crypto: Introduce ARIA symmetric cipher algorithm

PATCH Superseded Taehee Yoo 2 participants 7 messages Jun 20, 2022 – Jun 30, 2022

Subsystems (2)

Crypto API Arm STM32

lore

Share & tools

lore

Taehee Yoo

author

2022-06-20 16:41 UTC #0 lore raw

cover crypto: Introduce ARIA symmetric cipher algorithm

This patchset adds a new ARIA(RFC 5794) symmetric cipher algorithm.

Like SEED, the ARIA is a standard cipher algorithm in South Korea. Especially Government and Banking industry have been using this algorithm. So the implementation of ARIA will be useful for them and network vendors.

Usecases of this algorithm are TLS[1], and IPSec.
It would be very useful for them if it implements kTLS for ARIA.

It is tested in x86 and MIPS with the tcrypt module.

The first patch is an implementation of ARIA algorithm. The second patch adds tests for ARIA.

[1] https://datatracker.ietf.org/doc/html/rfc6209

Taehee Yoo (2):
crypto: Implement ARIA symmetric cipher algorithm
crypto: add ARIA testmgr tests

 crypto/Kconfig        |   15 +
 crypto/Makefile       |    1 +
 crypto/aria.c         |  288 +++++
 crypto/tcrypt.c       |   38 +-
 crypto/testmgr.c      |   31 +
 crypto/testmgr.h      | 2860 +++++++++++++++++++++++++++++++++++++++++
 include/crypto/aria.h |  461 +++++++
 7 files changed, 3693 insertions(+), 1 deletion(-)
 create mