+ "\x0a\xe3\xcc\x6a\x93\xd9\x7a\xb1"
+ "\xe9\x69\xaf\x0a\x6b\x75\x38\xa7"
+ "\xa1\xbf\xf7\xda\x95\x93\x4b\x78"
+ "\x19\xf5\x94\xf9\xd2\x00\x33\x37"
+ "\xcf\xf5\x9e\x9c\xf3\xcc\xa6\xee"
+ "\x42\xb2\x9e\x2c\x5f\x48\x23\x26"
+ "\x15\x25\x17\x03\x3d\xfe\x2c\xfc"
+ "\xeb\xba\xda\xe0\x00\x05\xb6\xa6"
+ "\x07\xb3\xe8\x36\x5b\xec\x5b\xbf"
+ "\xd6\x5b\x00\x74\xc6\x97\xf1\x6a"
+ "\x49\xa1\xc3\xfa\x10\x52\xb9\x14"
+ "\xad\xb7\x73\xf8\x78\x12\xc8\x59"
+ "\x17\x80\x4c\x57\x39\xf1\x6d\x80"
+ "\x25\x77\x0f\x5e\x7d\xf0\xaf\x21"
+ "\xec\xce\xb7\xc8\x02\x8a\xed\x53"
+ "\x2c\x25\x68\x2e\x1f\x85\x5e\x67"
+ "\xd1\x07\x7a\x3a\x89\x08\xe0\x34"
+ "\xdc\xdb\x26\xb4\x6b\x77\xfc\x40"
+ "\x31\x15\x72\xa0\xf0\x73\xd9\x3b"
+ "\xd5\xdb\xfe\xfc\x8f\xa9\x44\xa2"
+ "\x09\x9f\xc6\x33\xe5\xe2\x88\xe8"
+ "\xf3\xf0\x1a\xf4\xce\x12\x0f\xd6"
+ "\xf7\x36\xe6\xa4\xf4\x7a\x10\x58"
+ "\xcc\x1f\x48\x49\x65\x47\x75\xe9"
+ "\x28\xe1\x65\x7b\xf2\xc4\xb5\x07"
+ "\xf2\xec\x76\xd8\x8f\x09\xf3\x16"
+ "\xa1\x51\x89\x3b\xeb\x96\x42\xac"
+ "\x65\xe0\x67\x63\x29\xdc\xb4\x7d"
+ "\xf2\x41\x51\x6a\xcb\xde\x3c\xfb"
+ "\x66\x8d\x13\xca\xe0\x59\x2a\x00"
+ "\xc9\x53\x4c\xe6\x9e\xe2\x73\xd5"
+ "\x67\x19\xb2\xbd\x9a\x63\xd7\x5c",
+ .rlen = 512,
+ .also_non_np = 1,
+ .np = 3,
+ .tap = { 512 - 20, 4, 16 },
+ }
+};
+
+static const struct cipher_testvec speck128_xts_dec_tv_template[] = {
+ {
+ .key = "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .klen = 32,
+ .iv = "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\xbe\xa0\xe7\x03\xd7\xfe\xab\x62"
+ "\x3b\x99\x4a\x64\x74\x77\xac\xed"
+ "\xd8\xf4\xa6\xcf\xae\xb9\x07\x42"
+ "\x51\xd9\xb6\x1d\xe0\x5e\xbc\x54",
+ .ilen = 32,
+ .result = "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .rlen = 32,
+ }, {
+ .key = "\x11\x11\x11\x11\x11\x11\x11\x11"
+ "\x11\x11\x11\x11\x11\x11\x11\x11"
+ "\x22\x22\x22\x22\x22\x22\x22\x22"
+ "\x22\x22\x22\x22\x22\x22\x22\x22",
+ .klen = 32,
+ .iv = "\x33\x33\x33\x33\x33\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\xfb\x53\x81\x75\x6f\x9f\x34\xad"
+ "\x7e\x01\xed\x7b\xcc\xda\x4e\x4a"
+ "\xd4\x84\xa4\x53\xd5\x88\x73\x1b"
+ "\xfd\xcb\xae\x0d\xf3\x04\xee\xe6",
+ .ilen = 32,
+ .result = "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44",
+ .rlen = 32,
+ }, {
+ .key = "\xff\xfe\xfd\xfc\xfb\xfa\xf9\xf8"
+ "\xf7\xf6\xf5\xf4\xf3\xf2\xf1\xf0"
+ "\x22\x22\x22\x22\x22\x22\x22\x22"
+ "\x22\x22\x22\x22\x22\x22\x22\x22",
+ .klen = 32,
+ .iv = "\x33\x33\x33\x33\x33\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\x21\x52\x84\x15\xd1\xf7\x21\x55"
+ "\xd9\x75\x4a\xd3\xc5\xdb\x9f\x7d"
+ "\xda\x63\xb2\xf1\x82\xb0\x89\x59"
+ "\x86\xd4\xaa\xaa\xdd\xff\x4f\x92",
+ .ilen = 32,
+ .result = "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44"
+ "\x44\x44\x44\x44\x44\x44\x44\x44",
+ .rlen = 32,
+ }, {
+ .key = "\x27\x18\x28\x18\x28\x45\x90\x45"
+ "\x23\x53\x60\x28\x74\x71\x35\x26"
+ "\x31\x41\x59\x26\x53\x58\x97\x93"
+ "\x23\x84\x62\x64\x33\x83\x27\x95",
+ .klen = 32,
+ .iv = "\x00\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\x57\xb5\xf8\x71\x6e\x6d\xdd\x82"
+ "\x53\xd0\xed\x2d\x30\xc1\x20\xef"
+ "\x70\x67\x5e\xff\x09\x70\xbb\xc1"
+ "\x3a\x7b\x48\x26\xd9\x0b\xf4\x48"
+ "\xbe\xce\xb1\xc7\xb2\x67\xc4\xa7"
+ "\x76\xf8\x36\x30\xb7\xb4\x9a\xd9"
+ "\xf5\x9d\xd0\x7b\xc1\x06\x96\x44"
+ "\x19\xc5\x58\x84\x63\xb9\x12\x68"
+ "\x68\xc7\xaa\x18\x98\xf2\x1f\x5c"
+ "\x39\xa6\xd8\x32\x2b\xc3\x51\xfd"
+ "\x74\x79\x2e\xb4\x44\xd7\x69\xc4"
+ "\xfc\x29\xe6\xed\x26\x1e\xa6\x9d"
+ "\x1c\xbe\x00\x0e\x7f\x3a\xca\xfb"
+ "\x6d\x13\x65\xa0\xf9\x31\x12\xe2"
+ "\x26\xd1\xec\x2b\x0a\x8b\x59\x99"
+ "\xa7\x49\xa0\x0e\x09\x33\x85\x50"
+ "\xc3\x23\xca\x7a\xdd\x13\x45\x5f"
+ "\xde\x4c\xa7\xcb\x00\x8a\x66\x6f"
+ "\xa2\xb6\xb1\x2e\xe1\xa0\x18\xf6"
+ "\xad\xf3\xbd\xeb\xc7\xef\x55\x4f"
+ "\x79\x91\x8d\x36\x13\x7b\xd0\x4a"
+ "\x6c\x39\xfb\x53\xb8\x6f\x02\x51"
+ "\xa5\x20\xac\x24\x1c\x73\x59\x73"
+ "\x58\x61\x3a\x87\x58\xb3\x20\x56"
+ "\x39\x06\x2b\x4d\xd3\x20\x2b\x89"
+ "\x3f\xa2\xf0\x96\xeb\x7f\xa4\xcd"
+ "\x11\xae\xbd\xcb\x3a\xb4\xd9\x91"
+ "\x09\x35\x71\x50\x65\xac\x92\xe3"
+ "\x7b\x32\xc0\x7a\xdd\xd4\xc3\x92"
+ "\x6f\xeb\x79\xde\x6f\xd3\x25\xc9"
+ "\xcd\x63\xf5\x1e\x7a\x3b\x26\x9d"
+ "\x77\x04\x80\xa9\xbf\x38\xb5\xbd"
+ "\xb8\x05\x07\xbd\xfd\xab\x7b\xf8"
+ "\x2a\x26\xcc\x49\x14\x6d\x55\x01"
+ "\x06\x94\xd8\xb2\x2d\x53\x83\x1b"
+ "\x8f\xd4\xdd\x57\x12\x7e\x18\xba"
+ "\x8e\xe2\x4d\x80\xef\x7e\x6b\x9d"
+ "\x24\xa9\x60\xa4\x97\x85\x86\x2a"
+ "\x01\x00\x09\xf1\xcb\x4a\x24\x1c"
+ "\xd8\xf6\xe6\x5b\xe7\x5d\xf2\xc4"
+ "\x97\x1c\x10\xc6\x4d\x66\x4f\x98"
+ "\x87\x30\xac\xd5\xea\x73\x49\x10"
+ "\x80\xea\xe5\x5f\x4d\x5f\x03\x33"
+ "\x66\x02\x35\x3d\x60\x06\x36\x4f"
+ "\x14\x1c\xd8\x07\x1f\x78\xd0\xf8"
+ "\x4f\x6c\x62\x7c\x15\xa5\x7c\x28"
+ "\x7c\xcc\xeb\x1f\xd1\x07\x90\x93"
+ "\x7e\xc2\xa8\x3a\x80\xc0\xf5\x30"
+ "\xcc\x75\xcf\x16\x26\xa9\x26\x3b"
+ "\xe7\x68\x2f\x15\x21\x5b\xe4\x00"
+ "\xbd\x48\x50\xcd\x75\x70\xc4\x62"
+ "\xbb\x41\xfb\x89\x4a\x88\x3b\x3b"
+ "\x51\x66\x02\x69\x04\x97\x36\xd4"
+ "\x75\xae\x0b\xa3\x42\xf8\xca\x79"
+ "\x8f\x93\xe9\xcc\x38\xbd\xd6\xd2"
+ "\xf9\x70\x4e\xc3\x6a\x8e\x25\xbd"
+ "\xea\x15\x5a\xa0\x85\x7e\x81\x0d"
+ "\x03\xe7\x05\x39\xf5\x05\x26\xee"
+ "\xec\xaa\x1f\x3d\xc9\x98\x76\x01"
+ "\x2c\xf4\xfc\xa3\x88\x77\x38\xc4"
+ "\x50\x65\x50\x6d\x04\x1f\xdf\x5a"
+ "\xaa\xf2\x01\xa9\xc1\x8d\xee\xca"
+ "\x47\x26\xef\x39\xb8\xb4\xf2\xd1"
+ "\xd6\xbb\x1b\x2a\xc1\x34\x14\xcf",
+ .ilen = 512,
+ .result = "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17"
+ "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f"
+ "\x20\x21\x22\x23\x24\x25\x26\x27"
+ "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f"
+ "\x30\x31\x32\x33\x34\x35\x36\x37"
+ "\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f"
+ "\x40\x41\x42\x43\x44\x45\x46\x47"
+ "\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f"
+ "\x50\x51\x52\x53\x54\x55\x56\x57"
+ "\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f"
+ "\x60\x61\x62\x63\x64\x65\x66\x67"
+ "\x68\x69\x6a\x6b\x6c\x6d\x6e\x6f"
+ "\x70\x71\x72\x73\x74\x75\x76\x77"
+ "\x78\x79\x7a\x7b\x7c\x7d\x7e\x7f"
+ "\x80\x81\x82\x83\x84\x85\x86\x87"
+ "\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
+ "\x90\x91\x92\x93\x94\x95\x96\x97"
+ "\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
+ "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7"
+ "\xa8\xa9\xaa\xab\xac\xad\xae\xaf"
+ "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7"
+ "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf"
+ "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7"
+ "\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
+ "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7"
+ "\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf"
+ "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7"
+ "\xe8\xe9\xea\xeb\xec\xed\xee\xef"
+ "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7"
+ "\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff"
+ "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17"
+ "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f"
+ "\x20\x21\x22\x23\x24\x25\x26\x27"
+ "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f"
+ "\x30\x31\x32\x33\x34\x35\x36\x37"
+ "\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f"
+ "\x40\x41\x42\x43\x44\x45\x46\x47"
+ "\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f"
+ "\x50\x51\x52\x53\x54\x55\x56\x57"
+ "\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f"
+ "\x60\x61\x62\x63\x64\x65\x66\x67"
+ "\x68\x69\x6a\x6b\x6c\x6d\x6e\x6f"
+ "\x70\x71\x72\x73\x74\x75\x76\x77"
+ "\x78\x79\x7a\x7b\x7c\x7d\x7e\x7f"
+ "\x80\x81\x82\x83\x84\x85\x86\x87"
+ "\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
+ "\x90\x91\x92\x93\x94\x95\x96\x97"
+ "\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
+ "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7"
+ "\xa8\xa9\xaa\xab\xac\xad\xae\xaf"
+ "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7"
+ "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf"
+ "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7"
+ "\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
+ "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7"
+ "\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf"
+ "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7"
+ "\xe8\xe9\xea\xeb\xec\xed\xee\xef"
+ "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7"
+ "\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff",
+ .rlen = 512,
+ }, {
+ .key = "\x27\x18\x28\x18\x28\x45\x90\x45"
+ "\x23\x53\x60\x28\x74\x71\x35\x26"
+ "\x62\x49\x77\x57\x24\x70\x93\x69"
+ "\x99\x59\x57\x49\x66\x96\x76\x27"
+ "\x31\x41\x59\x26\x53\x58\x97\x93"
+ "\x23\x84\x62\x64\x33\x83\x27\x95"
+ "\x02\x88\x41\x97\x16\x93\x99\x37"
+ "\x51\x05\x82\x09\x74\x94\x45\x92",
+ .klen = 64,
+ .iv = "\xff\x00\x00\x00\x00\x00\x00\x00"
+ "\x00\x00\x00\x00\x00\x00\x00\x00",
+ .input = "\xc5\x85\x2a\x4b\x73\xe4\xf6\xf1"
+ "\x7e\xf9\xf6\xe9\xa3\x73\x36\xcb"
+ "\xaa\xb6\x22\xb0\x24\x6e\x3d\x73"
+ "\x92\x99\xde\xd3\x76\xed\xcd\x63"
+ "\x64\x3a\x22\x57\xc1\x43\x49\xd4"
+ "\x79\x36\x31\x19\x62\xae\x10\x7e"
+ "\x7d\xcf\x7a\xe2\x6b\xce\x27\xfa"
+ "\xdc\x3d\xd9\x83\xd3\x42\x4c\xe0"
+ "\x1b\xd6\x1d\x1a\x6f\xd2\x03\x00"
+ "\xfc\x81\x99\x8a\x14\x62\xf5\x7e"
+ "\x0d\xe7\x12\xe8\x17\x9d\x0b\xec"
+ "\xe2\xf7\xc9\xa7\x63\xd1\x79\xb6"
+ "\x62\x62\x37\xfe\x0a\x4c\x4a\x37"
+ "\x70\xc7\x5e\x96\x5f\xbc\x8e\x9e"
+ "\x85\x3c\x4f\x26\x64\x85\xbc\x68"
+ "\xb0\xe0\x86\x5e\x26\x41\xce\x11"
+ "\x50\xda\x97\x14\xe9\x9e\xc7\x6d"
+ "\x3b\xdc\x43\xde\x2b\x27\x69\x7d"
+ "\xfc\xb0\x28\xbd\x8f\xb1\xc6\x31"
+ "\x14\x4d\xf0\x74\x37\xfd\x07\x25"
+ "\x96\x55\xe5\xfc\x9e\x27\x2a\x74"
+ "\x1b\x83\x4d\x15\x83\xac\x57\xa0"
+ "\xac\xa5\xd0\x38\xef\x19\x56\x53"
+ "\x25\x4b\xfc\xce\x04\x23\xe5\x6b"
+ "\xf6\xc6\x6c\x32\x0b\xb3\x12\xc5"
+ "\xed\x22\x34\x1c\x5d\xed\x17\x06"
+ "\x36\xa3\xe6\x77\xb9\x97\x46\xb8"
+ "\xe9\x3f\x7e\xc7\xbc\x13\x5c\xdc"
+ "\x6e\x3f\x04\x5e\xd1\x59\xa5\x82"
+ "\x35\x91\x3d\x1b\xe4\x97\x9f\x92"
+ "\x1c\x5e\x5f\x6f\x41\xd4\x62\xa1"
+ "\x8d\x39\xfc\x42\xfb\x38\x80\xb9"
+ "\x0a\xe3\xcc\x6a\x93\xd9\x7a\xb1"
+ "\xe9\x69\xaf\x0a\x6b\x75\x38\xa7"
+ "\xa1\xbf\xf7\xda\x95\x93\x4b\x78"
+ "\x19\xf5\x94\xf9\xd2\x00\x33\x37"
+ "\xcf\xf5\x9e\x9c\xf3\xcc\xa6\xee"
+ "\x42\xb2\x9e\x2c\x5f\x48\x23\x26"
+ "\x15\x25\x17\x03\x3d\xfe\x2c\xfc"
+ "\xeb\xba\xda\xe0\x00\x05\xb6\xa6"
+ "\x07\xb3\xe8\x36\x5b\xec\x5b\xbf"
+ "\xd6\x5b\x00\x74\xc6\x97\xf1\x6a"
+ "\x49\xa1\xc3\xfa\x10\x52\xb9\x14"
+ "\xad\xb7\x73\xf8\x78\x12\xc8\x59"
+ "\x17\x80\x4c\x57\x39\xf1\x6d\x80"
+ "\x25\x77\x0f\x5e\x7d\xf0\xaf\x21"
+ "\xec\xce\xb7\xc8\x02\x8a\xed\x53"
+ "\x2c\x25\x68\x2e\x1f\x85\x5e\x67"
+ "\xd1\x07\x7a\x3a\x89\x08\xe0\x34"
+ "\xdc\xdb\x26\xb4\x6b\x77\xfc\x40"
+ "\x31\x15\x72\xa0\xf0\x73\xd9\x3b"
+ "\xd5\xdb\xfe\xfc\x8f\xa9\x44\xa2"
+ "\x09\x9f\xc6\x33\xe5\xe2\x88\xe8"
+ "\xf3\xf0\x1a\xf4\xce\x12\x0f\xd6"
+ "\xf7\x36\xe6\xa4\xf4\x7a\x10\x58"
+ "\xcc\x1f\x48\x49\x65\x47\x75\xe9"
+ "\x28\xe1\x65\x7b\xf2\xc4\xb5\x07"
+ "\xf2\xec\x76\xd8\x8f\x09\xf3\x16"
+ "\xa1\x51\x89\x3b\xeb\x96\x42\xac"
+ "\x65\xe0\x67\x63\x29\xdc\xb4\x7d"
+ "\xf2\x41\x51\x6a\xcb\xde\x3c\xfb"
+ "\x66\x8d\x13\xca\xe0\x59\x2a\x00"
+ "\xc9\x53\x4c\xe6\x9e\xe2\x73\xd5"
+ "\x67\x19\xb2\xbd\x9a\x63\xd7\x5c",
+ .ilen = 512,
+ .result = "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17"
+ "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f"
+ "\x20\x21\x22\x23\x24\x25\x26\x27"
+ "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f"
+ "\x30\x31\x32\x33\x34\x35\x36\x37"
+ "\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f"
+ "\x40\x41\x42\x43\x44\x45\x46\x47"
+ "\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f"
+ "\x50\x51\x52\x53\x54\x55\x56\x57"
+ "\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f"
+ "\x60\x61\x62\x63\x64\x65\x66\x67"
+ "\x68\x69\x6a\x6b\x6c\x6d\x6e\x6f"
+ "\x70\x71\x72\x73\x74\x75\x76\x77"
+ "\x78\x79\x7a\x7b\x7c\x7d\x7e\x7f"
+ "\x80\x81\x82\x83\x84\x85\x86\x87"
+ "\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
+ "\x90\x91\x92\x93\x94\x95\x96\x97"
+ "\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
+ "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7"
+ "\xa8\xa9\xaa\xab\xac\xad\xae\xaf"
+ "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7"
+ "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf"
+ "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7"
+ "\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
+ "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7"
+ "\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf"
+ "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7"
+ "\xe8\xe9\xea\xeb\xec\xed\xee\xef"
+ "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7"
+ "\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff"
+ "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17"
+ "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f"
+ "\x20\x21\x22\x23\x24\x25\x26\x27"
+ "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f"
+ "\x30\x31\x32\x33\x34\x35\x36\x37"
+ "\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f"
+ "\x40\x41\x42\x43\x44\x45\x46\x47"
+ "\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f"
+ "\x50\x51\x52\x53\x54\x55\x56\x57"
+ "\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f"
+ "\x60\x61\x62\x63\x64\x65\x66\x67"
+ "\x68\x69\x6a\x6b\x6c\x6d\x6e\x6f"
+ "\x70\x71\x72\x73\x74\x75\x76\x77"
+ "\x78\x79\x7a\x7b\x7c\x7d\x7e\x7f"
+ "\x80\x81\x82\x83\x84\x85\x86\x87"
+ "\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
+ "\x90\x91\x92\x93\x94\x95\x96\x97"
+ "\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
+ "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7"
+ "\xa8\xa9\xaa\xab\xac\xad\xae\xaf"
+ "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7"
+ "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf"
+ "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7"
+ "\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf"
+ "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7"
+ "\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf"
+ "\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7"
+ "\xe8\xe9\xea\xeb\xec\xed\xee\xef"
+ "\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7"
+ "\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff",
+ .rlen = 512,
+ .also_non_np = 1,
+ .np = 3,
+ .tap = { 512 - 20, 4, 16 },
+ }
+};
+
static const struct cipher_testvec speck64_enc_tv_template[] = {
{ /* Speck64/96 */
.key = "\x00\x01\x02\x03\x08\x09\x0a\x0b"
--
2.16.1.291.g4437f3f132-goog
Add a generic implementation of Speck, including the Speck128 and
Speck64 variants. Speck is a lightweight block cipher that can be much
faster than AES on processors that don't have AES instructions.
We are planning to offer Speck-XTS (probably Speck128/256-XTS) as an
option for dm-crypt and fscrypt on Android, for low-end mobile devices
with older CPUs such as ARMv7 which don't have the Cryptography
Extensions. Currently, such devices are unencrypted because AES is not
fast enough, even when the NEON bit-sliced implementation of AES is
used. Other AES alternatives such as Twofish, Threefish, Camellia,
CAST6, and Serpent aren't fast enough either; it seems that only a
modern ARX cipher can provide sufficient performance on these devices.
This is a replacement for our original proposal
(https://patchwork.kernel.org/patch/10101451/) which was to offer
ChaCha20 for these devices. However, the use of a stream cipher for
disk/file encryption with no space to store nonces would have been much
more insecure than we thought initially, given that it would be used on
top of flash storage as well as potentially on top of F2FS, neither of
which is guaranteed to overwrite data in-place.
Speck has been somewhat controversial due to its origin. Nevertheless,
it has a straightforward design (it's an ARX cipher), and it appears to
be the leading software-optimized lightweight block cipher currently,
with the most cryptanalysis. It's also easy to implement without side
channels, unlike AES. Moreover, we only intend Speck to be used when
the status quo is no encryption, due to AES not being fast enough.
We've also considered a novel length-preserving encryption mode based on
ChaCha20 and Poly1305. While theoretically attractive, such a mode
would be a brand new crypto construction and would be more complicated
and difficult to implement efficiently in comparison to Speck-XTS.
There is confusion about the byte and word orders of Speck, since the
original paper doesn't specify them. But we have implemented it using
the orders the authors recommended in a correspondence with them. The
test vectors are taken from the original paper but were mapped to byte
arrays using the recommended byte and word orders.
Signed-off-by: Eric Biggers <[email protected]>
---
crypto/Kconfig | 14 +++
crypto/Makefile | 1 +
crypto/speck.c | 299 +++++++++++++++++++++++++++++++++++++++++++++++
crypto/testmgr.c | 18 +++
crypto/testmgr.h | 128 ++++++++++++++++++++
5 files changed, 460 insertions(+)
create mode 100644 crypto/speck.c
diff --git a/crypto/Kconfig b/crypto/Kconfig
index b75264b09a46..558eff07b799 100644
--- a/crypto/Kconfig
+++ b/crypto/Kconfig
@@ -1508,6 +1508,20 @@ config CRYPTO_SERPENT_AVX2_X86_64
See also:
<http://www.cl.cam.ac.uk/~rja14/serpent.html>
+config CRYPTO_SPECK
+ tristate "Speck cipher algorithm"
+ select CRYPTO_ALGAPI
+ help
+ Speck is a lightweight block cipher that is tuned for optimal
+ performance in software (rather than hardware).
+
+ Speck may not be as secure as AES, and should only be used on systems
+ where AES is not fast enough.
+
+ See also: <https://eprint.iacr.org/2013/404.pdf>
+
+ If unsure, say N.
+
config CRYPTO_TEA
tristate "TEA, XTEA and XETA cipher algorithms"
select CRYPTO_ALGAPI
diff --git a/crypto/Makefile b/crypto/Makefile
index cdbc03b35510..ba6019471447 100644
--- a/crypto/Makefile
+++ b/crypto/Makefile
@@ -110,6 +110,7 @@ obj-$(CONFIG_CRYPTO_TEA) += tea.o
obj-$(CONFIG_CRYPTO_KHAZAD) += khazad.o
obj-$(CONFIG_CRYPTO_ANUBIS) += anubis.o
obj-$(CONFIG_CRYPTO_SEED) += seed.o
+obj-$(CONFIG_CRYPTO_SPECK) += speck.o
obj-$(CONFIG_CRYPTO_SALSA20) += salsa20_generic.o
obj-$(CONFIG_CRYPTO_CHACHA20) += chacha20_generic.o
obj-$(CONFIG_CRYPTO_POLY1305) += poly1305_generic.o
diff --git a/crypto/speck.c b/crypto/speck.c
new file mode 100644
index 000000000000..4e80ad76bcd7
--- /dev/null
+++ b/crypto/speck.c
@@ -0,0 +1,299 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Speck: a lightweight block cipher
+ *
+ * Copyright (c) 2018 Google, Inc
+ *
+ * Speck has 10 variants, including 5 block sizes. For now we only implement
+ * the variants Speck128/128, Speck128/192, Speck128/256, Speck64/96, and
+ * Speck64/128. Speck${B}/${K} denotes the variant with a block size of B bits
+ * and a key size of K bits. The Speck128 variants are believed to be the most
+ * secure variants, and they use the same block size and key sizes as AES. The
+ * Speck64 variants are less secure, but on 32-bit processors are usually
+ * faster. The remaining variants (Speck32, Speck48, and Speck96) are even less
+ * secure and/or not as well suited for implementation on either 32-bit or
+ * 64-bit processors, so are omitted.
+ *
+ * Reference: "The Simon and Speck Families of Lightweight Block Ciphers"
+ * https://eprint.iacr.org/2013/404.pdf
+ *
+ * In a correspondence, the Speck designers have also clarified that the words
+ * should be interpreted in little-endian format, and the words should be
+ * ordered such that the first word of each block is 'y' rather than 'x', and
+ * the first key word (rather than the last) becomes the first round key.
+ */
+
+#include <asm/unaligned.h>
+#include <linux/bitops.h>
+#include <linux/crypto.h>
+#include <linux/init.h>
+#include <linux/module.h>
+
+/* Speck128 */
+
+#define SPECK128_BLOCK_SIZE 16
+
+#define SPECK128_128_KEY_SIZE 16
+#define SPECK128_128_NROUNDS 32
+
+#define SPECK128_192_KEY_SIZE 24
+#define SPECK128_192_NROUNDS 33
+
+#define SPECK128_256_KEY_SIZE 32
+#define SPECK128_256_NROUNDS 34
+
+struct speck128_tfm_ctx {
+ u64 round_keys[SPECK128_256_NROUNDS];
+ int nrounds;
+};
+
+static __always_inline void speck128_round(u64 *x, u64 *y, u64 k)
+{
+ *x = ror64(*x, 8);
+ *x += *y;
+ *x ^= k;
+ *y = rol64(*y, 3);
+ *y ^= *x;
+}
+
+static __always_inline void speck128_unround(u64 *x, u64 *y, u64 k)
+{
+ *y ^= *x;
+ *y = ror64(*y, 3);
+ *x ^= k;
+ *x -= *y;
+ *x = rol64(*x, 8);
+}
+
+static void speck128_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
+{
+ const struct speck128_tfm_ctx *ctx = crypto_tfm_ctx(tfm);
+ u64 y = get_unaligned_le64(in);
+ u64 x = get_unaligned_le64(in + 8);
+ int i;
+
+ for (i = 0; i < ctx->nrounds; i++)
+ speck128_round(&x, &y, ctx->round_keys[i]);
+
+ put_unaligned_le64(y, out);
+ put_unaligned_le64(x, out + 8);
+}
+
+static void speck128_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
+{
+ const struct speck128_tfm_ctx *ctx = crypto_tfm_ctx(tfm);
+ u64 y = get_unaligned_le64(in);
+ u64 x = get_unaligned_le64(in + 8);
+ int i;
+
+ for (i = ctx->nrounds - 1; i >= 0; i--)
+ speck128_unround(&x, &y, ctx->round_keys[i]);
+
+ put_unaligned_le64(y, out);
+ put_unaligned_le64(x, out + 8);
+}
+
+static int speck128_setkey(struct crypto_tfm *tfm, const u8 *key,
+ unsigned int keylen)
+{
+ struct speck128_tfm_ctx *ctx = crypto_tfm_ctx(tfm);
+ u64 l[3];
+ u64 k;
+ int i;
+
+ switch (keylen) {
+ case SPECK128_128_KEY_SIZE:
+ k = get_unaligned_le64(key);
+ l[0] = get_unaligned_le64(key + 8);
+ ctx->nrounds = SPECK128_128_NROUNDS;
+ for (i = 0; i < ctx->nrounds; i++) {
+ ctx->round_keys[i] = k;
+ speck128_round(&l[0], &k, i);
+ }
+ break;
+ case SPECK128_192_KEY_SIZE:
+ k = get_unaligned_le64(key);
+ l[0] = get_unaligned_le64(key + 8);
+ l[1] = get_unaligned_le64(key + 16);
+ ctx->nrounds = SPECK128_192_NROUNDS;
+ for (i = 0; i < ctx->nrounds; i++) {
+ ctx->round_keys[i] = k;
+ speck128_round(&l[i % 2], &k, i);
+ }
+ break;
+ case SPECK128_256_KEY_SIZE:
+ k = get_unaligned_le64(key);
+ l[0] = get_unaligned_le64(key + 8);
+ l[1] = get_unaligned_le64(key + 16);
+ l[2] = get_unaligned_le64(key + 24);
+ ctx->nrounds = SPECK128_256_NROUNDS;
+ for (i = 0; i < ctx->nrounds; i++) {
+ ctx->round_keys[i] = k;
+ speck128_round(&l[i % 3], &k, i);
+ }
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+/* Speck64 */
+
+#define SPECK64_BLOCK_SIZE 8
+
+#define SPECK64_96_KEY_SIZE 12
+#define SPECK64_96_NROUNDS 26
+
+#define SPECK64_128_KEY_SIZE 16
+#define SPECK64_128_NROUNDS 27
+
+struct speck64_tfm_ctx {
+ u32 round_keys[SPECK64_128_NROUNDS];
+ int nrounds;
+};
+
+static __always_inline void speck64_round(u32 *x, u32 *y, u32 k)
+{
+ *x = ror32(*x, 8);
+ *x += *y;
+ *x ^= k;
+ *y = rol32(*y, 3);
+ *y ^= *x;
+}
+
+static __always_inline void speck64_unround(u32 *x, u32 *y, u32 k)
+{
+ *y ^= *x;
+ *y = ror32(*y, 3);
+ *x ^= k;
+ *x -= *y;
+ *x = rol32(*x, 8);
+}
+
+static void speck64_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
+{
+ const struct speck64_tfm_ctx *ctx = crypto_tfm_ctx(tfm);
+ u32 y = get_unaligned_le32(in);
+ u32 x = get_unaligned_le32(in + 4);
+ int i;
+
+ for (i = 0; i < ctx->nrounds; i++)
+ speck64_round(&x, &y, ctx->round_keys[i]);
+
+ put_unaligned_le32(y, out);
+ put_unaligned_le32(x, out + 4);
+}
+
+static void speck64_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
+{
+ const struct speck64_tfm_ctx *ctx = crypto_tfm_ctx(tfm);
+ u32 y = get_unaligned_le32(in);
+ u32 x = get_unaligned_le32(in + 4);
+ int i;
+
+ for (i = ctx->nrounds - 1; i >= 0; i--)
+ speck64_unround(&x, &y, ctx->round_keys[i]);
+
+ put_unaligned_le32(y, out);
+ put_unaligned_le32(x, out + 4);
+}
+
+static int speck64_setkey(struct crypto_tfm *tfm, const u8 *key,
+ unsigned int keylen)
+{
+ struct speck64_tfm_ctx *ctx = crypto_tfm_ctx(tfm);
+ u32 l[3];
+ u32 k;
+ int i;
+
+ switch (keylen) {
+ case SPECK64_96_KEY_SIZE:
+ k = get_unaligned_le32(key);
+ l[0] = get_unaligned_le32(key + 4);
+ l[1] = get_unaligned_le32(key + 8);
+ ctx->nrounds = SPECK64_96_NROUNDS;
+ for (i = 0; i < ctx->nrounds; i++) {
+ ctx->round_keys[i] = k;
+ speck64_round(&l[i % 2], &k, i);
+ }
+ break;
+ case SPECK64_128_KEY_SIZE:
+ k = get_unaligned_le32(key);
+ l[0] = get_unaligned_le32(key + 4);
+ l[1] = get_unaligned_le32(key + 8);
+ l[2] = get_unaligned_le32(key + 12);
+ ctx->nrounds = SPECK64_128_NROUNDS;
+ for (i = 0; i < ctx->nrounds; i++) {
+ ctx->round_keys[i] = k;
+ speck64_round(&l[i % 3], &k, i);
+ }
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+/* Algorithm definitions */
+
+static struct crypto_alg speck_algs[] = {
+ {
+ .cra_name = "speck128",
+ .cra_driver_name = "speck128-generic",
+ .cra_priority = 100,
+ .cra_flags = CRYPTO_ALG_TYPE_CIPHER,
+ .cra_blocksize = SPECK128_BLOCK_SIZE,
+ .cra_ctxsize = sizeof(struct speck128_tfm_ctx),
+ .cra_module = THIS_MODULE,
+ .cra_u = {
+ .cipher = {
+ .cia_min_keysize = SPECK128_128_KEY_SIZE,
+ .cia_max_keysize = SPECK128_256_KEY_SIZE,
+ .cia_setkey = speck128_setkey,
+ .cia_encrypt = speck128_encrypt,
+ .cia_decrypt = speck128_decrypt
+ }
+ }
+ }, {
+ .cra_name = "speck64",
+ .cra_driver_name = "speck64-generic",
+ .cra_priority = 100,
+ .cra_flags = CRYPTO_ALG_TYPE_CIPHER,
+ .cra_blocksize = SPECK64_BLOCK_SIZE,
+ .cra_ctxsize = sizeof(struct speck64_tfm_ctx),
+ .cra_module = THIS_MODULE,
+ .cra_u = {
+ .cipher = {
+ .cia_min_keysize = SPECK64_96_KEY_SIZE,
+ .cia_max_keysize = SPECK64_128_KEY_SIZE,
+ .cia_setkey = speck64_setkey,
+ .cia_encrypt = speck64_encrypt,
+ .cia_decrypt = speck64_decrypt
+ }
+ }
+ }
+};
+
+static int __init speck_module_init(void)
+{
+ return crypto_register_algs(speck_algs, ARRAY_SIZE(speck_algs));
+}
+
+static void __exit speck_module_exit(void)
+{
+ crypto_unregister_algs(speck_algs, ARRAY_SIZE(speck_algs));
+}
+
+module_init(speck_module_init);
+module_exit(speck_module_exit);
+
+MODULE_DESCRIPTION("Speck block cipher (generic)");
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Eric Biggers <[email protected]>");
+MODULE_ALIAS_CRYPTO("speck128");
+MODULE_ALIAS_CRYPTO("speck128-generic");
+MODULE_ALIAS_CRYPTO("speck64");
+MODULE_ALIAS_CRYPTO("speck64-generic");
diff --git a/crypto/testmgr.c b/crypto/testmgr.c
index d5e23a142a04..058ed5eb6620 100644
--- a/crypto/testmgr.c
+++ b/crypto/testmgr.c
@@ -3000,6 +3000,24 @@ static const struct alg_test_desc alg_test_descs[] = {
.dec = __VECS(serpent_dec_tv_template)
}
}
+ }, {
+ .alg = "ecb(speck128)",
+ .test = alg_test_skcipher,
+ .suite = {
+ .cipher = {
+ .enc = __VECS(speck128_enc_tv_template),
+ .dec = __VECS(speck128_dec_tv_template)
+ }
+ }
+ }, {
+ .alg = "ecb(speck64)",
+ .test = alg_test_skcipher,
+ .suite = {
+ .cipher = {
+ .enc = __VECS(speck64_enc_tv_template),
+ .dec = __VECS(speck64_dec_tv_template)
+ }
+ }
}, {
.alg = "ecb(tea)",
.test = alg_test_skcipher,
diff --git a/crypto/testmgr.h b/crypto/testmgr.h
index 6044f6906bd6..3818210f77cf 100644
--- a/crypto/testmgr.h
+++ b/crypto/testmgr.h
@@ -14323,6 +14323,134 @@ static const struct cipher_testvec serpent_xts_dec_tv_template[] = {
},
};
+/*
+ * Speck test vectors taken from the original paper:
+ * "The Simon and Speck Families of Lightweight Block Ciphers"
+ * https://eprint.iacr.org/2013/404.pdf
+ *
+ * Note that the paper does not make byte and word order clear. But it was
+ * confirmed with the authors that the intended orders are little endian byte
+ * order and (y, x) word order. Equivalently, the printed test vectors, when
+ * looking at only the bytes (ignoring the whitespace that divides them into
+ * words), are backwards: the left-most byte is actually the one with the
+ * highest memory address, while the right-most byte is actually the one with
+ * the lowest memory address.
+ */
+
+static const struct cipher_testvec speck128_enc_tv_template[] = {
+ { /* Speck128/128 */
+ .key = "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f",
+ .klen = 16,
+ .input = "\x20\x6d\x61\x64\x65\x20\x69\x74"
+ "\x20\x65\x71\x75\x69\x76\x61\x6c",
+ .ilen = 16,
+ .result = "\x18\x0d\x57\x5c\xdf\xfe\x60\x78"
+ "\x65\x32\x78\x79\x51\x98\x5d\xa6",
+ .rlen = 16,
+ }, { /* Speck128/192 */
+ .key = "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17",
+ .klen = 24,
+ .input = "\x65\x6e\x74\x20\x74\x6f\x20\x43"
+ "\x68\x69\x65\x66\x20\x48\x61\x72",
+ .ilen = 16,
+ .result = "\x86\x18\x3c\xe0\x5d\x18\xbc\xf9"
+ "\x66\x55\x13\x13\x3a\xcf\xe4\x1b",
+ .rlen = 16,
+ }, { /* Speck128/256 */
+ .key = "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17"
+ "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f",
+ .klen = 32,
+ .input = "\x70\x6f\x6f\x6e\x65\x72\x2e\x20"
+ "\x49\x6e\x20\x74\x68\x6f\x73\x65",
+ .ilen = 16,
+ .result = "\x43\x8f\x18\x9c\x8d\xb4\xee\x4e"
+ "\x3e\xf5\xc0\x05\x04\x01\x09\x41",
+ .rlen = 16,
+ },
+};
+
+static const struct cipher_testvec speck128_dec_tv_template[] = {
+ { /* Speck128/128 */
+ .key = "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f",
+ .klen = 16,
+ .input = "\x18\x0d\x57\x5c\xdf\xfe\x60\x78"
+ "\x65\x32\x78\x79\x51\x98\x5d\xa6",
+ .ilen = 16,
+ .result = "\x20\x6d\x61\x64\x65\x20\x69\x74"
+ "\x20\x65\x71\x75\x69\x76\x61\x6c",
+ .rlen = 16,
+ }, { /* Speck128/192 */
+ .key = "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17",
+ .klen = 24,
+ .input = "\x86\x18\x3c\xe0\x5d\x18\xbc\xf9"
+ "\x66\x55\x13\x13\x3a\xcf\xe4\x1b",
+ .ilen = 16,
+ .result = "\x65\x6e\x74\x20\x74\x6f\x20\x43"
+ "\x68\x69\x65\x66\x20\x48\x61\x72",
+ .rlen = 16,
+ }, { /* Speck128/256 */
+ .key = "\x00\x01\x02\x03\x04\x05\x06\x07"
+ "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f"
+ "\x10\x11\x12\x13\x14\x15\x16\x17"
+ "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f",
+ .klen = 32,
+ .input = "\x43\x8f\x18\x9c\x8d\xb4\xee\x4e"
+ "\x3e\xf5\xc0\x05\x04\x01\x09\x41",
+ .ilen = 16,
+ .result = "\x70\x6f\x6f\x6e\x65\x72\x2e\x20"
+ "\x49\x6e\x20\x74\x68\x6f\x73\x65",
+ .rlen = 16,
+ },
+};
+
+static const struct cipher_testvec speck64_enc_tv_template[] = {
+ { /* Speck64/96 */
+ .key = "\x00\x01\x02\x03\x08\x09\x0a\x0b"
+ "\x10\x11\x12\x13",
+ .klen = 12,
+ .input = "\x65\x61\x6e\x73\x20\x46\x61\x74",
+ .ilen = 8,
+ .result = "\x6c\x94\x75\x41\xec\x52\x79\x9f",
+ .rlen = 8,
+ }, { /* Speck64/128 */
+ .key = "\x00\x01\x02\x03\x08\x09\x0a\x0b"
+ "\x10\x11\x12\x13\x18\x19\x1a\x1b",
+ .klen = 16,
+ .input = "\x2d\x43\x75\x74\x74\x65\x72\x3b",
+ .ilen = 8,
+ .result = "\x8b\x02\x4e\x45\x48\xa5\x6f\x8c",
+ .rlen = 8,
+ },
+};
+
+static const struct cipher_testvec speck64_dec_tv_template[] = {
+ { /* Speck64/96 */
+ .key = "\x00\x01\x02\x03\x08\x09\x0a\x0b"
+ "\x10\x11\x12\x13",
+ .klen = 12,
+ .input = "\x6c\x94\x75\x41\xec\x52\x79\x9f",
+ .ilen = 8,
+ .result = "\x65\x61\x6e\x73\x20\x46\x61\x74",
+ .rlen = 8,
+ }, { /* Speck64/128 */
+ .key = "\x00\x01\x02\x03\x08\x09\x0a\x0b"
+ "\x10\x11\x12\x13\x18\x19\x1a\x1b",
+ .klen = 16,
+ .input = "\x8b\x02\x4e\x45\x48\xa5\x6f\x8c",
+ .ilen = 8,
+ .result = "\x2d\x43\x75\x74\x74\x65\x72\x3b",
+ .rlen = 8,
+ },
+};
+
/* Cast6 test vectors from RFC 2612 */
static const struct cipher_testvec cast6_enc_tv_template[] = {
{
--
2.16.1.291.g4437f3f132-goog
On Wed, Feb 14, 2018 at 10:42:18AM -0800, Eric Biggers wrote:
> Hello,
>
> This series adds Speck support to the crypto API, including the Speck128
> and Speck64 variants. Speck is a lightweight block cipher that can be
> much faster than AES on processors that don't have AES instructions.
>
> We are planning to offer Speck-XTS (probably Speck128/256-XTS) as an
> option for dm-crypt and fscrypt on Android, for low-end mobile devices
> with older CPUs such as ARMv7 which don't have the Cryptography
> Extensions. Currently, such devices are unencrypted because AES is not
> fast enough, even when the NEON bit-sliced implementation of AES is
> used. Other AES alternatives such as Twofish, Threefish, Camellia,
> CAST6, and Serpent aren't fast enough either; it seems that only a
> modern ARX cipher can provide sufficient performance on these devices.
>
> This is a replacement for our original proposal
> (https://patchwork.kernel.org/patch/10101451/) which was to offer
> ChaCha20 for these devices. However, the use of a stream cipher for
> disk/file encryption with no space to store nonces would have been much
> more insecure than we thought initially, given that it would be used on
> top of flash storage as well as potentially on top of F2FS, neither of
> which is guaranteed to overwrite data in-place.
>
> Speck has been somewhat controversial due to its origin. Nevertheless,
> it has a straightforward design (it's an ARX cipher), and it appears to
> be the leading software-optimized lightweight block cipher currently,
> with the most cryptanalysis. It's also easy to implement without side
> channels, unlike AES. Moreover, we only intend Speck to be used when
> the status quo is no encryption, due to AES not being fast enough.
>
> We've also considered a novel length-preserving encryption mode based on
> ChaCha20 and Poly1305. While theoretically attractive, such a mode
> would be a brand new crypto construction and would be more complicated
> and difficult to implement efficiently in comparison to Speck-XTS.
>
> Thus, patch 1 adds a generic implementation of Speck, and the following
> patches add a 32-bit ARM NEON implementation of Speck-XTS. The
> NEON-accelerated implementation is much faster than the generic
> implementation and therefore is the implementation that would primarily
> be used in practice on the devices we are targeting.
>
> There is no AArch64 implementation included, since most such CPUs have
> the Cryptography Extensions, allowing the use of AES. An AArch64
> implementation can be added later if there is interest though.
>
> Changed since v2:
>
> - Fix __speck64_xts_crypt() to work on big endian CPUs.
>
> Changed since v1:
>
> - Use the word order recommended by the Speck authors. All test
> vectors were updated.
>
> Eric Biggers (5):
> crypto: add support for the Speck block cipher
> crypto: speck - export common helpers
> crypto: arm/speck - add NEON-accelerated implementation of Speck-XTS
> crypto: speck - add test vectors for Speck128-XTS
> crypto: speck - add test vectors for Speck64-XTS
>
> arch/arm/crypto/Kconfig | 6 +
> arch/arm/crypto/Makefile | 2 +
> arch/arm/crypto/speck-neon-core.S | 432 +++++++++
> arch/arm/crypto/speck-neon-glue.c | 288 ++++++
> crypto/Kconfig | 14 +
> crypto/Makefile | 1 +
> crypto/speck.c | 307 ++++++
> crypto/testmgr.c | 36 +
> crypto/testmgr.h | 1486 +++++++++++++++++++++++++++++
> include/crypto/speck.h | 62 ++
> 10 files changed, 2634 insertions(+)
> create mode 100644 arch/arm/crypto/speck-neon-core.S
> create mode 100644 arch/arm/crypto/speck-neon-glue.c
> create mode 100644 crypto/speck.c
> create mode 100644 include/crypto/speck.h
All applied. Thanks.
--
Email: Herbert Xu <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
Hi Eric,
On 14.02.2018 19:42, Eric Biggers wrote:
> Add an ARM NEON-accelerated implementation of Speck-XTS. It operates on
> 128-byte chunks at a time, i.e. 8 blocks for Speck128 or 16 blocks for
> Speck64. Each 128-byte chunk goes through XTS preprocessing, then is
> encrypted/decrypted (doing one cipher round for all the blocks, then the
> next round, etc.), then goes through XTS postprocessing.
>
> The performance depends on the processor but can be about 3 times faster
> than the generic code. For example, on an ARMv7 processor we observe
> the following performance with Speck128/256-XTS:
>
> xts-speck128-neon: Encryption 107.9 MB/s, Decryption 108.1 MB/s
> xts(speck128-generic): Encryption 32.1 MB/s, Decryption 36.6 MB/s
>
> In comparison to AES-256-XTS without the Cryptography Extensions:
>
> xts-aes-neonbs: Encryption 41.2 MB/s, Decryption 36.7 MB/s
> xts(aes-asm): Encryption 31.7 MB/s, Decryption 30.8 MB/s
> xts(aes-generic): Encryption 21.2 MB/s, Decryption 20.9 MB/s
>
> Speck64/128-XTS is even faster:
>
> xts-speck64-neon: Encryption 138.6 MB/s, Decryption 139.1 MB/s
>
> Note that as with the generic code, only the Speck128 and Speck64
> variants are supported. Also, for now only the XTS mode of operation is
> supported, to target the disk and file encryption use cases. The NEON
> code also only handles the portion of the data that is evenly divisible
> into 128-byte chunks, with any remainder handled by a C fallback. Of
> course, other modes of operation could be added later if needed, and/or
> the NEON code could be updated to handle other buffer sizes.
>
> The XTS specification is only defined for AES which has a 128-bit block
> size, so for the GF(2^64) math needed for Speck64-XTS we use the
> reducing polynomial 'x^64 + x^4 + x^3 + x + 1' given by the original XEX
> paper. Of course, when possible users should use Speck128-XTS, but even
> that may be too slow on some processors; Speck64-XTS can be faster.
>
> Signed-off-by: Eric Biggers <[email protected]>
> ---
> arch/arm/crypto/Kconfig | 6 +
> arch/arm/crypto/Makefile | 2 +
> arch/arm/crypto/speck-neon-core.S | 432 ++++++++++++++++++++++++++++++
> arch/arm/crypto/speck-neon-glue.c | 288 ++++++++++++++++++++
> 4 files changed, 728 insertions(+)
> create mode 100644 arch/arm/crypto/speck-neon-core.S
> create mode 100644 arch/arm/crypto/speck-neon-glue.c
>
> diff --git a/arch/arm/crypto/Kconfig b/arch/arm/crypto/Kconfig
> index b8e69fe282b8..925d1364727a 100644
> --- a/arch/arm/crypto/Kconfig
> +++ b/arch/arm/crypto/Kconfig
> @@ -121,4 +121,10 @@ config CRYPTO_CHACHA20_NEON
> select CRYPTO_BLKCIPHER
> select CRYPTO_CHACHA20
>
> +config CRYPTO_SPECK_NEON
> + tristate "NEON accelerated Speck cipher algorithms"
> + depends on KERNEL_MODE_NEON
> + select CRYPTO_BLKCIPHER
> + select CRYPTO_SPECK
> +
> endif
> diff --git a/arch/arm/crypto/Makefile b/arch/arm/crypto/Makefile
> index 30ef8e291271..a758107c5525 100644
> --- a/arch/arm/crypto/Makefile
> +++ b/arch/arm/crypto/Makefile
> @@ -10,6 +10,7 @@ obj-$(CONFIG_CRYPTO_SHA1_ARM_NEON) += sha1-arm-neon.o
> obj-$(CONFIG_CRYPTO_SHA256_ARM) += sha256-arm.o
> obj-$(CONFIG_CRYPTO_SHA512_ARM) += sha512-arm.o
> obj-$(CONFIG_CRYPTO_CHACHA20_NEON) += chacha20-neon.o
> +obj-$(CONFIG_CRYPTO_SPECK_NEON) += speck-neon.o
>
> ce-obj-$(CONFIG_CRYPTO_AES_ARM_CE) += aes-arm-ce.o
> ce-obj-$(CONFIG_CRYPTO_SHA1_ARM_CE) += sha1-arm-ce.o
> @@ -53,6 +54,7 @@ ghash-arm-ce-y := ghash-ce-core.o ghash-ce-glue.o
> crct10dif-arm-ce-y := crct10dif-ce-core.o crct10dif-ce-glue.o
> crc32-arm-ce-y:= crc32-ce-core.o crc32-ce-glue.o
> chacha20-neon-y := chacha20-neon-core.o chacha20-neon-glue.o
> +speck-neon-y := speck-neon-core.o speck-neon-glue.o
>
> quiet_cmd_perl = PERL $@
> cmd_perl = $(PERL) $(<) > $(@)
> diff --git a/arch/arm/crypto/speck-neon-core.S
> b/arch/arm/crypto/speck-neon-core.S
> new file mode 100644
> index 000000000000..3c1e203e53b9
> --- /dev/null
> +++ b/arch/arm/crypto/speck-neon-core.S
> @@ -0,0 +1,432 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * NEON-accelerated implementation of Speck128-XTS and Speck64-XTS
> + *
> + * Copyright (c) 2018 Google, Inc
> + *
> + * Author: Eric Biggers <[email protected]>
> + */
> +
> +#include <linux/linkage.h>
> +
> + .text
> + .fpu neon
> +
> + // arguments
> + ROUND_KEYS .req r0 // const {u64,u32} *round_keys
> + NROUNDS .req r1 // int nrounds
> + DST .req r2 // void *dst
> + SRC .req r3 // const void *src
> + NBYTES .req r4 // unsigned int nbytes
> + TWEAK .req r5 // void *tweak
> +
> + // registers which hold the data being encrypted/decrypted
> + X0 .req q0
> + X0_L .req d0
> + X0_H .req d1
> + Y0 .req q1
> + Y0_H .req d3
> + X1 .req q2
> + X1_L .req d4
> + X1_H .req d5
> + Y1 .req q3
> + Y1_H .req d7
> + X2 .req q4
> + X2_L .req d8
> + X2_H .req d9
> + Y2 .req q5
> + Y2_H .req d11
> + X3 .req q6
> + X3_L .req d12
> + X3_H .req d13
> + Y3 .req q7
> + Y3_H .req d15
> +
> + // the round key, duplicated in all lanes
> + ROUND_KEY .req q8
> + ROUND_KEY_L .req d16
> + ROUND_KEY_H .req d17
> +
> + // index vector for vtbl-based 8-bit rotates
> + ROTATE_TABLE .req d18
> +
> + // multiplication table for updating XTS tweaks
> + GF128MUL_TABLE .req d19
> + GF64MUL_TABLE .req d19
> +
> + // current XTS tweak value(s)
> + TWEAKV .req q10
> + TWEAKV_L .req d20
> + TWEAKV_H .req d21
> +
> + TMP0 .req q12
> + TMP0_L .req d24
> + TMP0_H .req d25
> + TMP1 .req q13
> + TMP2 .req q14
> + TMP3 .req q15
> +
> + .align 4
> +.Lror64_8_table:
> + .byte 1, 2, 3, 4, 5, 6, 7, 0
> +.Lror32_8_table:
> + .byte 1, 2, 3, 0, 5, 6, 7, 4
> +.Lrol64_8_table:
> + .byte 7, 0, 1, 2, 3, 4, 5, 6
> +.Lrol32_8_table:
> + .byte 3, 0, 1, 2, 7, 4, 5, 6
> +.Lgf128mul_table:
> + .byte 0, 0x87
> + .fill 14
> +.Lgf64mul_table:
> + .byte 0, 0x1b, (0x1b << 1), (0x1b << 1) ^ 0x1b
> + .fill 12
> +
> +/*
> + * _speck_round_128bytes() - Speck encryption round on 128 bytes at a time
> + *
> + * Do one Speck encryption round on the 128 bytes (8 blocks for
> Speck128, 16 for
> + * Speck64) stored in X0-X3 and Y0-Y3, using the round key stored in all lanes
> + * of ROUND_KEY. 'n' is the lane size: 64 for Speck128, or 32 for Speck64.
> + *
> + * The 8-bit rotates are implemented using vtbl instead of vshr + vsli because
> + * the vtbl approach is faster on some processors and the same speed on others.
> + */
> +.macro _speck_round_128bytes n
> +
> + // x = ror(x, 8)
> + vtbl.8 X0_L, {X0_L}, ROTATE_TABLE
> + vtbl.8 X0_H, {X0_H}, ROTATE_TABLE
> + vtbl.8 X1_L, {X1_L}, ROTATE_TABLE
> + vtbl.8 X1_H, {X1_H}, ROTATE_TABLE
> + vtbl.8 X2_L, {X2_L}, ROTATE_TABLE
> + vtbl.8 X2_H, {X2_H}, ROTATE_TABLE
> + vtbl.8 X3_L, {X3_L}, ROTATE_TABLE
> + vtbl.8 X3_H, {X3_H}, ROTATE_TABLE
> +
> + // x += y
> + vadd.u\n X0, Y0
> + vadd.u\n X1, Y1
> + vadd.u\n X2, Y2
> + vadd.u\n X3, Y3
> +
> + // x ^= k
> + veor X0, ROUND_KEY
> + veor X1, ROUND_KEY
> + veor X2, ROUND_KEY
> + veor X3, ROUND_KEY
> +
> + // y = rol(y, 3)
> + vshl.u\n TMP0, Y0, #3
> + vshl.u\n TMP1, Y1, #3
> + vshl.u\n TMP2, Y2, #3
> + vshl.u\n TMP3, Y3, #3
> + vsri.u\n TMP0, Y0, #(\n - 3)
> + vsri.u\n TMP1, Y1, #(\n - 3)
> + vsri.u\n TMP2, Y2, #(\n - 3)
> + vsri.u\n TMP3, Y3, #(\n - 3)
> +
> + // y ^= x
> + veor Y0, TMP0, X0
> + veor Y1, TMP1, X1
> + veor Y2, TMP2, X2
> + veor Y3, TMP3, X3
> +.endm
> +
> +/*
> + * _speck_unround_128bytes() - Speck decryption round on 128 bytes at a time
> + *
> + * This is the inverse of _speck_round_128bytes().
> + */
> +.macro _speck_unround_128bytes n
> +
> + // y ^= x
> + veor TMP0, Y0, X0
> + veor TMP1, Y1, X1
> + veor TMP2, Y2, X2
> + veor TMP3, Y3, X3
> +
> + // y = ror(y, 3)
> + vshr.u\n Y0, TMP0, #3
> + vshr.u\n Y1, TMP1, #3
> + vshr.u\n Y2, TMP2, #3
> + vshr.u\n Y3, TMP3, #3
> + vsli.u\n Y0, TMP0, #(\n - 3)
> + vsli.u\n Y1, TMP1, #(\n - 3)
> + vsli.u\n Y2, TMP2, #(\n - 3)
> + vsli.u\n Y3, TMP3, #(\n - 3)
> +
> + // x ^= k
> + veor X0, ROUND_KEY
> + veor X1, ROUND_KEY
> + veor X2, ROUND_KEY
> + veor X3, ROUND_KEY
> +
> + // x -= y
> + vsub.u\n X0, Y0
> + vsub.u\n X1, Y1
> + vsub.u\n X2, Y2
> + vsub.u\n X3, Y3
> +
> + // x = rol(x, 8);
> + vtbl.8 X0_L, {X0_L}, ROTATE_TABLE
> + vtbl.8 X0_H, {X0_H}, ROTATE_TABLE
> + vtbl.8 X1_L, {X1_L}, ROTATE_TABLE
> + vtbl.8 X1_H, {X1_H}, ROTATE_TABLE
> + vtbl.8 X2_L, {X2_L}, ROTATE_TABLE
> + vtbl.8 X2_H, {X2_H}, ROTATE_TABLE
> + vtbl.8 X3_L, {X3_L}, ROTATE_TABLE
> + vtbl.8 X3_H, {X3_H}, ROTATE_TABLE
> +.endm
> +
> +.macro _xts128_precrypt_one dst_reg, tweak_buf, tmp
> +
> + // Load the next source block
> + vld1.8 {\dst_reg}, [SRC]!
> +
> + // Save the current tweak in the tweak buffer
> + vst1.8 {TWEAKV}, [\tweak_buf:128]!
> +
> + // XOR the next source block with the current tweak
> + veor \dst_reg, TWEAKV
> +
> + /*
> + * Calculate the next tweak by multiplying the current one by x,
> + * modulo p(x) = x^128 + x^7 + x^2 + x + 1.
> + */
> + vshr.u64 \tmp, TWEAKV, #63
> + vshl.u64 TWEAKV, #1
> + veor TWEAKV_H, \tmp\()_L
> + vtbl.8 \tmp\()_H, {GF128MUL_TABLE}, \tmp\()_H
> + veor TWEAKV_L, \tmp\()_H
> +.endm
> +
> +.macro _xts64_precrypt_two dst_reg, tweak_buf, tmp
> +
> + // Load the next two source blocks
> + vld1.8 {\dst_reg}, [SRC]!
> +
> + // Save the current two tweaks in the tweak buffer
> + vst1.8 {TWEAKV}, [\tweak_buf:128]!
> +
> + // XOR the next two source blocks with the current two tweaks
> + veor \dst_reg, TWEAKV
> +
> + /*
> + * Calculate the next two tweaks by multiplying the current ones by x^2,
> + * modulo p(x) = x^64 + x^4 + x^3 + x + 1.
> + */
> + vshr.u64 \tmp, TWEAKV, #62
> + vshl.u64 TWEAKV, #2
> + vtbl.8 \tmp\()_L, {GF64MUL_TABLE}, \tmp\()_L
> + vtbl.8 \tmp\()_H, {GF64MUL_TABLE}, \tmp\()_H
> + veor TWEAKV, \tmp
> +.endm
> +
> +/*
> + * _speck_xts_crypt() - Speck-XTS encryption/decryption
> + *
> + * Encrypt or decrypt NBYTES bytes of data from the SRC buffer to the
> DST buffer
> + * using Speck-XTS, specifically the variant with a block size of
> '2n' and round
> + * count given by NROUNDS. The expanded round keys are given in
> ROUND_KEYS, and
> + * the current XTS tweak value is given in TWEAK. It's assumed that
> NBYTES is a
> + * nonzero multiple of 128.
> + */
> +.macro _speck_xts_crypt n, decrypting
> + push {r4-r7}
> + mov r7, sp
> +
> + /*
> + * The first four parameters were passed in registers r0-r3. Load the
> + * additional parameters, which were passed on the stack.
> + */
> + ldr NBYTES, [sp, #16]
> + ldr TWEAK, [sp, #20]
> +
> + /*
> + * If decrypting, modify the ROUND_KEYS parameter to point to the last
> + * round key rather than the first, since for decryption the round keys
> + * are used in reverse order.
> + */
> +.if \decrypting
> +.if \n == 64
> + add ROUND_KEYS, ROUND_KEYS, NROUNDS, lsl #3
> + sub ROUND_KEYS, #8
> +.else
> + add ROUND_KEYS, ROUND_KEYS, NROUNDS, lsl #2
> + sub ROUND_KEYS, #4
> +.endif
> +.endif
> +
> + // Load the index vector for vtbl-based 8-bit rotates
> +.if \decrypting
> + ldr r12, =.Lrol\n\()_8_table
> +.else
> + ldr r12, =.Lror\n\()_8_table
> +.endif
> + vld1.8 {ROTATE_TABLE}, [r12:64]
> +
> + // One-time XTS preparation
> +
> + /*
> + * Allocate stack space to store 128 bytes worth of tweaks. For
> + * performance, this space is aligned to a 16-byte boundary so that we
> + * can use the load/store instructions that declare 16-byte alignment.
> + */
> + sub sp, #128
> + bic sp, #0xf
This fails here when building with CONFIG_THUMB2_KERNEL=y
AS arch/arm/crypto/speck-neon-core.o
arch/arm/crypto/speck-neon-core.S: Assembler messages:
arch/arm/crypto/speck-neon-core.S:419: Error: r13 not allowed here --
`bic sp,#0xf'
arch/arm/crypto/speck-neon-core.S:423: Error: r13 not allowed here --
`bic sp,#0xf'
arch/arm/crypto/speck-neon-core.S:427: Error: r13 not allowed here --
`bic sp,#0xf'
arch/arm/crypto/speck-neon-core.S:431: Error: r13 not allowed here --
`bic sp,#0xf'
In a quick hack this change seems to address it:
- sub sp, #128
- bic sp, #0xf
+ mov r6, sp
+ sub r6, #128
+ bic r6, #0xf
+ mov sp, r6
But there is probably a better solution to address this.
--
Stefan
> +
> +.if \n == 64
> + // Load first tweak
> + vld1.8 {TWEAKV}, [TWEAK]
> +
> + // Load GF(2^128) multiplication table
> + ldr r12, =.Lgf128mul_table
> + vld1.8 {GF128MUL_TABLE}, [r12:64]
> +.else
> + // Load first tweak
> + vld1.8 {TWEAKV_L}, [TWEAK]
> +
> + // Load GF(2^64) multiplication table
> + ldr r12, =.Lgf64mul_table
> + vld1.8 {GF64MUL_TABLE}, [r12:64]
> +
> + // Calculate second tweak, packing it together with the first
> + vshr.u64 TMP0_L, TWEAKV_L, #63
> + vtbl.u8 TMP0_L, {GF64MUL_TABLE}, TMP0_L
> + vshl.u64 TWEAKV_H, TWEAKV_L, #1
> + veor TWEAKV_H, TMP0_L
> +.endif
> +
> +.Lnext_128bytes_\@:
> +
> + /*
> + * Load the source blocks into {X,Y}[0-3], XOR them with their XTS tweak
> + * values, and save the tweaks on the stack for later. Then
> + * de-interleave the 'x' and 'y' elements of each block, i.e. make it so
> + * that the X[0-3] registers contain only the second halves of blocks,
> + * and the Y[0-3] registers contain only the first halves of blocks.
> + * (Speck uses the order (y, x) rather than the more intuitive (x, y).)
> + */
> + mov r12, sp
> +.if \n == 64
> + _xts128_precrypt_one X0, r12, TMP0
> + _xts128_precrypt_one Y0, r12, TMP0
> + _xts128_precrypt_one X1, r12, TMP0
> + _xts128_precrypt_one Y1, r12, TMP0
> + _xts128_precrypt_one X2, r12, TMP0
> + _xts128_precrypt_one Y2, r12, TMP0
> + _xts128_precrypt_one X3, r12, TMP0
> + _xts128_precrypt_one Y3, r12, TMP0
> + vswp X0_L, Y0_H
> + vswp X1_L, Y1_H
> + vswp X2_L, Y2_H
> + vswp X3_L, Y3_H
> +.else
> + _xts64_precrypt_two X0, r12, TMP0
> + _xts64_precrypt_two Y0, r12, TMP0
> + _xts64_precrypt_two X1, r12, TMP0
> + _xts64_precrypt_two Y1, r12, TMP0
> + _xts64_precrypt_two X2, r12, TMP0
> + _xts64_precrypt_two Y2, r12, TMP0
> + _xts64_precrypt_two X3, r12, TMP0
> + _xts64_precrypt_two Y3, r12, TMP0
> + vuzp.32 Y0, X0
> + vuzp.32 Y1, X1
> + vuzp.32 Y2, X2
> + vuzp.32 Y3, X3
> +.endif
> +
> + // Do the cipher rounds
> +
> + mov r12, ROUND_KEYS
> + mov r6, NROUNDS
> +
> +.Lnext_round_\@:
> +.if \decrypting
> +.if \n == 64
> + vld1.64 ROUND_KEY_L, [r12]
> + sub r12, #8
> + vmov ROUND_KEY_H, ROUND_KEY_L
> +.else
> + vld1.32 {ROUND_KEY_L[],ROUND_KEY_H[]}, [r12]
> + sub r12, #4
> +.endif
> + _speck_unround_128bytes \n
> +.else
> +.if \n == 64
> + vld1.64 ROUND_KEY_L, [r12]!
> + vmov ROUND_KEY_H, ROUND_KEY_L
> +.else
> + vld1.32 {ROUND_KEY_L[],ROUND_KEY_H[]}, [r12]!
> +.endif
> + _speck_round_128bytes \n
> +.endif
> + subs r6, r6, #1
> + bne .Lnext_round_\@
> +
> + // Re-interleave the 'x' and 'y' elements of each block
> +.if \n == 64
> + vswp X0_L, Y0_H
> + vswp X1_L, Y1_H
> + vswp X2_L, Y2_H
> + vswp X3_L, Y3_H
> +.else
> + vzip.32 Y0, X0
> + vzip.32 Y1, X1
> + vzip.32 Y2, X2
> + vzip.32 Y3, X3
> +.endif
> +
> + // XOR the encrypted/decrypted blocks with the tweaks we saved earlier
> + mov r12, sp
> + vld1.8 {TMP0, TMP1}, [r12:128]!
> + vld1.8 {TMP2, TMP3}, [r12:128]!
> + veor X0, TMP0
> + veor Y0, TMP1
> + veor X1, TMP2
> + veor Y1, TMP3
> + vld1.8 {TMP0, TMP1}, [r12:128]!
> + vld1.8 {TMP2, TMP3}, [r12:128]!
> + veor X2, TMP0
> + veor Y2, TMP1
> + veor X3, TMP2
> + veor Y3, TMP3
> +
> + // Store the ciphertext in the destination buffer
> + vst1.8 {X0, Y0}, [DST]!
> + vst1.8 {X1, Y1}, [DST]!
> + vst1.8 {X2, Y2}, [DST]!
> + vst1.8 {X3, Y3}, [DST]!
> +
> + // Continue if there are more 128-byte chunks remaining, else return
> + subs NBYTES, #128
> + bne .Lnext_128bytes_\@
> +
> + // Store the next tweak
> +.if \n == 64
> + vst1.8 {TWEAKV}, [TWEAK]
> +.else
> + vst1.8 {TWEAKV_L}, [TWEAK]
> +.endif
> +
> + mov sp, r7
> + pop {r4-r7}
> + bx lr
> +.endm
> +
> +ENTRY(speck128_xts_encrypt_neon)
> + _speck_xts_crypt n=64, decrypting=0
> +ENDPROC(speck128_xts_encrypt_neon)
> +
> +ENTRY(speck128_xts_decrypt_neon)
> + _speck_xts_crypt n=64, decrypting=1
> +ENDPROC(speck128_xts_decrypt_neon)
> +
> +ENTRY(speck64_xts_encrypt_neon)
> + _speck_xts_crypt n=32, decrypting=0
> +ENDPROC(speck64_xts_encrypt_neon)
> +
> +ENTRY(speck64_xts_decrypt_neon)
> + _speck_xts_crypt n=32, decrypting=1
> +ENDPROC(speck64_xts_decrypt_neon)
> diff --git a/arch/arm/crypto/speck-neon-glue.c
> b/arch/arm/crypto/speck-neon-glue.c
> new file mode 100644
> index 000000000000..f012c3ea998f
> --- /dev/null
> +++ b/arch/arm/crypto/speck-neon-glue.c
> @@ -0,0 +1,288 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * NEON-accelerated implementation of Speck128-XTS and Speck64-XTS
> + *
> + * Copyright (c) 2018 Google, Inc
> + *
> + * Note: the NIST recommendation for XTS only specifies a 128-bit block size,
> + * but a 64-bit version (needed for Speck64) is fairly
> straightforward; the math
> + * is just done in GF(2^64) instead of GF(2^128), with the reducing polynomial
> + * x^64 + x^4 + x^3 + x + 1 from the original XEX paper (Rogaway, 2004:
> + * "Efficient Instantiations of Tweakable Blockciphers and Refinements to Modes
> + * OCB and PMAC"), represented as 0x1B.
> + */
> +
> +#include <asm/hwcap.h>
> +#include <asm/neon.h>
> +#include <asm/simd.h>
> +#include <crypto/algapi.h>
> +#include <crypto/gf128mul.h>
> +#include <crypto/internal/skcipher.h>
> +#include <crypto/speck.h>
> +#include <crypto/xts.h>
> +#include <linux/kernel.h>
> +#include <linux/module.h>
> +
> +/* The assembly functions only handle multiples of 128 bytes */
> +#define SPECK_NEON_CHUNK_SIZE 128
> +
> +/* Speck128 */
> +
> +struct speck128_xts_tfm_ctx {
> + struct speck128_tfm_ctx main_key;
> + struct speck128_tfm_ctx tweak_key;
> +};
> +
> +asmlinkage void speck128_xts_encrypt_neon(const u64 *round_keys, int nrounds,
> + void *dst, const void *src,
> + unsigned int nbytes, void *tweak);
> +
> +asmlinkage void speck128_xts_decrypt_neon(const u64 *round_keys, int nrounds,
> + void *dst, const void *src,
> + unsigned int nbytes, void *tweak);
> +
> +typedef void (*speck128_crypt_one_t)(const struct speck128_tfm_ctx *,
> + u8 *, const u8 *);
> +typedef void (*speck128_xts_crypt_many_t)(const u64 *, int, void *,
> + const void *, unsigned int, void *);
> +
> +static __always_inline int
> +__speck128_xts_crypt(struct skcipher_request *req,
> + speck128_crypt_one_t crypt_one,
> + speck128_xts_crypt_many_t crypt_many)
> +{
> + struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
> + const struct speck128_xts_tfm_ctx *ctx = crypto_skcipher_ctx(tfm);
> + struct skcipher_walk walk;
> + le128 tweak;
> + int err;
> +
> + err = skcipher_walk_virt(&walk, req, true);
> +
> + crypto_speck128_encrypt(&ctx->tweak_key, (u8 *)&tweak, walk.iv);
> +
> + while (walk.nbytes > 0) {
> + unsigned int nbytes = walk.nbytes;
> + u8 *dst = walk.dst.virt.addr;
> + const u8 *src = walk.src.virt.addr;
> +
> + if (nbytes >= SPECK_NEON_CHUNK_SIZE && may_use_simd()) {
> + unsigned int count;
> +
> + count = round_down(nbytes, SPECK_NEON_CHUNK_SIZE);
> + kernel_neon_begin();
> + (*crypt_many)(ctx->main_key.round_keys,
> + ctx->main_key.nrounds,
> + dst, src, count, &tweak);
> + kernel_neon_end();
> + dst += count;
> + src += count;
> + nbytes -= count;
> + }
> +
> + /* Handle any remainder with generic code */
> + while (nbytes >= sizeof(tweak)) {
> + le128_xor((le128 *)dst, (const le128 *)src, &tweak);
> + (*crypt_one)(&ctx->main_key, dst, dst);
> + le128_xor((le128 *)dst, (const le128 *)dst, &tweak);
> + gf128mul_x_ble(&tweak, &tweak);
> +
> + dst += sizeof(tweak);
> + src += sizeof(tweak);
> + nbytes -= sizeof(tweak);
> + }
> + err = skcipher_walk_done(&walk, nbytes);
> + }
> +
> + return err;
> +}
> +
> +static int speck128_xts_encrypt(struct skcipher_request *req)
> +{
> + return __speck128_xts_crypt(req, crypto_speck128_encrypt,
> + speck128_xts_encrypt_neon);
> +}
> +
> +static int speck128_xts_decrypt(struct skcipher_request *req)
> +{
> + return __speck128_xts_crypt(req, crypto_speck128_decrypt,
> + speck128_xts_decrypt_neon);
> +}
> +
> +static int speck128_xts_setkey(struct crypto_skcipher *tfm, const u8 *key,
> + unsigned int keylen)
> +{
> + struct speck128_xts_tfm_ctx *ctx = crypto_skcipher_ctx(tfm);
> + int err;
> +
> + err = xts_verify_key(tfm, key, keylen);
> + if (err)
> + return err;
> +
> + keylen /= 2;
> +
> + err = crypto_speck128_setkey(&ctx->main_key, key, keylen);
> + if (err)
> + return err;
> +
> + return crypto_speck128_setkey(&ctx->tweak_key, key + keylen, keylen);
> +}
> +
> +/* Speck64 */
> +
> +struct speck64_xts_tfm_ctx {
> + struct speck64_tfm_ctx main_key;
> + struct speck64_tfm_ctx tweak_key;
> +};
> +
> +asmlinkage void speck64_xts_encrypt_neon(const u32 *round_keys, int nrounds,
> + void *dst, const void *src,
> + unsigned int nbytes, void *tweak);
> +
> +asmlinkage void speck64_xts_decrypt_neon(const u32 *round_keys, int nrounds,
> + void *dst, const void *src,
> + unsigned int nbytes, void *tweak);
> +
> +typedef void (*speck64_crypt_one_t)(const struct speck64_tfm_ctx *,
> + u8 *, const u8 *);
> +typedef void (*speck64_xts_crypt_many_t)(const u32 *, int, void *,
> + const void *, unsigned int, void *);
> +
> +static __always_inline int
> +__speck64_xts_crypt(struct skcipher_request *req, speck64_crypt_one_t
> crypt_one,
> + speck64_xts_crypt_many_t crypt_many)
> +{
> + struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
> + const struct speck64_xts_tfm_ctx *ctx = crypto_skcipher_ctx(tfm);
> + struct skcipher_walk walk;
> + __le64 tweak;
> + int err;
> +
> + err = skcipher_walk_virt(&walk, req, true);
> +
> + crypto_speck64_encrypt(&ctx->tweak_key, (u8 *)&tweak, walk.iv);
> +
> + while (walk.nbytes > 0) {
> + unsigned int nbytes = walk.nbytes;
> + u8 *dst = walk.dst.virt.addr;
> + const u8 *src = walk.src.virt.addr;
> +
> + if (nbytes >= SPECK_NEON_CHUNK_SIZE && may_use_simd()) {
> + unsigned int count;
> +
> + count = round_down(nbytes, SPECK_NEON_CHUNK_SIZE);
> + kernel_neon_begin();
> + (*crypt_many)(ctx->main_key.round_keys,
> + ctx->main_key.nrounds,
> + dst, src, count, &tweak);
> + kernel_neon_end();
> + dst += count;
> + src += count;
> + nbytes -= count;
> + }
> +
> + /* Handle any remainder with generic code */
> + while (nbytes >= sizeof(tweak)) {
> + *(__le64 *)dst = *(__le64 *)src ^ tweak;
> + (*crypt_one)(&ctx->main_key, dst, dst);
> + *(__le64 *)dst ^= tweak;
> + tweak = cpu_to_le64((le64_to_cpu(tweak) << 1) ^
> + ((tweak & cpu_to_le64(1ULL << 63)) ?
> + 0x1B : 0));
> + dst += sizeof(tweak);
> + src += sizeof(tweak);
> + nbytes -= sizeof(tweak);
> + }
> + err = skcipher_walk_done(&walk, nbytes);
> + }
> +
> + return err;
> +}
> +
> +static int speck64_xts_encrypt(struct skcipher_request *req)
> +{
> + return __speck64_xts_crypt(req, crypto_speck64_encrypt,
> + speck64_xts_encrypt_neon);
> +}
> +
> +static int speck64_xts_decrypt(struct skcipher_request *req)
> +{
> + return __speck64_xts_crypt(req, crypto_speck64_decrypt,
> + speck64_xts_decrypt_neon);
> +}
> +
> +static int speck64_xts_setkey(struct crypto_skcipher *tfm, const u8 *key,
> + unsigned int keylen)
> +{
> + struct speck64_xts_tfm_ctx *ctx = crypto_skcipher_ctx(tfm);
> + int err;
> +
> + err = xts_verify_key(tfm, key, keylen);
> + if (err)
> + return err;
> +
> + keylen /= 2;
> +
> + err = crypto_speck64_setkey(&ctx->main_key, key, keylen);
> + if (err)
> + return err;
> +
> + return crypto_speck64_setkey(&ctx->tweak_key, key + keylen, keylen);
> +}
> +
> +static struct skcipher_alg speck_algs[] = {
> + {
> + .base.cra_name = "xts(speck128)",
> + .base.cra_driver_name = "xts-speck128-neon",
> + .base.cra_priority = 300,
> + .base.cra_blocksize = SPECK128_BLOCK_SIZE,
> + .base.cra_ctxsize = sizeof(struct speck128_xts_tfm_ctx),
> + .base.cra_alignmask = 7,
> + .base.cra_module = THIS_MODULE,
> + .min_keysize = 2 * SPECK128_128_KEY_SIZE,
> + .max_keysize = 2 * SPECK128_256_KEY_SIZE,
> + .ivsize = SPECK128_BLOCK_SIZE,
> + .walksize = SPECK_NEON_CHUNK_SIZE,
> + .setkey = speck128_xts_setkey,
> + .encrypt = speck128_xts_encrypt,
> + .decrypt = speck128_xts_decrypt,
> + }, {
> + .base.cra_name = "xts(speck64)",
> + .base.cra_driver_name = "xts-speck64-neon",
> + .base.cra_priority = 300,
> + .base.cra_blocksize = SPECK64_BLOCK_SIZE,
> + .base.cra_ctxsize = sizeof(struct speck64_xts_tfm_ctx),
> + .base.cra_alignmask = 7,
> + .base.cra_module = THIS_MODULE,
> + .min_keysize = 2 * SPECK64_96_KEY_SIZE,
> + .max_keysize = 2 * SPECK64_128_KEY_SIZE,
> + .ivsize = SPECK64_BLOCK_SIZE,
> + .walksize = SPECK_NEON_CHUNK_SIZE,
> + .setkey = speck64_xts_setkey,
> + .encrypt = speck64_xts_encrypt,
> + .decrypt = speck64_xts_decrypt,
> + }
> +};
> +
> +static int __init speck_neon_module_init(void)
> +{
> + if (!(elf_hwcap & HWCAP_NEON))
> + return -ENODEV;
> + return crypto_register_skciphers(speck_algs, ARRAY_SIZE(speck_algs));
> +}
> +
> +static void __exit speck_neon_module_exit(void)
> +{
> + crypto_unregister_skciphers(speck_algs, ARRAY_SIZE(speck_algs));
> +}
> +
> +module_init(speck_neon_module_init);
> +module_exit(speck_neon_module_exit);
> +
> +MODULE_DESCRIPTION("Speck block cipher (NEON-accelerated)");
> +MODULE_LICENSE("GPL");
> +MODULE_AUTHOR("Eric Biggers <[email protected]>");
> +MODULE_ALIAS_CRYPTO("xts(speck128)");
> +MODULE_ALIAS_CRYPTO("xts-speck128-neon");
> +MODULE_ALIAS_CRYPTO("xts(speck64)");
> +MODULE_ALIAS_CRYPTO("xts-speck64-neon");
On 17 June 2018 at 00:40, Stefan Agner <[email protected]> wrote:
> Hi Eric,
>
> On 14.02.2018 19:42, Eric Biggers wrote:
>> Add an ARM NEON-accelerated implementation of Speck-XTS. It operates on
>> 128-byte chunks at a time, i.e. 8 blocks for Speck128 or 16 blocks for
>> Speck64. Each 128-byte chunk goes through XTS preprocessing, then is
>> encrypted/decrypted (doing one cipher round for all the blocks, then the
>> next round, etc.), then goes through XTS postprocessing.
>>
>> The performance depends on the processor but can be about 3 times faster
>> than the generic code. For example, on an ARMv7 processor we observe
>> the following performance with Speck128/256-XTS:
>>
>> xts-speck128-neon: Encryption 107.9 MB/s, Decryption 108.1 MB/s
>> xts(speck128-generic): Encryption 32.1 MB/s, Decryption 36.6 MB/s
>>
>> In comparison to AES-256-XTS without the Cryptography Extensions:
>>
>> xts-aes-neonbs: Encryption 41.2 MB/s, Decryption 36.7 MB/s
>> xts(aes-asm): Encryption 31.7 MB/s, Decryption 30.8 MB/s
>> xts(aes-generic): Encryption 21.2 MB/s, Decryption 20.9 MB/s
>>
>> Speck64/128-XTS is even faster:
>>
>> xts-speck64-neon: Encryption 138.6 MB/s, Decryption 139.1 MB/s
>>
>> Note that as with the generic code, only the Speck128 and Speck64
>> variants are supported. Also, for now only the XTS mode of operation is
>> supported, to target the disk and file encryption use cases. The NEON
>> code also only handles the portion of the data that is evenly divisible
>> into 128-byte chunks, with any remainder handled by a C fallback. Of
>> course, other modes of operation could be added later if needed, and/or
>> the NEON code could be updated to handle other buffer sizes.
>>
>> The XTS specification is only defined for AES which has a 128-bit block
>> size, so for the GF(2^64) math needed for Speck64-XTS we use the
>> reducing polynomial 'x^64 + x^4 + x^3 + x + 1' given by the original XEX
>> paper. Of course, when possible users should use Speck128-XTS, but even
>> that may be too slow on some processors; Speck64-XTS can be faster.
>>
>> Signed-off-by: Eric Biggers <[email protected]>
>> ---
>> arch/arm/crypto/Kconfig | 6 +
>> arch/arm/crypto/Makefile | 2 +
>> arch/arm/crypto/speck-neon-core.S | 432 ++++++++++++++++++++++++++++++
>> arch/arm/crypto/speck-neon-glue.c | 288 ++++++++++++++++++++
>> 4 files changed, 728 insertions(+)
>> create mode 100644 arch/arm/crypto/speck-neon-core.S
>> create mode 100644 arch/arm/crypto/speck-neon-glue.c
>>
>> diff --git a/arch/arm/crypto/Kconfig b/arch/arm/crypto/Kconfig
>> index b8e69fe282b8..925d1364727a 100644
>> --- a/arch/arm/crypto/Kconfig
>> +++ b/arch/arm/crypto/Kconfig
>> @@ -121,4 +121,10 @@ config CRYPTO_CHACHA20_NEON
>> select CRYPTO_BLKCIPHER
>> select CRYPTO_CHACHA20
>>
>> +config CRYPTO_SPECK_NEON
>> + tristate "NEON accelerated Speck cipher algorithms"
>> + depends on KERNEL_MODE_NEON
>> + select CRYPTO_BLKCIPHER
>> + select CRYPTO_SPECK
>> +
>> endif
>> diff --git a/arch/arm/crypto/Makefile b/arch/arm/crypto/Makefile
>> index 30ef8e291271..a758107c5525 100644
>> --- a/arch/arm/crypto/Makefile
>> +++ b/arch/arm/crypto/Makefile
>> @@ -10,6 +10,7 @@ obj-$(CONFIG_CRYPTO_SHA1_ARM_NEON) += sha1-arm-neon.o
>> obj-$(CONFIG_CRYPTO_SHA256_ARM) += sha256-arm.o
>> obj-$(CONFIG_CRYPTO_SHA512_ARM) += sha512-arm.o
>> obj-$(CONFIG_CRYPTO_CHACHA20_NEON) += chacha20-neon.o
>> +obj-$(CONFIG_CRYPTO_SPECK_NEON) += speck-neon.o
>>
>> ce-obj-$(CONFIG_CRYPTO_AES_ARM_CE) += aes-arm-ce.o
>> ce-obj-$(CONFIG_CRYPTO_SHA1_ARM_CE) += sha1-arm-ce.o
>> @@ -53,6 +54,7 @@ ghash-arm-ce-y := ghash-ce-core.o ghash-ce-glue.o
>> crct10dif-arm-ce-y := crct10dif-ce-core.o crct10dif-ce-glue.o
>> crc32-arm-ce-y:= crc32-ce-core.o crc32-ce-glue.o
>> chacha20-neon-y := chacha20-neon-core.o chacha20-neon-glue.o
>> +speck-neon-y := speck-neon-core.o speck-neon-glue.o
>>
>> quiet_cmd_perl = PERL $@
>> cmd_perl = $(PERL) $(<) > $(@)
>> diff --git a/arch/arm/crypto/speck-neon-core.S
>> b/arch/arm/crypto/speck-neon-core.S
>> new file mode 100644
>> index 000000000000..3c1e203e53b9
>> --- /dev/null
>> +++ b/arch/arm/crypto/speck-neon-core.S
>> @@ -0,0 +1,432 @@
>> +// SPDX-License-Identifier: GPL-2.0
>> +/*
>> + * NEON-accelerated implementation of Speck128-XTS and Speck64-XTS
>> + *
>> + * Copyright (c) 2018 Google, Inc
>> + *
>> + * Author: Eric Biggers <[email protected]>
>> + */
>> +
>> +#include <linux/linkage.h>
>> +
>> + .text
>> + .fpu neon
>> +
>> + // arguments
>> + ROUND_KEYS .req r0 // const {u64,u32} *round_keys
>> + NROUNDS .req r1 // int nrounds
>> + DST .req r2 // void *dst
>> + SRC .req r3 // const void *src
>> + NBYTES .req r4 // unsigned int nbytes
>> + TWEAK .req r5 // void *tweak
>> +
>> + // registers which hold the data being encrypted/decrypted
>> + X0 .req q0
>> + X0_L .req d0
>> + X0_H .req d1
>> + Y0 .req q1
>> + Y0_H .req d3
>> + X1 .req q2
>> + X1_L .req d4
>> + X1_H .req d5
>> + Y1 .req q3
>> + Y1_H .req d7
>> + X2 .req q4
>> + X2_L .req d8
>> + X2_H .req d9
>> + Y2 .req q5
>> + Y2_H .req d11
>> + X3 .req q6
>> + X3_L .req d12
>> + X3_H .req d13
>> + Y3 .req q7
>> + Y3_H .req d15
>> +
>> + // the round key, duplicated in all lanes
>> + ROUND_KEY .req q8
>> + ROUND_KEY_L .req d16
>> + ROUND_KEY_H .req d17
>> +
>> + // index vector for vtbl-based 8-bit rotates
>> + ROTATE_TABLE .req d18
>> +
>> + // multiplication table for updating XTS tweaks
>> + GF128MUL_TABLE .req d19
>> + GF64MUL_TABLE .req d19
>> +
>> + // current XTS tweak value(s)
>> + TWEAKV .req q10
>> + TWEAKV_L .req d20
>> + TWEAKV_H .req d21
>> +
>> + TMP0 .req q12
>> + TMP0_L .req d24
>> + TMP0_H .req d25
>> + TMP1 .req q13
>> + TMP2 .req q14
>> + TMP3 .req q15
>> +
>> + .align 4
>> +.Lror64_8_table:
>> + .byte 1, 2, 3, 4, 5, 6, 7, 0
>> +.Lror32_8_table:
>> + .byte 1, 2, 3, 0, 5, 6, 7, 4
>> +.Lrol64_8_table:
>> + .byte 7, 0, 1, 2, 3, 4, 5, 6
>> +.Lrol32_8_table:
>> + .byte 3, 0, 1, 2, 7, 4, 5, 6
>> +.Lgf128mul_table:
>> + .byte 0, 0x87
>> + .fill 14
>> +.Lgf64mul_table:
>> + .byte 0, 0x1b, (0x1b << 1), (0x1b << 1) ^ 0x1b
>> + .fill 12
>> +
>> +/*
>> + * _speck_round_128bytes() - Speck encryption round on 128 bytes at a time
>> + *
>> + * Do one Speck encryption round on the 128 bytes (8 blocks for
>> Speck128, 16 for
>> + * Speck64) stored in X0-X3 and Y0-Y3, using the round key stored in all lanes
>> + * of ROUND_KEY. 'n' is the lane size: 64 for Speck128, or 32 for Speck64.
>> + *
>> + * The 8-bit rotates are implemented using vtbl instead of vshr + vsli because
>> + * the vtbl approach is faster on some processors and the same speed on others.
>> + */
>> +.macro _speck_round_128bytes n
>> +
>> + // x = ror(x, 8)
>> + vtbl.8 X0_L, {X0_L}, ROTATE_TABLE
>> + vtbl.8 X0_H, {X0_H}, ROTATE_TABLE
>> + vtbl.8 X1_L, {X1_L}, ROTATE_TABLE
>> + vtbl.8 X1_H, {X1_H}, ROTATE_TABLE
>> + vtbl.8 X2_L, {X2_L}, ROTATE_TABLE
>> + vtbl.8 X2_H, {X2_H}, ROTATE_TABLE
>> + vtbl.8 X3_L, {X3_L}, ROTATE_TABLE
>> + vtbl.8 X3_H, {X3_H}, ROTATE_TABLE
>> +
>> + // x += y
>> + vadd.u\n X0, Y0
>> + vadd.u\n X1, Y1
>> + vadd.u\n X2, Y2
>> + vadd.u\n X3, Y3
>> +
>> + // x ^= k
>> + veor X0, ROUND_KEY
>> + veor X1, ROUND_KEY
>> + veor X2, ROUND_KEY
>> + veor X3, ROUND_KEY
>> +
>> + // y = rol(y, 3)
>> + vshl.u\n TMP0, Y0, #3
>> + vshl.u\n TMP1, Y1, #3
>> + vshl.u\n TMP2, Y2, #3
>> + vshl.u\n TMP3, Y3, #3
>> + vsri.u\n TMP0, Y0, #(\n - 3)
>> + vsri.u\n TMP1, Y1, #(\n - 3)
>> + vsri.u\n TMP2, Y2, #(\n - 3)
>> + vsri.u\n TMP3, Y3, #(\n - 3)
>> +
>> + // y ^= x
>> + veor Y0, TMP0, X0
>> + veor Y1, TMP1, X1
>> + veor Y2, TMP2, X2
>> + veor Y3, TMP3, X3
>> +.endm
>> +
>> +/*
>> + * _speck_unround_128bytes() - Speck decryption round on 128 bytes at a time
>> + *
>> + * This is the inverse of _speck_round_128bytes().
>> + */
>> +.macro _speck_unround_128bytes n
>> +
>> + // y ^= x
>> + veor TMP0, Y0, X0
>> + veor TMP1, Y1, X1
>> + veor TMP2, Y2, X2
>> + veor TMP3, Y3, X3
>> +
>> + // y = ror(y, 3)
>> + vshr.u\n Y0, TMP0, #3
>> + vshr.u\n Y1, TMP1, #3
>> + vshr.u\n Y2, TMP2, #3
>> + vshr.u\n Y3, TMP3, #3
>> + vsli.u\n Y0, TMP0, #(\n - 3)
>> + vsli.u\n Y1, TMP1, #(\n - 3)
>> + vsli.u\n Y2, TMP2, #(\n - 3)
>> + vsli.u\n Y3, TMP3, #(\n - 3)
>> +
>> + // x ^= k
>> + veor X0, ROUND_KEY
>> + veor X1, ROUND_KEY
>> + veor X2, ROUND_KEY
>> + veor X3, ROUND_KEY
>> +
>> + // x -= y
>> + vsub.u\n X0, Y0
>> + vsub.u\n X1, Y1
>> + vsub.u\n X2, Y2
>> + vsub.u\n X3, Y3
>> +
>> + // x = rol(x, 8);
>> + vtbl.8 X0_L, {X0_L}, ROTATE_TABLE
>> + vtbl.8 X0_H, {X0_H}, ROTATE_TABLE
>> + vtbl.8 X1_L, {X1_L}, ROTATE_TABLE
>> + vtbl.8 X1_H, {X1_H}, ROTATE_TABLE
>> + vtbl.8 X2_L, {X2_L}, ROTATE_TABLE
>> + vtbl.8 X2_H, {X2_H}, ROTATE_TABLE
>> + vtbl.8 X3_L, {X3_L}, ROTATE_TABLE
>> + vtbl.8 X3_H, {X3_H}, ROTATE_TABLE
>> +.endm
>> +
>> +.macro _xts128_precrypt_one dst_reg, tweak_buf, tmp
>> +
>> + // Load the next source block
>> + vld1.8 {\dst_reg}, [SRC]!
>> +
>> + // Save the current tweak in the tweak buffer
>> + vst1.8 {TWEAKV}, [\tweak_buf:128]!
>> +
>> + // XOR the next source block with the current tweak
>> + veor \dst_reg, TWEAKV
>> +
>> + /*
>> + * Calculate the next tweak by multiplying the current one by x,
>> + * modulo p(x) = x^128 + x^7 + x^2 + x + 1.
>> + */
>> + vshr.u64 \tmp, TWEAKV, #63
>> + vshl.u64 TWEAKV, #1
>> + veor TWEAKV_H, \tmp\()_L
>> + vtbl.8 \tmp\()_H, {GF128MUL_TABLE}, \tmp\()_H
>> + veor TWEAKV_L, \tmp\()_H
>> +.endm
>> +
>> +.macro _xts64_precrypt_two dst_reg, tweak_buf, tmp
>> +
>> + // Load the next two source blocks
>> + vld1.8 {\dst_reg}, [SRC]!
>> +
>> + // Save the current two tweaks in the tweak buffer
>> + vst1.8 {TWEAKV}, [\tweak_buf:128]!
>> +
>> + // XOR the next two source blocks with the current two tweaks
>> + veor \dst_reg, TWEAKV
>> +
>> + /*
>> + * Calculate the next two tweaks by multiplying the current ones by x^2,
>> + * modulo p(x) = x^64 + x^4 + x^3 + x + 1.
>> + */
>> + vshr.u64 \tmp, TWEAKV, #62
>> + vshl.u64 TWEAKV, #2
>> + vtbl.8 \tmp\()_L, {GF64MUL_TABLE}, \tmp\()_L
>> + vtbl.8 \tmp\()_H, {GF64MUL_TABLE}, \tmp\()_H
>> + veor TWEAKV, \tmp
>> +.endm
>> +
>> +/*
>> + * _speck_xts_crypt() - Speck-XTS encryption/decryption
>> + *
>> + * Encrypt or decrypt NBYTES bytes of data from the SRC buffer to the
>> DST buffer
>> + * using Speck-XTS, specifically the variant with a block size of
>> '2n' and round
>> + * count given by NROUNDS. The expanded round keys are given in
>> ROUND_KEYS, and
>> + * the current XTS tweak value is given in TWEAK. It's assumed that
>> NBYTES is a
>> + * nonzero multiple of 128.
>> + */
>> +.macro _speck_xts_crypt n, decrypting
>> + push {r4-r7}
>> + mov r7, sp
>> +
>> + /*
>> + * The first four parameters were passed in registers r0-r3. Load the
>> + * additional parameters, which were passed on the stack.
>> + */
>> + ldr NBYTES, [sp, #16]
>> + ldr TWEAK, [sp, #20]
>> +
>> + /*
>> + * If decrypting, modify the ROUND_KEYS parameter to point to the last
>> + * round key rather than the first, since for decryption the round keys
>> + * are used in reverse order.
>> + */
>> +.if \decrypting
>> +.if \n == 64
>> + add ROUND_KEYS, ROUND_KEYS, NROUNDS, lsl #3
>> + sub ROUND_KEYS, #8
>> +.else
>> + add ROUND_KEYS, ROUND_KEYS, NROUNDS, lsl #2
>> + sub ROUND_KEYS, #4
>> +.endif
>> +.endif
>> +
>> + // Load the index vector for vtbl-based 8-bit rotates
>> +.if \decrypting
>> + ldr r12, =.Lrol\n\()_8_table
>> +.else
>> + ldr r12, =.Lror\n\()_8_table
>> +.endif
>> + vld1.8 {ROTATE_TABLE}, [r12:64]
>> +
>> + // One-time XTS preparation
>> +
>> + /*
>> + * Allocate stack space to store 128 bytes worth of tweaks. For
>> + * performance, this space is aligned to a 16-byte boundary so that we
>> + * can use the load/store instructions that declare 16-byte alignment.
>> + */
>> + sub sp, #128
>> + bic sp, #0xf
>
>
> This fails here when building with CONFIG_THUMB2_KERNEL=y
>
> AS arch/arm/crypto/speck-neon-core.o
>
> arch/arm/crypto/speck-neon-core.S: Assembler messages:
>
> arch/arm/crypto/speck-neon-core.S:419: Error: r13 not allowed here --
> `bic sp,#0xf'
> arch/arm/crypto/speck-neon-core.S:423: Error: r13 not allowed here --
> `bic sp,#0xf'
> arch/arm/crypto/speck-neon-core.S:427: Error: r13 not allowed here --
> `bic sp,#0xf'
> arch/arm/crypto/speck-neon-core.S:431: Error: r13 not allowed here --
> `bic sp,#0xf'
>
> In a quick hack this change seems to address it:
>
>
> - sub sp, #128
> - bic sp, #0xf
> + mov r6, sp
> + sub r6, #128
> + bic r6, #0xf
> + mov sp, r6
>
> But there is probably a better solution to address this.
>
Given that there is no NEON on M class cores, I recommend we put something like
THUMB(bx pc)
THUMB(nop.w)
THUMB(.arm)
at the beginning and be done with it.
On 17 June 2018 at 11:30, Ard Biesheuvel <[email protected]> wrote:
> On 17 June 2018 at 00:40, Stefan Agner <[email protected]> wrote:
>> Hi Eric,
>>
>> On 14.02.2018 19:42, Eric Biggers wrote:
>>> Add an ARM NEON-accelerated implementation of Speck-XTS. It operates on
>>> 128-byte chunks at a time, i.e. 8 blocks for Speck128 or 16 blocks for
>>> Speck64. Each 128-byte chunk goes through XTS preprocessing, then is
>>> encrypted/decrypted (doing one cipher round for all the blocks, then the
>>> next round, etc.), then goes through XTS postprocessing.
>>>
>>> The performance depends on the processor but can be about 3 times faster
>>> than the generic code. For example, on an ARMv7 processor we observe
>>> the following performance with Speck128/256-XTS:
>>>
>>> xts-speck128-neon: Encryption 107.9 MB/s, Decryption 108.1 MB/s
>>> xts(speck128-generic): Encryption 32.1 MB/s, Decryption 36.6 MB/s
>>>
>>> In comparison to AES-256-XTS without the Cryptography Extensions:
>>>
>>> xts-aes-neonbs: Encryption 41.2 MB/s, Decryption 36.7 MB/s
>>> xts(aes-asm): Encryption 31.7 MB/s, Decryption 30.8 MB/s
>>> xts(aes-generic): Encryption 21.2 MB/s, Decryption 20.9 MB/s
>>>
>>> Speck64/128-XTS is even faster:
>>>
>>> xts-speck64-neon: Encryption 138.6 MB/s, Decryption 139.1 MB/s
>>>
>>> Note that as with the generic code, only the Speck128 and Speck64
>>> variants are supported. Also, for now only the XTS mode of operation is
>>> supported, to target the disk and file encryption use cases. The NEON
>>> code also only handles the portion of the data that is evenly divisible
>>> into 128-byte chunks, with any remainder handled by a C fallback. Of
>>> course, other modes of operation could be added later if needed, and/or
>>> the NEON code could be updated to handle other buffer sizes.
>>>
>>> The XTS specification is only defined for AES which has a 128-bit block
>>> size, so for the GF(2^64) math needed for Speck64-XTS we use the
>>> reducing polynomial 'x^64 + x^4 + x^3 + x + 1' given by the original XEX
>>> paper. Of course, when possible users should use Speck128-XTS, but even
>>> that may be too slow on some processors; Speck64-XTS can be faster.
>>>
>>> Signed-off-by: Eric Biggers <[email protected]>
>>> ---
>>> arch/arm/crypto/Kconfig | 6 +
>>> arch/arm/crypto/Makefile | 2 +
>>> arch/arm/crypto/speck-neon-core.S | 432 ++++++++++++++++++++++++++++++
>>> arch/arm/crypto/speck-neon-glue.c | 288 ++++++++++++++++++++
>>> 4 files changed, 728 insertions(+)
>>> create mode 100644 arch/arm/crypto/speck-neon-core.S
>>> create mode 100644 arch/arm/crypto/speck-neon-glue.c
>>>
>>> diff --git a/arch/arm/crypto/Kconfig b/arch/arm/crypto/Kconfig
>>> index b8e69fe282b8..925d1364727a 100644
>>> --- a/arch/arm/crypto/Kconfig
>>> +++ b/arch/arm/crypto/Kconfig
>>> @@ -121,4 +121,10 @@ config CRYPTO_CHACHA20_NEON
>>> select CRYPTO_BLKCIPHER
>>> select CRYPTO_CHACHA20
>>>
>>> +config CRYPTO_SPECK_NEON
>>> + tristate "NEON accelerated Speck cipher algorithms"
>>> + depends on KERNEL_MODE_NEON
>>> + select CRYPTO_BLKCIPHER
>>> + select CRYPTO_SPECK
>>> +
>>> endif
>>> diff --git a/arch/arm/crypto/Makefile b/arch/arm/crypto/Makefile
>>> index 30ef8e291271..a758107c5525 100644
>>> --- a/arch/arm/crypto/Makefile
>>> +++ b/arch/arm/crypto/Makefile
>>> @@ -10,6 +10,7 @@ obj-$(CONFIG_CRYPTO_SHA1_ARM_NEON) += sha1-arm-neon.o
>>> obj-$(CONFIG_CRYPTO_SHA256_ARM) += sha256-arm.o
>>> obj-$(CONFIG_CRYPTO_SHA512_ARM) += sha512-arm.o
>>> obj-$(CONFIG_CRYPTO_CHACHA20_NEON) += chacha20-neon.o
>>> +obj-$(CONFIG_CRYPTO_SPECK_NEON) += speck-neon.o
>>>
>>> ce-obj-$(CONFIG_CRYPTO_AES_ARM_CE) += aes-arm-ce.o
>>> ce-obj-$(CONFIG_CRYPTO_SHA1_ARM_CE) += sha1-arm-ce.o
>>> @@ -53,6 +54,7 @@ ghash-arm-ce-y := ghash-ce-core.o ghash-ce-glue.o
>>> crct10dif-arm-ce-y := crct10dif-ce-core.o crct10dif-ce-glue.o
>>> crc32-arm-ce-y:= crc32-ce-core.o crc32-ce-glue.o
>>> chacha20-neon-y := chacha20-neon-core.o chacha20-neon-glue.o
>>> +speck-neon-y := speck-neon-core.o speck-neon-glue.o
>>>
>>> quiet_cmd_perl = PERL $@
>>> cmd_perl = $(PERL) $(<) > $(@)
>>> diff --git a/arch/arm/crypto/speck-neon-core.S
>>> b/arch/arm/crypto/speck-neon-core.S
>>> new file mode 100644
>>> index 000000000000..3c1e203e53b9
>>> --- /dev/null
>>> +++ b/arch/arm/crypto/speck-neon-core.S
>>> @@ -0,0 +1,432 @@
>>> +// SPDX-License-Identifier: GPL-2.0
>>> +/*
>>> + * NEON-accelerated implementation of Speck128-XTS and Speck64-XTS
>>> + *
>>> + * Copyright (c) 2018 Google, Inc
>>> + *
>>> + * Author: Eric Biggers <[email protected]>
>>> + */
>>> +
>>> +#include <linux/linkage.h>
>>> +
>>> + .text
>>> + .fpu neon
>>> +
>>> + // arguments
>>> + ROUND_KEYS .req r0 // const {u64,u32} *round_keys
>>> + NROUNDS .req r1 // int nrounds
>>> + DST .req r2 // void *dst
>>> + SRC .req r3 // const void *src
>>> + NBYTES .req r4 // unsigned int nbytes
>>> + TWEAK .req r5 // void *tweak
>>> +
>>> + // registers which hold the data being encrypted/decrypted
>>> + X0 .req q0
>>> + X0_L .req d0
>>> + X0_H .req d1
>>> + Y0 .req q1
>>> + Y0_H .req d3
>>> + X1 .req q2
>>> + X1_L .req d4
>>> + X1_H .req d5
>>> + Y1 .req q3
>>> + Y1_H .req d7
>>> + X2 .req q4
>>> + X2_L .req d8
>>> + X2_H .req d9
>>> + Y2 .req q5
>>> + Y2_H .req d11
>>> + X3 .req q6
>>> + X3_L .req d12
>>> + X3_H .req d13
>>> + Y3 .req q7
>>> + Y3_H .req d15
>>> +
>>> + // the round key, duplicated in all lanes
>>> + ROUND_KEY .req q8
>>> + ROUND_KEY_L .req d16
>>> + ROUND_KEY_H .req d17
>>> +
>>> + // index vector for vtbl-based 8-bit rotates
>>> + ROTATE_TABLE .req d18
>>> +
>>> + // multiplication table for updating XTS tweaks
>>> + GF128MUL_TABLE .req d19
>>> + GF64MUL_TABLE .req d19
>>> +
>>> + // current XTS tweak value(s)
>>> + TWEAKV .req q10
>>> + TWEAKV_L .req d20
>>> + TWEAKV_H .req d21
>>> +
>>> + TMP0 .req q12
>>> + TMP0_L .req d24
>>> + TMP0_H .req d25
>>> + TMP1 .req q13
>>> + TMP2 .req q14
>>> + TMP3 .req q15
>>> +
>>> + .align 4
>>> +.Lror64_8_table:
>>> + .byte 1, 2, 3, 4, 5, 6, 7, 0
>>> +.Lror32_8_table:
>>> + .byte 1, 2, 3, 0, 5, 6, 7, 4
>>> +.Lrol64_8_table:
>>> + .byte 7, 0, 1, 2, 3, 4, 5, 6
>>> +.Lrol32_8_table:
>>> + .byte 3, 0, 1, 2, 7, 4, 5, 6
>>> +.Lgf128mul_table:
>>> + .byte 0, 0x87
>>> + .fill 14
>>> +.Lgf64mul_table:
>>> + .byte 0, 0x1b, (0x1b << 1), (0x1b << 1) ^ 0x1b
>>> + .fill 12
>>> +
>>> +/*
>>> + * _speck_round_128bytes() - Speck encryption round on 128 bytes at a time
>>> + *
>>> + * Do one Speck encryption round on the 128 bytes (8 blocks for
>>> Speck128, 16 for
>>> + * Speck64) stored in X0-X3 and Y0-Y3, using the round key stored in all lanes
>>> + * of ROUND_KEY. 'n' is the lane size: 64 for Speck128, or 32 for Speck64.
>>> + *
>>> + * The 8-bit rotates are implemented using vtbl instead of vshr + vsli because
>>> + * the vtbl approach is faster on some processors and the same speed on others.
>>> + */
>>> +.macro _speck_round_128bytes n
>>> +
>>> + // x = ror(x, 8)
>>> + vtbl.8 X0_L, {X0_L}, ROTATE_TABLE
>>> + vtbl.8 X0_H, {X0_H}, ROTATE_TABLE
>>> + vtbl.8 X1_L, {X1_L}, ROTATE_TABLE
>>> + vtbl.8 X1_H, {X1_H}, ROTATE_TABLE
>>> + vtbl.8 X2_L, {X2_L}, ROTATE_TABLE
>>> + vtbl.8 X2_H, {X2_H}, ROTATE_TABLE
>>> + vtbl.8 X3_L, {X3_L}, ROTATE_TABLE
>>> + vtbl.8 X3_H, {X3_H}, ROTATE_TABLE
>>> +
>>> + // x += y
>>> + vadd.u\n X0, Y0
>>> + vadd.u\n X1, Y1
>>> + vadd.u\n X2, Y2
>>> + vadd.u\n X3, Y3
>>> +
>>> + // x ^= k
>>> + veor X0, ROUND_KEY
>>> + veor X1, ROUND_KEY
>>> + veor X2, ROUND_KEY
>>> + veor X3, ROUND_KEY
>>> +
>>> + // y = rol(y, 3)
>>> + vshl.u\n TMP0, Y0, #3
>>> + vshl.u\n TMP1, Y1, #3
>>> + vshl.u\n TMP2, Y2, #3
>>> + vshl.u\n TMP3, Y3, #3
>>> + vsri.u\n TMP0, Y0, #(\n - 3)
>>> + vsri.u\n TMP1, Y1, #(\n - 3)
>>> + vsri.u\n TMP2, Y2, #(\n - 3)
>>> + vsri.u\n TMP3, Y3, #(\n - 3)
>>> +
>>> + // y ^= x
>>> + veor Y0, TMP0, X0
>>> + veor Y1, TMP1, X1
>>> + veor Y2, TMP2, X2
>>> + veor Y3, TMP3, X3
>>> +.endm
>>> +
>>> +/*
>>> + * _speck_unround_128bytes() - Speck decryption round on 128 bytes at a time
>>> + *
>>> + * This is the inverse of _speck_round_128bytes().
>>> + */
>>> +.macro _speck_unround_128bytes n
>>> +
>>> + // y ^= x
>>> + veor TMP0, Y0, X0
>>> + veor TMP1, Y1, X1
>>> + veor TMP2, Y2, X2
>>> + veor TMP3, Y3, X3
>>> +
>>> + // y = ror(y, 3)
>>> + vshr.u\n Y0, TMP0, #3
>>> + vshr.u\n Y1, TMP1, #3
>>> + vshr.u\n Y2, TMP2, #3
>>> + vshr.u\n Y3, TMP3, #3
>>> + vsli.u\n Y0, TMP0, #(\n - 3)
>>> + vsli.u\n Y1, TMP1, #(\n - 3)
>>> + vsli.u\n Y2, TMP2, #(\n - 3)
>>> + vsli.u\n Y3, TMP3, #(\n - 3)
>>> +
>>> + // x ^= k
>>> + veor X0, ROUND_KEY
>>> + veor X1, ROUND_KEY
>>> + veor X2, ROUND_KEY
>>> + veor X3, ROUND_KEY
>>> +
>>> + // x -= y
>>> + vsub.u\n X0, Y0
>>> + vsub.u\n X1, Y1
>>> + vsub.u\n X2, Y2
>>> + vsub.u\n X3, Y3
>>> +
>>> + // x = rol(x, 8);
>>> + vtbl.8 X0_L, {X0_L}, ROTATE_TABLE
>>> + vtbl.8 X0_H, {X0_H}, ROTATE_TABLE
>>> + vtbl.8 X1_L, {X1_L}, ROTATE_TABLE
>>> + vtbl.8 X1_H, {X1_H}, ROTATE_TABLE
>>> + vtbl.8 X2_L, {X2_L}, ROTATE_TABLE
>>> + vtbl.8 X2_H, {X2_H}, ROTATE_TABLE
>>> + vtbl.8 X3_L, {X3_L}, ROTATE_TABLE
>>> + vtbl.8 X3_H, {X3_H}, ROTATE_TABLE
>>> +.endm
>>> +
>>> +.macro _xts128_precrypt_one dst_reg, tweak_buf, tmp
>>> +
>>> + // Load the next source block
>>> + vld1.8 {\dst_reg}, [SRC]!
>>> +
>>> + // Save the current tweak in the tweak buffer
>>> + vst1.8 {TWEAKV}, [\tweak_buf:128]!
>>> +
>>> + // XOR the next source block with the current tweak
>>> + veor \dst_reg, TWEAKV
>>> +
>>> + /*
>>> + * Calculate the next tweak by multiplying the current one by x,
>>> + * modulo p(x) = x^128 + x^7 + x^2 + x + 1.
>>> + */
>>> + vshr.u64 \tmp, TWEAKV, #63
>>> + vshl.u64 TWEAKV, #1
>>> + veor TWEAKV_H, \tmp\()_L
>>> + vtbl.8 \tmp\()_H, {GF128MUL_TABLE}, \tmp\()_H
>>> + veor TWEAKV_L, \tmp\()_H
>>> +.endm
>>> +
>>> +.macro _xts64_precrypt_two dst_reg, tweak_buf, tmp
>>> +
>>> + // Load the next two source blocks
>>> + vld1.8 {\dst_reg}, [SRC]!
>>> +
>>> + // Save the current two tweaks in the tweak buffer
>>> + vst1.8 {TWEAKV}, [\tweak_buf:128]!
>>> +
>>> + // XOR the next two source blocks with the current two tweaks
>>> + veor \dst_reg, TWEAKV
>>> +
>>> + /*
>>> + * Calculate the next two tweaks by multiplying the current ones by x^2,
>>> + * modulo p(x) = x^64 + x^4 + x^3 + x + 1.
>>> + */
>>> + vshr.u64 \tmp, TWEAKV, #62
>>> + vshl.u64 TWEAKV, #2
>>> + vtbl.8 \tmp\()_L, {GF64MUL_TABLE}, \tmp\()_L
>>> + vtbl.8 \tmp\()_H, {GF64MUL_TABLE}, \tmp\()_H
>>> + veor TWEAKV, \tmp
>>> +.endm
>>> +
>>> +/*
>>> + * _speck_xts_crypt() - Speck-XTS encryption/decryption
>>> + *
>>> + * Encrypt or decrypt NBYTES bytes of data from the SRC buffer to the
>>> DST buffer
>>> + * using Speck-XTS, specifically the variant with a block size of
>>> '2n' and round
>>> + * count given by NROUNDS. The expanded round keys are given in
>>> ROUND_KEYS, and
>>> + * the current XTS tweak value is given in TWEAK. It's assumed that
>>> NBYTES is a
>>> + * nonzero multiple of 128.
>>> + */
>>> +.macro _speck_xts_crypt n, decrypting
>>> + push {r4-r7}
>>> + mov r7, sp
>>> +
>>> + /*
>>> + * The first four parameters were passed in registers r0-r3. Load the
>>> + * additional parameters, which were passed on the stack.
>>> + */
>>> + ldr NBYTES, [sp, #16]
>>> + ldr TWEAK, [sp, #20]
>>> +
>>> + /*
>>> + * If decrypting, modify the ROUND_KEYS parameter to point to the last
>>> + * round key rather than the first, since for decryption the round keys
>>> + * are used in reverse order.
>>> + */
>>> +.if \decrypting
>>> +.if \n == 64
>>> + add ROUND_KEYS, ROUND_KEYS, NROUNDS, lsl #3
>>> + sub ROUND_KEYS, #8
>>> +.else
>>> + add ROUND_KEYS, ROUND_KEYS, NROUNDS, lsl #2
>>> + sub ROUND_KEYS, #4
>>> +.endif
>>> +.endif
>>> +
>>> + // Load the index vector for vtbl-based 8-bit rotates
>>> +.if \decrypting
>>> + ldr r12, =.Lrol\n\()_8_table
>>> +.else
>>> + ldr r12, =.Lror\n\()_8_table
>>> +.endif
>>> + vld1.8 {ROTATE_TABLE}, [r12:64]
>>> +
>>> + // One-time XTS preparation
>>> +
>>> + /*
>>> + * Allocate stack space to store 128 bytes worth of tweaks. For
>>> + * performance, this space is aligned to a 16-byte boundary so that we
>>> + * can use the load/store instructions that declare 16-byte alignment.
>>> + */
>>> + sub sp, #128
>>> + bic sp, #0xf
>>
>>
>> This fails here when building with CONFIG_THUMB2_KERNEL=y
>>
>> AS arch/arm/crypto/speck-neon-core.o
>>
>> arch/arm/crypto/speck-neon-core.S: Assembler messages:
>>
>> arch/arm/crypto/speck-neon-core.S:419: Error: r13 not allowed here --
>> `bic sp,#0xf'
>> arch/arm/crypto/speck-neon-core.S:423: Error: r13 not allowed here --
>> `bic sp,#0xf'
>> arch/arm/crypto/speck-neon-core.S:427: Error: r13 not allowed here --
>> `bic sp,#0xf'
>> arch/arm/crypto/speck-neon-core.S:431: Error: r13 not allowed here --
>> `bic sp,#0xf'
>>
>> In a quick hack this change seems to address it:
>>
>>
>> - sub sp, #128
>> - bic sp, #0xf
>> + mov r6, sp
>> + sub r6, #128
>> + bic r6, #0xf
>> + mov sp, r6
>>
>> But there is probably a better solution to address this.
>>
>
> Given that there is no NEON on M class cores, I recommend we put something like
>
> THUMB(bx pc)
> THUMB(nop.w)
> THUMB(.arm)
>
> at the beginning and be done with it.
I mean nop.n or just nop, of course, and we may need a '.align 2' at
the beginning as well.
On 17.06.2018 11:40, Ard Biesheuvel wrote:
> On 17 June 2018 at 11:30, Ard Biesheuvel <[email protected]> wrote:
>> On 17 June 2018 at 00:40, Stefan Agner <[email protected]> wrote:
>>> Hi Eric,
>>>
>>> On 14.02.2018 19:42, Eric Biggers wrote:
>>>> Add an ARM NEON-accelerated implementation of Speck-XTS. It operates on
>>>> 128-byte chunks at a time, i.e. 8 blocks for Speck128 or 16 blocks for
>>>> Speck64. Each 128-byte chunk goes through XTS preprocessing, then is
>>>> encrypted/decrypted (doing one cipher round for all the blocks, then the
>>>> next round, etc.), then goes through XTS postprocessing.
>>>>
>>>> The performance depends on the processor but can be about 3 times faster
>>>> than the generic code. For example, on an ARMv7 processor we observe
>>>> the following performance with Speck128/256-XTS:
>>>>
>>>> xts-speck128-neon: Encryption 107.9 MB/s, Decryption 108.1 MB/s
>>>> xts(speck128-generic): Encryption 32.1 MB/s, Decryption 36.6 MB/s
>>>>
>>>> In comparison to AES-256-XTS without the Cryptography Extensions:
>>>>
>>>> xts-aes-neonbs: Encryption 41.2 MB/s, Decryption 36.7 MB/s
>>>> xts(aes-asm): Encryption 31.7 MB/s, Decryption 30.8 MB/s
>>>> xts(aes-generic): Encryption 21.2 MB/s, Decryption 20.9 MB/s
>>>>
>>>> Speck64/128-XTS is even faster:
>>>>
>>>> xts-speck64-neon: Encryption 138.6 MB/s, Decryption 139.1 MB/s
>>>>
>>>> Note that as with the generic code, only the Speck128 and Speck64
>>>> variants are supported. Also, for now only the XTS mode of operation is
>>>> supported, to target the disk and file encryption use cases. The NEON
>>>> code also only handles the portion of the data that is evenly divisible
>>>> into 128-byte chunks, with any remainder handled by a C fallback. Of
>>>> course, other modes of operation could be added later if needed, and/or
>>>> the NEON code could be updated to handle other buffer sizes.
>>>>
>>>> The XTS specification is only defined for AES which has a 128-bit block
>>>> size, so for the GF(2^64) math needed for Speck64-XTS we use the
>>>> reducing polynomial 'x^64 + x^4 + x^3 + x + 1' given by the original XEX
>>>> paper. Of course, when possible users should use Speck128-XTS, but even
>>>> that may be too slow on some processors; Speck64-XTS can be faster.
>>>>
>>>> Signed-off-by: Eric Biggers <[email protected]>
>>>> ---
>>>> arch/arm/crypto/Kconfig | 6 +
>>>> arch/arm/crypto/Makefile | 2 +
>>>> arch/arm/crypto/speck-neon-core.S | 432 ++++++++++++++++++++++++++++++
>>>> arch/arm/crypto/speck-neon-glue.c | 288 ++++++++++++++++++++
>>>> 4 files changed, 728 insertions(+)
>>>> create mode 100644 arch/arm/crypto/speck-neon-core.S
>>>> create mode 100644 arch/arm/crypto/speck-neon-glue.c
>>>>
>>>> diff --git a/arch/arm/crypto/Kconfig b/arch/arm/crypto/Kconfig
>>>> index b8e69fe282b8..925d1364727a 100644
>>>> --- a/arch/arm/crypto/Kconfig
>>>> +++ b/arch/arm/crypto/Kconfig
>>>> @@ -121,4 +121,10 @@ config CRYPTO_CHACHA20_NEON
>>>> select CRYPTO_BLKCIPHER
>>>> select CRYPTO_CHACHA20
>>>>
>>>> +config CRYPTO_SPECK_NEON
>>>> + tristate "NEON accelerated Speck cipher algorithms"
>>>> + depends on KERNEL_MODE_NEON
>>>> + select CRYPTO_BLKCIPHER
>>>> + select CRYPTO_SPECK
>>>> +
>>>> endif
>>>> diff --git a/arch/arm/crypto/Makefile b/arch/arm/crypto/Makefile
>>>> index 30ef8e291271..a758107c5525 100644
>>>> --- a/arch/arm/crypto/Makefile
>>>> +++ b/arch/arm/crypto/Makefile
>>>> @@ -10,6 +10,7 @@ obj-$(CONFIG_CRYPTO_SHA1_ARM_NEON) += sha1-arm-neon.o
>>>> obj-$(CONFIG_CRYPTO_SHA256_ARM) += sha256-arm.o
>>>> obj-$(CONFIG_CRYPTO_SHA512_ARM) += sha512-arm.o
>>>> obj-$(CONFIG_CRYPTO_CHACHA20_NEON) += chacha20-neon.o
>>>> +obj-$(CONFIG_CRYPTO_SPECK_NEON) += speck-neon.o
>>>>
>>>> ce-obj-$(CONFIG_CRYPTO_AES_ARM_CE) += aes-arm-ce.o
>>>> ce-obj-$(CONFIG_CRYPTO_SHA1_ARM_CE) += sha1-arm-ce.o
>>>> @@ -53,6 +54,7 @@ ghash-arm-ce-y := ghash-ce-core.o ghash-ce-glue.o
>>>> crct10dif-arm-ce-y := crct10dif-ce-core.o crct10dif-ce-glue.o
>>>> crc32-arm-ce-y:= crc32-ce-core.o crc32-ce-glue.o
>>>> chacha20-neon-y := chacha20-neon-core.o chacha20-neon-glue.o
>>>> +speck-neon-y := speck-neon-core.o speck-neon-glue.o
>>>>
>>>> quiet_cmd_perl = PERL $@
>>>> cmd_perl = $(PERL) $(<) > $(@)
>>>> diff --git a/arch/arm/crypto/speck-neon-core.S
>>>> b/arch/arm/crypto/speck-neon-core.S
>>>> new file mode 100644
>>>> index 000000000000..3c1e203e53b9
>>>> --- /dev/null
>>>> +++ b/arch/arm/crypto/speck-neon-core.S
>>>> @@ -0,0 +1,432 @@
>>>> +// SPDX-License-Identifier: GPL-2.0
>>>> +/*
>>>> + * NEON-accelerated implementation of Speck128-XTS and Speck64-XTS
>>>> + *
>>>> + * Copyright (c) 2018 Google, Inc
>>>> + *
>>>> + * Author: Eric Biggers <[email protected]>
>>>> + */
>>>> +
>>>> +#include <linux/linkage.h>
>>>> +
>>>> + .text
>>>> + .fpu neon
>>>> +
>>>> + // arguments
>>>> + ROUND_KEYS .req r0 // const {u64,u32} *round_keys
>>>> + NROUNDS .req r1 // int nrounds
>>>> + DST .req r2 // void *dst
>>>> + SRC .req r3 // const void *src
>>>> + NBYTES .req r4 // unsigned int nbytes
>>>> + TWEAK .req r5 // void *tweak
>>>> +
>>>> + // registers which hold the data being encrypted/decrypted
>>>> + X0 .req q0
>>>> + X0_L .req d0
>>>> + X0_H .req d1
>>>> + Y0 .req q1
>>>> + Y0_H .req d3
>>>> + X1 .req q2
>>>> + X1_L .req d4
>>>> + X1_H .req d5
>>>> + Y1 .req q3
>>>> + Y1_H .req d7
>>>> + X2 .req q4
>>>> + X2_L .req d8
>>>> + X2_H .req d9
>>>> + Y2 .req q5
>>>> + Y2_H .req d11
>>>> + X3 .req q6
>>>> + X3_L .req d12
>>>> + X3_H .req d13
>>>> + Y3 .req q7
>>>> + Y3_H .req d15
>>>> +
>>>> + // the round key, duplicated in all lanes
>>>> + ROUND_KEY .req q8
>>>> + ROUND_KEY_L .req d16
>>>> + ROUND_KEY_H .req d17
>>>> +
>>>> + // index vector for vtbl-based 8-bit rotates
>>>> + ROTATE_TABLE .req d18
>>>> +
>>>> + // multiplication table for updating XTS tweaks
>>>> + GF128MUL_TABLE .req d19
>>>> + GF64MUL_TABLE .req d19
>>>> +
>>>> + // current XTS tweak value(s)
>>>> + TWEAKV .req q10
>>>> + TWEAKV_L .req d20
>>>> + TWEAKV_H .req d21
>>>> +
>>>> + TMP0 .req q12
>>>> + TMP0_L .req d24
>>>> + TMP0_H .req d25
>>>> + TMP1 .req q13
>>>> + TMP2 .req q14
>>>> + TMP3 .req q15
>>>> +
>>>> + .align 4
>>>> +.Lror64_8_table:
>>>> + .byte 1, 2, 3, 4, 5, 6, 7, 0
>>>> +.Lror32_8_table:
>>>> + .byte 1, 2, 3, 0, 5, 6, 7, 4
>>>> +.Lrol64_8_table:
>>>> + .byte 7, 0, 1, 2, 3, 4, 5, 6
>>>> +.Lrol32_8_table:
>>>> + .byte 3, 0, 1, 2, 7, 4, 5, 6
>>>> +.Lgf128mul_table:
>>>> + .byte 0, 0x87
>>>> + .fill 14
>>>> +.Lgf64mul_table:
>>>> + .byte 0, 0x1b, (0x1b << 1), (0x1b << 1) ^ 0x1b
>>>> + .fill 12
>>>> +
>>>> +/*
>>>> + * _speck_round_128bytes() - Speck encryption round on 128 bytes at a time
>>>> + *
>>>> + * Do one Speck encryption round on the 128 bytes (8 blocks for
>>>> Speck128, 16 for
>>>> + * Speck64) stored in X0-X3 and Y0-Y3, using the round key stored in all lanes
>>>> + * of ROUND_KEY. 'n' is the lane size: 64 for Speck128, or 32 for Speck64.
>>>> + *
>>>> + * The 8-bit rotates are implemented using vtbl instead of vshr + vsli because
>>>> + * the vtbl approach is faster on some processors and the same speed on others.
>>>> + */
>>>> +.macro _speck_round_128bytes n
>>>> +
>>>> + // x = ror(x, 8)
>>>> + vtbl.8 X0_L, {X0_L}, ROTATE_TABLE
>>>> + vtbl.8 X0_H, {X0_H}, ROTATE_TABLE
>>>> + vtbl.8 X1_L, {X1_L}, ROTATE_TABLE
>>>> + vtbl.8 X1_H, {X1_H}, ROTATE_TABLE
>>>> + vtbl.8 X2_L, {X2_L}, ROTATE_TABLE
>>>> + vtbl.8 X2_H, {X2_H}, ROTATE_TABLE
>>>> + vtbl.8 X3_L, {X3_L}, ROTATE_TABLE
>>>> + vtbl.8 X3_H, {X3_H}, ROTATE_TABLE
>>>> +
>>>> + // x += y
>>>> + vadd.u\n X0, Y0
>>>> + vadd.u\n X1, Y1
>>>> + vadd.u\n X2, Y2
>>>> + vadd.u\n X3, Y3
>>>> +
>>>> + // x ^= k
>>>> + veor X0, ROUND_KEY
>>>> + veor X1, ROUND_KEY
>>>> + veor X2, ROUND_KEY
>>>> + veor X3, ROUND_KEY
>>>> +
>>>> + // y = rol(y, 3)
>>>> + vshl.u\n TMP0, Y0, #3
>>>> + vshl.u\n TMP1, Y1, #3
>>>> + vshl.u\n TMP2, Y2, #3
>>>> + vshl.u\n TMP3, Y3, #3
>>>> + vsri.u\n TMP0, Y0, #(\n - 3)
>>>> + vsri.u\n TMP1, Y1, #(\n - 3)
>>>> + vsri.u\n TMP2, Y2, #(\n - 3)
>>>> + vsri.u\n TMP3, Y3, #(\n - 3)
>>>> +
>>>> + // y ^= x
>>>> + veor Y0, TMP0, X0
>>>> + veor Y1, TMP1, X1
>>>> + veor Y2, TMP2, X2
>>>> + veor Y3, TMP3, X3
>>>> +.endm
>>>> +
>>>> +/*
>>>> + * _speck_unround_128bytes() - Speck decryption round on 128 bytes at a time
>>>> + *
>>>> + * This is the inverse of _speck_round_128bytes().
>>>> + */
>>>> +.macro _speck_unround_128bytes n
>>>> +
>>>> + // y ^= x
>>>> + veor TMP0, Y0, X0
>>>> + veor TMP1, Y1, X1
>>>> + veor TMP2, Y2, X2
>>>> + veor TMP3, Y3, X3
>>>> +
>>>> + // y = ror(y, 3)
>>>> + vshr.u\n Y0, TMP0, #3
>>>> + vshr.u\n Y1, TMP1, #3
>>>> + vshr.u\n Y2, TMP2, #3
>>>> + vshr.u\n Y3, TMP3, #3
>>>> + vsli.u\n Y0, TMP0, #(\n - 3)
>>>> + vsli.u\n Y1, TMP1, #(\n - 3)
>>>> + vsli.u\n Y2, TMP2, #(\n - 3)
>>>> + vsli.u\n Y3, TMP3, #(\n - 3)
>>>> +
>>>> + // x ^= k
>>>> + veor X0, ROUND_KEY
>>>> + veor X1, ROUND_KEY
>>>> + veor X2, ROUND_KEY
>>>> + veor X3, ROUND_KEY
>>>> +
>>>> + // x -= y
>>>> + vsub.u\n X0, Y0
>>>> + vsub.u\n X1, Y1
>>>> + vsub.u\n X2, Y2
>>>> + vsub.u\n X3, Y3
>>>> +
>>>> + // x = rol(x, 8);
>>>> + vtbl.8 X0_L, {X0_L}, ROTATE_TABLE
>>>> + vtbl.8 X0_H, {X0_H}, ROTATE_TABLE
>>>> + vtbl.8 X1_L, {X1_L}, ROTATE_TABLE
>>>> + vtbl.8 X1_H, {X1_H}, ROTATE_TABLE
>>>> + vtbl.8 X2_L, {X2_L}, ROTATE_TABLE
>>>> + vtbl.8 X2_H, {X2_H}, ROTATE_TABLE
>>>> + vtbl.8 X3_L, {X3_L}, ROTATE_TABLE
>>>> + vtbl.8 X3_H, {X3_H}, ROTATE_TABLE
>>>> +.endm
>>>> +
>>>> +.macro _xts128_precrypt_one dst_reg, tweak_buf, tmp
>>>> +
>>>> + // Load the next source block
>>>> + vld1.8 {\dst_reg}, [SRC]!
>>>> +
>>>> + // Save the current tweak in the tweak buffer
>>>> + vst1.8 {TWEAKV}, [\tweak_buf:128]!
>>>> +
>>>> + // XOR the next source block with the current tweak
>>>> + veor \dst_reg, TWEAKV
>>>> +
>>>> + /*
>>>> + * Calculate the next tweak by multiplying the current one by x,
>>>> + * modulo p(x) = x^128 + x^7 + x^2 + x + 1.
>>>> + */
>>>> + vshr.u64 \tmp, TWEAKV, #63
>>>> + vshl.u64 TWEAKV, #1
>>>> + veor TWEAKV_H, \tmp\()_L
>>>> + vtbl.8 \tmp\()_H, {GF128MUL_TABLE}, \tmp\()_H
>>>> + veor TWEAKV_L, \tmp\()_H
>>>> +.endm
>>>> +
>>>> +.macro _xts64_precrypt_two dst_reg, tweak_buf, tmp
>>>> +
>>>> + // Load the next two source blocks
>>>> + vld1.8 {\dst_reg}, [SRC]!
>>>> +
>>>> + // Save the current two tweaks in the tweak buffer
>>>> + vst1.8 {TWEAKV}, [\tweak_buf:128]!
>>>> +
>>>> + // XOR the next two source blocks with the current two tweaks
>>>> + veor \dst_reg, TWEAKV
>>>> +
>>>> + /*
>>>> + * Calculate the next two tweaks by multiplying the current ones by x^2,
>>>> + * modulo p(x) = x^64 + x^4 + x^3 + x + 1.
>>>> + */
>>>> + vshr.u64 \tmp, TWEAKV, #62
>>>> + vshl.u64 TWEAKV, #2
>>>> + vtbl.8 \tmp\()_L, {GF64MUL_TABLE}, \tmp\()_L
>>>> + vtbl.8 \tmp\()_H, {GF64MUL_TABLE}, \tmp\()_H
>>>> + veor TWEAKV, \tmp
>>>> +.endm
>>>> +
>>>> +/*
>>>> + * _speck_xts_crypt() - Speck-XTS encryption/decryption
>>>> + *
>>>> + * Encrypt or decrypt NBYTES bytes of data from the SRC buffer to the
>>>> DST buffer
>>>> + * using Speck-XTS, specifically the variant with a block size of
>>>> '2n' and round
>>>> + * count given by NROUNDS. The expanded round keys are given in
>>>> ROUND_KEYS, and
>>>> + * the current XTS tweak value is given in TWEAK. It's assumed that
>>>> NBYTES is a
>>>> + * nonzero multiple of 128.
>>>> + */
>>>> +.macro _speck_xts_crypt n, decrypting
>>>> + push {r4-r7}
>>>> + mov r7, sp
>>>> +
>>>> + /*
>>>> + * The first four parameters were passed in registers r0-r3. Load the
>>>> + * additional parameters, which were passed on the stack.
>>>> + */
>>>> + ldr NBYTES, [sp, #16]
>>>> + ldr TWEAK, [sp, #20]
>>>> +
>>>> + /*
>>>> + * If decrypting, modify the ROUND_KEYS parameter to point to the last
>>>> + * round key rather than the first, since for decryption the round keys
>>>> + * are used in reverse order.
>>>> + */
>>>> +.if \decrypting
>>>> +.if \n == 64
>>>> + add ROUND_KEYS, ROUND_KEYS, NROUNDS, lsl #3
>>>> + sub ROUND_KEYS, #8
>>>> +.else
>>>> + add ROUND_KEYS, ROUND_KEYS, NROUNDS, lsl #2
>>>> + sub ROUND_KEYS, #4
>>>> +.endif
>>>> +.endif
>>>> +
>>>> + // Load the index vector for vtbl-based 8-bit rotates
>>>> +.if \decrypting
>>>> + ldr r12, =.Lrol\n\()_8_table
>>>> +.else
>>>> + ldr r12, =.Lror\n\()_8_table
>>>> +.endif
>>>> + vld1.8 {ROTATE_TABLE}, [r12:64]
>>>> +
>>>> + // One-time XTS preparation
>>>> +
>>>> + /*
>>>> + * Allocate stack space to store 128 bytes worth of tweaks. For
>>>> + * performance, this space is aligned to a 16-byte boundary so that we
>>>> + * can use the load/store instructions that declare 16-byte alignment.
>>>> + */
>>>> + sub sp, #128
>>>> + bic sp, #0xf
>>>
>>>
>>> This fails here when building with CONFIG_THUMB2_KERNEL=y
>>>
>>> AS arch/arm/crypto/speck-neon-core.o
>>>
>>> arch/arm/crypto/speck-neon-core.S: Assembler messages:
>>>
>>> arch/arm/crypto/speck-neon-core.S:419: Error: r13 not allowed here --
>>> `bic sp,#0xf'
>>> arch/arm/crypto/speck-neon-core.S:423: Error: r13 not allowed here --
>>> `bic sp,#0xf'
>>> arch/arm/crypto/speck-neon-core.S:427: Error: r13 not allowed here --
>>> `bic sp,#0xf'
>>> arch/arm/crypto/speck-neon-core.S:431: Error: r13 not allowed here --
>>> `bic sp,#0xf'
>>>
>>> In a quick hack this change seems to address it:
>>>
>>>
>>> - sub sp, #128
>>> - bic sp, #0xf
>>> + mov r6, sp
>>> + sub r6, #128
>>> + bic r6, #0xf
>>> + mov sp, r6
>>>
>>> But there is probably a better solution to address this.
>>>
>>
>> Given that there is no NEON on M class cores, I recommend we put something like
>>
>> THUMB(bx pc)
>> THUMB(nop.w)
>> THUMB(.arm)
>>
>> at the beginning and be done with it.
>
> I mean nop.n or just nop, of course, and we may need a '.align 2' at
> the beginning as well.
Wouldn't it be preferable to have it assemble it in Thumb2 too? It seems
that bic sp,#0xf is the only issue...
--
Stefan
On 17 June 2018 at 12:41, Stefan Agner <[email protected]> wrote:
> On 17.06.2018 11:40, Ard Biesheuvel wrote:
>> On 17 June 2018 at 11:30, Ard Biesheuvel <[email protected]> wrote:
>>> On 17 June 2018 at 00:40, Stefan Agner <[email protected]> wrote:
>>>> Hi Eric,
>>>>
>>>> On 14.02.2018 19:42, Eric Biggers wrote:
>>>>> Add an ARM NEON-accelerated implementation of Speck-XTS. It operates on
>>>>> 128-byte chunks at a time, i.e. 8 blocks for Speck128 or 16 blocks for
>>>>> Speck64. Each 128-byte chunk goes through XTS preprocessing, then is
>>>>> encrypted/decrypted (doing one cipher round for all the blocks, then the
>>>>> next round, etc.), then goes through XTS postprocessing.
>>>>>
>>>>> The performance depends on the processor but can be about 3 times faster
>>>>> than the generic code. For example, on an ARMv7 processor we observe
>>>>> the following performance with Speck128/256-XTS:
>>>>>
>>>>> xts-speck128-neon: Encryption 107.9 MB/s, Decryption 108.1 MB/s
>>>>> xts(speck128-generic): Encryption 32.1 MB/s, Decryption 36.6 MB/s
>>>>>
>>>>> In comparison to AES-256-XTS without the Cryptography Extensions:
>>>>>
>>>>> xts-aes-neonbs: Encryption 41.2 MB/s, Decryption 36.7 MB/s
>>>>> xts(aes-asm): Encryption 31.7 MB/s, Decryption 30.8 MB/s
>>>>> xts(aes-generic): Encryption 21.2 MB/s, Decryption 20.9 MB/s
>>>>>
>>>>> Speck64/128-XTS is even faster:
>>>>>
>>>>> xts-speck64-neon: Encryption 138.6 MB/s, Decryption 139.1 MB/s
>>>>>
>>>>> Note that as with the generic code, only the Speck128 and Speck64
>>>>> variants are supported. Also, for now only the XTS mode of operation is
>>>>> supported, to target the disk and file encryption use cases. The NEON
>>>>> code also only handles the portion of the data that is evenly divisible
>>>>> into 128-byte chunks, with any remainder handled by a C fallback. Of
>>>>> course, other modes of operation could be added later if needed, and/or
>>>>> the NEON code could be updated to handle other buffer sizes.
>>>>>
>>>>> The XTS specification is only defined for AES which has a 128-bit block
>>>>> size, so for the GF(2^64) math needed for Speck64-XTS we use the
>>>>> reducing polynomial 'x^64 + x^4 + x^3 + x + 1' given by the original XEX
>>>>> paper. Of course, when possible users should use Speck128-XTS, but even
>>>>> that may be too slow on some processors; Speck64-XTS can be faster.
>>>>>
>>>>> Signed-off-by: Eric Biggers <[email protected]>
>>>>> ---
>>>>> arch/arm/crypto/Kconfig | 6 +
>>>>> arch/arm/crypto/Makefile | 2 +
>>>>> arch/arm/crypto/speck-neon-core.S | 432 ++++++++++++++++++++++++++++++
>>>>> arch/arm/crypto/speck-neon-glue.c | 288 ++++++++++++++++++++
>>>>> 4 files changed, 728 insertions(+)
>>>>> create mode 100644 arch/arm/crypto/speck-neon-core.S
>>>>> create mode 100644 arch/arm/crypto/speck-neon-glue.c
>>>>>
>>>>> diff --git a/arch/arm/crypto/Kconfig b/arch/arm/crypto/Kconfig
>>>>> index b8e69fe282b8..925d1364727a 100644
>>>>> --- a/arch/arm/crypto/Kconfig
>>>>> +++ b/arch/arm/crypto/Kconfig
>>>>> @@ -121,4 +121,10 @@ config CRYPTO_CHACHA20_NEON
>>>>> select CRYPTO_BLKCIPHER
>>>>> select CRYPTO_CHACHA20
>>>>>
>>>>> +config CRYPTO_SPECK_NEON
>>>>> + tristate "NEON accelerated Speck cipher algorithms"
>>>>> + depends on KERNEL_MODE_NEON
>>>>> + select CRYPTO_BLKCIPHER
>>>>> + select CRYPTO_SPECK
>>>>> +
>>>>> endif
>>>>> diff --git a/arch/arm/crypto/Makefile b/arch/arm/crypto/Makefile
>>>>> index 30ef8e291271..a758107c5525 100644
>>>>> --- a/arch/arm/crypto/Makefile
>>>>> +++ b/arch/arm/crypto/Makefile
>>>>> @@ -10,6 +10,7 @@ obj-$(CONFIG_CRYPTO_SHA1_ARM_NEON) += sha1-arm-neon.o
>>>>> obj-$(CONFIG_CRYPTO_SHA256_ARM) += sha256-arm.o
>>>>> obj-$(CONFIG_CRYPTO_SHA512_ARM) += sha512-arm.o
>>>>> obj-$(CONFIG_CRYPTO_CHACHA20_NEON) += chacha20-neon.o
>>>>> +obj-$(CONFIG_CRYPTO_SPECK_NEON) += speck-neon.o
>>>>>
>>>>> ce-obj-$(CONFIG_CRYPTO_AES_ARM_CE) += aes-arm-ce.o
>>>>> ce-obj-$(CONFIG_CRYPTO_SHA1_ARM_CE) += sha1-arm-ce.o
>>>>> @@ -53,6 +54,7 @@ ghash-arm-ce-y := ghash-ce-core.o ghash-ce-glue.o
>>>>> crct10dif-arm-ce-y := crct10dif-ce-core.o crct10dif-ce-glue.o
>>>>> crc32-arm-ce-y:= crc32-ce-core.o crc32-ce-glue.o
>>>>> chacha20-neon-y := chacha20-neon-core.o chacha20-neon-glue.o
>>>>> +speck-neon-y := speck-neon-core.o speck-neon-glue.o
>>>>>
>>>>> quiet_cmd_perl = PERL $@
>>>>> cmd_perl = $(PERL) $(<) > $(@)
>>>>> diff --git a/arch/arm/crypto/speck-neon-core.S
>>>>> b/arch/arm/crypto/speck-neon-core.S
>>>>> new file mode 100644
>>>>> index 000000000000..3c1e203e53b9
>>>>> --- /dev/null
>>>>> +++ b/arch/arm/crypto/speck-neon-core.S
>>>>> @@ -0,0 +1,432 @@
>>>>> +// SPDX-License-Identifier: GPL-2.0
>>>>> +/*
>>>>> + * NEON-accelerated implementation of Speck128-XTS and Speck64-XTS
>>>>> + *
>>>>> + * Copyright (c) 2018 Google, Inc
>>>>> + *
>>>>> + * Author: Eric Biggers <[email protected]>
>>>>> + */
>>>>> +
>>>>> +#include <linux/linkage.h>
>>>>> +
>>>>> + .text
>>>>> + .fpu neon
>>>>> +
>>>>> + // arguments
>>>>> + ROUND_KEYS .req r0 // const {u64,u32} *round_keys
>>>>> + NROUNDS .req r1 // int nrounds
>>>>> + DST .req r2 // void *dst
>>>>> + SRC .req r3 // const void *src
>>>>> + NBYTES .req r4 // unsigned int nbytes
>>>>> + TWEAK .req r5 // void *tweak
>>>>> +
>>>>> + // registers which hold the data being encrypted/decrypted
>>>>> + X0 .req q0
>>>>> + X0_L .req d0
>>>>> + X0_H .req d1
>>>>> + Y0 .req q1
>>>>> + Y0_H .req d3
>>>>> + X1 .req q2
>>>>> + X1_L .req d4
>>>>> + X1_H .req d5
>>>>> + Y1 .req q3
>>>>> + Y1_H .req d7
>>>>> + X2 .req q4
>>>>> + X2_L .req d8
>>>>> + X2_H .req d9
>>>>> + Y2 .req q5
>>>>> + Y2_H .req d11
>>>>> + X3 .req q6
>>>>> + X3_L .req d12
>>>>> + X3_H .req d13
>>>>> + Y3 .req q7
>>>>> + Y3_H .req d15
>>>>> +
>>>>> + // the round key, duplicated in all lanes
>>>>> + ROUND_KEY .req q8
>>>>> + ROUND_KEY_L .req d16
>>>>> + ROUND_KEY_H .req d17
>>>>> +
>>>>> + // index vector for vtbl-based 8-bit rotates
>>>>> + ROTATE_TABLE .req d18
>>>>> +
>>>>> + // multiplication table for updating XTS tweaks
>>>>> + GF128MUL_TABLE .req d19
>>>>> + GF64MUL_TABLE .req d19
>>>>> +
>>>>> + // current XTS tweak value(s)
>>>>> + TWEAKV .req q10
>>>>> + TWEAKV_L .req d20
>>>>> + TWEAKV_H .req d21
>>>>> +
>>>>> + TMP0 .req q12
>>>>> + TMP0_L .req d24
>>>>> + TMP0_H .req d25
>>>>> + TMP1 .req q13
>>>>> + TMP2 .req q14
>>>>> + TMP3 .req q15
>>>>> +
>>>>> + .align 4
>>>>> +.Lror64_8_table:
>>>>> + .byte 1, 2, 3, 4, 5, 6, 7, 0
>>>>> +.Lror32_8_table:
>>>>> + .byte 1, 2, 3, 0, 5, 6, 7, 4
>>>>> +.Lrol64_8_table:
>>>>> + .byte 7, 0, 1, 2, 3, 4, 5, 6
>>>>> +.Lrol32_8_table:
>>>>> + .byte 3, 0, 1, 2, 7, 4, 5, 6
>>>>> +.Lgf128mul_table:
>>>>> + .byte 0, 0x87
>>>>> + .fill 14
>>>>> +.Lgf64mul_table:
>>>>> + .byte 0, 0x1b, (0x1b << 1), (0x1b << 1) ^ 0x1b
>>>>> + .fill 12
>>>>> +
>>>>> +/*
>>>>> + * _speck_round_128bytes() - Speck encryption round on 128 bytes at a time
>>>>> + *
>>>>> + * Do one Speck encryption round on the 128 bytes (8 blocks for
>>>>> Speck128, 16 for
>>>>> + * Speck64) stored in X0-X3 and Y0-Y3, using the round key stored in all lanes
>>>>> + * of ROUND_KEY. 'n' is the lane size: 64 for Speck128, or 32 for Speck64.
>>>>> + *
>>>>> + * The 8-bit rotates are implemented using vtbl instead of vshr + vsli because
>>>>> + * the vtbl approach is faster on some processors and the same speed on others.
>>>>> + */
>>>>> +.macro _speck_round_128bytes n
>>>>> +
>>>>> + // x = ror(x, 8)
>>>>> + vtbl.8 X0_L, {X0_L}, ROTATE_TABLE
>>>>> + vtbl.8 X0_H, {X0_H}, ROTATE_TABLE
>>>>> + vtbl.8 X1_L, {X1_L}, ROTATE_TABLE
>>>>> + vtbl.8 X1_H, {X1_H}, ROTATE_TABLE
>>>>> + vtbl.8 X2_L, {X2_L}, ROTATE_TABLE
>>>>> + vtbl.8 X2_H, {X2_H}, ROTATE_TABLE
>>>>> + vtbl.8 X3_L, {X3_L}, ROTATE_TABLE
>>>>> + vtbl.8 X3_H, {X3_H}, ROTATE_TABLE
>>>>> +
>>>>> + // x += y
>>>>> + vadd.u\n X0, Y0
>>>>> + vadd.u\n X1, Y1
>>>>> + vadd.u\n X2, Y2
>>>>> + vadd.u\n X3, Y3
>>>>> +
>>>>> + // x ^= k
>>>>> + veor X0, ROUND_KEY
>>>>> + veor X1, ROUND_KEY
>>>>> + veor X2, ROUND_KEY
>>>>> + veor X3, ROUND_KEY
>>>>> +
>>>>> + // y = rol(y, 3)
>>>>> + vshl.u\n TMP0, Y0, #3
>>>>> + vshl.u\n TMP1, Y1, #3
>>>>> + vshl.u\n TMP2, Y2, #3
>>>>> + vshl.u\n TMP3, Y3, #3
>>>>> + vsri.u\n TMP0, Y0, #(\n - 3)
>>>>> + vsri.u\n TMP1, Y1, #(\n - 3)
>>>>> + vsri.u\n TMP2, Y2, #(\n - 3)
>>>>> + vsri.u\n TMP3, Y3, #(\n - 3)
>>>>> +
>>>>> + // y ^= x
>>>>> + veor Y0, TMP0, X0
>>>>> + veor Y1, TMP1, X1
>>>>> + veor Y2, TMP2, X2
>>>>> + veor Y3, TMP3, X3
>>>>> +.endm
>>>>> +
>>>>> +/*
>>>>> + * _speck_unround_128bytes() - Speck decryption round on 128 bytes at a time
>>>>> + *
>>>>> + * This is the inverse of _speck_round_128bytes().
>>>>> + */
>>>>> +.macro _speck_unround_128bytes n
>>>>> +
>>>>> + // y ^= x
>>>>> + veor TMP0, Y0, X0
>>>>> + veor TMP1, Y1, X1
>>>>> + veor TMP2, Y2, X2
>>>>> + veor TMP3, Y3, X3
>>>>> +
>>>>> + // y = ror(y, 3)
>>>>> + vshr.u\n Y0, TMP0, #3
>>>>> + vshr.u\n Y1, TMP1, #3
>>>>> + vshr.u\n Y2, TMP2, #3
>>>>> + vshr.u\n Y3, TMP3, #3
>>>>> + vsli.u\n Y0, TMP0, #(\n - 3)
>>>>> + vsli.u\n Y1, TMP1, #(\n - 3)
>>>>> + vsli.u\n Y2, TMP2, #(\n - 3)
>>>>> + vsli.u\n Y3, TMP3, #(\n - 3)
>>>>> +
>>>>> + // x ^= k
>>>>> + veor X0, ROUND_KEY
>>>>> + veor X1, ROUND_KEY
>>>>> + veor X2, ROUND_KEY
>>>>> + veor X3, ROUND_KEY
>>>>> +
>>>>> + // x -= y
>>>>> + vsub.u\n X0, Y0
>>>>> + vsub.u\n X1, Y1
>>>>> + vsub.u\n X2, Y2
>>>>> + vsub.u\n X3, Y3
>>>>> +
>>>>> + // x = rol(x, 8);
>>>>> + vtbl.8 X0_L, {X0_L}, ROTATE_TABLE
>>>>> + vtbl.8 X0_H, {X0_H}, ROTATE_TABLE
>>>>> + vtbl.8 X1_L, {X1_L}, ROTATE_TABLE
>>>>> + vtbl.8 X1_H, {X1_H}, ROTATE_TABLE
>>>>> + vtbl.8 X2_L, {X2_L}, ROTATE_TABLE
>>>>> + vtbl.8 X2_H, {X2_H}, ROTATE_TABLE
>>>>> + vtbl.8 X3_L, {X3_L}, ROTATE_TABLE
>>>>> + vtbl.8 X3_H, {X3_H}, ROTATE_TABLE
>>>>> +.endm
>>>>> +
>>>>> +.macro _xts128_precrypt_one dst_reg, tweak_buf, tmp
>>>>> +
>>>>> + // Load the next source block
>>>>> + vld1.8 {\dst_reg}, [SRC]!
>>>>> +
>>>>> + // Save the current tweak in the tweak buffer
>>>>> + vst1.8 {TWEAKV}, [\tweak_buf:128]!
>>>>> +
>>>>> + // XOR the next source block with the current tweak
>>>>> + veor \dst_reg, TWEAKV
>>>>> +
>>>>> + /*
>>>>> + * Calculate the next tweak by multiplying the current one by x,
>>>>> + * modulo p(x) = x^128 + x^7 + x^2 + x + 1.
>>>>> + */
>>>>> + vshr.u64 \tmp, TWEAKV, #63
>>>>> + vshl.u64 TWEAKV, #1
>>>>> + veor TWEAKV_H, \tmp\()_L
>>>>> + vtbl.8 \tmp\()_H, {GF128MUL_TABLE}, \tmp\()_H
>>>>> + veor TWEAKV_L, \tmp\()_H
>>>>> +.endm
>>>>> +
>>>>> +.macro _xts64_precrypt_two dst_reg, tweak_buf, tmp
>>>>> +
>>>>> + // Load the next two source blocks
>>>>> + vld1.8 {\dst_reg}, [SRC]!
>>>>> +
>>>>> + // Save the current two tweaks in the tweak buffer
>>>>> + vst1.8 {TWEAKV}, [\tweak_buf:128]!
>>>>> +
>>>>> + // XOR the next two source blocks with the current two tweaks
>>>>> + veor \dst_reg, TWEAKV
>>>>> +
>>>>> + /*
>>>>> + * Calculate the next two tweaks by multiplying the current ones by x^2,
>>>>> + * modulo p(x) = x^64 + x^4 + x^3 + x + 1.
>>>>> + */
>>>>> + vshr.u64 \tmp, TWEAKV, #62
>>>>> + vshl.u64 TWEAKV, #2
>>>>> + vtbl.8 \tmp\()_L, {GF64MUL_TABLE}, \tmp\()_L
>>>>> + vtbl.8 \tmp\()_H, {GF64MUL_TABLE}, \tmp\()_H
>>>>> + veor TWEAKV, \tmp
>>>>> +.endm
>>>>> +
>>>>> +/*
>>>>> + * _speck_xts_crypt() - Speck-XTS encryption/decryption
>>>>> + *
>>>>> + * Encrypt or decrypt NBYTES bytes of data from the SRC buffer to the
>>>>> DST buffer
>>>>> + * using Speck-XTS, specifically the variant with a block size of
>>>>> '2n' and round
>>>>> + * count given by NROUNDS. The expanded round keys are given in
>>>>> ROUND_KEYS, and
>>>>> + * the current XTS tweak value is given in TWEAK. It's assumed that
>>>>> NBYTES is a
>>>>> + * nonzero multiple of 128.
>>>>> + */
>>>>> +.macro _speck_xts_crypt n, decrypting
>>>>> + push {r4-r7}
>>>>> + mov r7, sp
>>>>> +
>>>>> + /*
>>>>> + * The first four parameters were passed in registers r0-r3. Load the
>>>>> + * additional parameters, which were passed on the stack.
>>>>> + */
>>>>> + ldr NBYTES, [sp, #16]
>>>>> + ldr TWEAK, [sp, #20]
>>>>> +
>>>>> + /*
>>>>> + * If decrypting, modify the ROUND_KEYS parameter to point to the last
>>>>> + * round key rather than the first, since for decryption the round keys
>>>>> + * are used in reverse order.
>>>>> + */
>>>>> +.if \decrypting
>>>>> +.if \n == 64
>>>>> + add ROUND_KEYS, ROUND_KEYS, NROUNDS, lsl #3
>>>>> + sub ROUND_KEYS, #8
>>>>> +.else
>>>>> + add ROUND_KEYS, ROUND_KEYS, NROUNDS, lsl #2
>>>>> + sub ROUND_KEYS, #4
>>>>> +.endif
>>>>> +.endif
>>>>> +
>>>>> + // Load the index vector for vtbl-based 8-bit rotates
>>>>> +.if \decrypting
>>>>> + ldr r12, =.Lrol\n\()_8_table
>>>>> +.else
>>>>> + ldr r12, =.Lror\n\()_8_table
>>>>> +.endif
>>>>> + vld1.8 {ROTATE_TABLE}, [r12:64]
>>>>> +
>>>>> + // One-time XTS preparation
>>>>> +
>>>>> + /*
>>>>> + * Allocate stack space to store 128 bytes worth of tweaks. For
>>>>> + * performance, this space is aligned to a 16-byte boundary so that we
>>>>> + * can use the load/store instructions that declare 16-byte alignment.
>>>>> + */
>>>>> + sub sp, #128
>>>>> + bic sp, #0xf
>>>>
>>>>
>>>> This fails here when building with CONFIG_THUMB2_KERNEL=y
>>>>
>>>> AS arch/arm/crypto/speck-neon-core.o
>>>>
>>>> arch/arm/crypto/speck-neon-core.S: Assembler messages:
>>>>
>>>> arch/arm/crypto/speck-neon-core.S:419: Error: r13 not allowed here --
>>>> `bic sp,#0xf'
>>>> arch/arm/crypto/speck-neon-core.S:423: Error: r13 not allowed here --
>>>> `bic sp,#0xf'
>>>> arch/arm/crypto/speck-neon-core.S:427: Error: r13 not allowed here --
>>>> `bic sp,#0xf'
>>>> arch/arm/crypto/speck-neon-core.S:431: Error: r13 not allowed here --
>>>> `bic sp,#0xf'
>>>>
>>>> In a quick hack this change seems to address it:
>>>>
>>>>
>>>> - sub sp, #128
>>>> - bic sp, #0xf
>>>> + mov r6, sp
>>>> + sub r6, #128
>>>> + bic r6, #0xf
>>>> + mov sp, r6
>>>>
>>>> But there is probably a better solution to address this.
>>>>
>>>
>>> Given that there is no NEON on M class cores, I recommend we put something like
>>>
>>> THUMB(bx pc)
>>> THUMB(nop.w)
>>> THUMB(.arm)
>>>
>>> at the beginning and be done with it.
>>
>> I mean nop.n or just nop, of course, and we may need a '.align 2' at
>> the beginning as well.
>
> Wouldn't it be preferable to have it assemble it in Thumb2 too? It seems
> that bic sp,#0xf is the only issue...
>
Well, in general, yes. In the case of NEON code, not really, since the
resulting code will not be smaller anyway, because the Thumb2 NEON
opcodes are all 4 bytes. Also, Thumb2-only cores don't have NEON
units, so all cores that this code can run on will be able to run in
ARM mode.
So from a maintainability pov, having code that only assembles in one
way is better than having code that must compile both to ARM and to
Thumb2 opcodes.
Just my 2 cents, anyway.
On Sun, Jun 17, 2018 at 01:10:41PM +0200, Ard Biesheuvel wrote:
> >>>>> +
> >>>>> + // One-time XTS preparation
> >>>>> +
> >>>>> + /*
> >>>>> + * Allocate stack space to store 128 bytes worth of tweaks. For
> >>>>> + * performance, this space is aligned to a 16-byte boundary so that we
> >>>>> + * can use the load/store instructions that declare 16-byte alignment.
> >>>>> + */
> >>>>> + sub sp, #128
> >>>>> + bic sp, #0xf
> >>>>
> >>>>
> >>>> This fails here when building with CONFIG_THUMB2_KERNEL=y
> >>>>
> >>>> AS arch/arm/crypto/speck-neon-core.o
> >>>>
> >>>> arch/arm/crypto/speck-neon-core.S: Assembler messages:
> >>>>
> >>>> arch/arm/crypto/speck-neon-core.S:419: Error: r13 not allowed here --
> >>>> `bic sp,#0xf'
> >>>> arch/arm/crypto/speck-neon-core.S:423: Error: r13 not allowed here --
> >>>> `bic sp,#0xf'
> >>>> arch/arm/crypto/speck-neon-core.S:427: Error: r13 not allowed here --
> >>>> `bic sp,#0xf'
> >>>> arch/arm/crypto/speck-neon-core.S:431: Error: r13 not allowed here --
> >>>> `bic sp,#0xf'
> >>>>
> >>>> In a quick hack this change seems to address it:
> >>>>
> >>>>
> >>>> - sub sp, #128
> >>>> - bic sp, #0xf
> >>>> + mov r6, sp
> >>>> + sub r6, #128
> >>>> + bic r6, #0xf
> >>>> + mov sp, r6
> >>>>
> >>>> But there is probably a better solution to address this.
> >>>>
> >>>
> >>> Given that there is no NEON on M class cores, I recommend we put something like
> >>>
> >>> THUMB(bx pc)
> >>> THUMB(nop.w)
> >>> THUMB(.arm)
> >>>
> >>> at the beginning and be done with it.
> >>
> >> I mean nop.n or just nop, of course, and we may need a '.align 2' at
> >> the beginning as well.
> >
> > Wouldn't it be preferable to have it assemble it in Thumb2 too? It seems
> > that bic sp,#0xf is the only issue...
> >
>
> Well, in general, yes. In the case of NEON code, not really, since the
> resulting code will not be smaller anyway, because the Thumb2 NEON
> opcodes are all 4 bytes. Also, Thumb2-only cores don't have NEON
> units, so all cores that this code can run on will be able to run in
> ARM mode.
>
> So from a maintainability pov, having code that only assembles in one
> way is better than having code that must compile both to ARM and to
> Thumb2 opcodes.
>
> Just my 2 cents, anyway.
I don't have too much of a preference, though Stefan's suggested 4 instructions
can be reduced to 3, which also matches what aes-neonbs-core.S does:
sub r12, sp, #128
bic r12, #0xf
mov sp, r12
Ard, is the following what you're suggesting instead?
diff --git a/arch/arm/crypto/speck-neon-core.S b/arch/arm/crypto/speck-neon-core.S
index 3c1e203e53b9..c989ce3dc057 100644
--- a/arch/arm/crypto/speck-neon-core.S
+++ b/arch/arm/crypto/speck-neon-core.S
@@ -8,6 +8,7 @@
*/
#include <linux/linkage.h>
+#include <asm/assembler.h>
.text
.fpu neon
@@ -233,6 +234,12 @@
* nonzero multiple of 128.
*/
.macro _speck_xts_crypt n, decrypting
+
+ .align 2
+ THUMB(bx pc)
+ THUMB(nop)
+ THUMB(.arm)
+
push {r4-r7}
mov r7, sp
@@ -413,6 +420,8 @@
mov sp, r7
pop {r4-r7}
bx lr
+
+ THUMB(.thumb)
.endm
ENTRY(speck128_xts_encrypt_neon)
On 18 June 2018 at 23:56, Eric Biggers <[email protected]> wrote:
> On Sun, Jun 17, 2018 at 01:10:41PM +0200, Ard Biesheuvel wrote:
>> >>>>> +
>> >>>>> + // One-time XTS preparation
>> >>>>> +
>> >>>>> + /*
>> >>>>> + * Allocate stack space to store 128 bytes worth of tweaks. For
>> >>>>> + * performance, this space is aligned to a 16-byte boundary so that we
>> >>>>> + * can use the load/store instructions that declare 16-byte alignment.
>> >>>>> + */
>> >>>>> + sub sp, #128
>> >>>>> + bic sp, #0xf
>> >>>>
>> >>>>
>> >>>> This fails here when building with CONFIG_THUMB2_KERNEL=y
>> >>>>
>> >>>> AS arch/arm/crypto/speck-neon-core.o
>> >>>>
>> >>>> arch/arm/crypto/speck-neon-core.S: Assembler messages:
>> >>>>
>> >>>> arch/arm/crypto/speck-neon-core.S:419: Error: r13 not allowed here --
>> >>>> `bic sp,#0xf'
>> >>>> arch/arm/crypto/speck-neon-core.S:423: Error: r13 not allowed here --
>> >>>> `bic sp,#0xf'
>> >>>> arch/arm/crypto/speck-neon-core.S:427: Error: r13 not allowed here --
>> >>>> `bic sp,#0xf'
>> >>>> arch/arm/crypto/speck-neon-core.S:431: Error: r13 not allowed here --
>> >>>> `bic sp,#0xf'
>> >>>>
>> >>>> In a quick hack this change seems to address it:
>> >>>>
>> >>>>
>> >>>> - sub sp, #128
>> >>>> - bic sp, #0xf
>> >>>> + mov r6, sp
>> >>>> + sub r6, #128
>> >>>> + bic r6, #0xf
>> >>>> + mov sp, r6
>> >>>>
>> >>>> But there is probably a better solution to address this.
>> >>>>
>> >>>
>> >>> Given that there is no NEON on M class cores, I recommend we put something like
>> >>>
>> >>> THUMB(bx pc)
>> >>> THUMB(nop.w)
>> >>> THUMB(.arm)
>> >>>
>> >>> at the beginning and be done with it.
>> >>
>> >> I mean nop.n or just nop, of course, and we may need a '.align 2' at
>> >> the beginning as well.
>> >
>> > Wouldn't it be preferable to have it assemble it in Thumb2 too? It seems
>> > that bic sp,#0xf is the only issue...
>> >
>>
>> Well, in general, yes. In the case of NEON code, not really, since the
>> resulting code will not be smaller anyway, because the Thumb2 NEON
>> opcodes are all 4 bytes. Also, Thumb2-only cores don't have NEON
>> units, so all cores that this code can run on will be able to run in
>> ARM mode.
>>
>> So from a maintainability pov, having code that only assembles in one
>> way is better than having code that must compile both to ARM and to
>> Thumb2 opcodes.
>>
>> Just my 2 cents, anyway.
>
> I don't have too much of a preference, though Stefan's suggested 4 instructions
> can be reduced to 3, which also matches what aes-neonbs-core.S does:
>
> sub r12, sp, #128
> bic r12, #0xf
> mov sp, r12
>
> Ard, is the following what you're suggesting instead?
>
Yes, but after looking at the actual code, I prefer the change above.
The access occurs only once, not in the loop so the additional
instructions should not affect performance.
Apologies for the noise.
> diff --git a/arch/arm/crypto/speck-neon-core.S b/arch/arm/crypto/speck-neon-core.S
> index 3c1e203e53b9..c989ce3dc057 100644
> --- a/arch/arm/crypto/speck-neon-core.S
> +++ b/arch/arm/crypto/speck-neon-core.S
> @@ -8,6 +8,7 @@
> */
>
> #include <linux/linkage.h>
> +#include <asm/assembler.h>
>
> .text
> .fpu neon
> @@ -233,6 +234,12 @@
> * nonzero multiple of 128.
> */
> .macro _speck_xts_crypt n, decrypting
> +
> + .align 2
> + THUMB(bx pc)
> + THUMB(nop)
> + THUMB(.arm)
> +
> push {r4-r7}
> mov r7, sp
>
> @@ -413,6 +420,8 @@
> mov sp, r7
> pop {r4-r7}
> bx lr
> +
> + THUMB(.thumb)
> .endm
>
> ENTRY(speck128_xts_encrypt_neon)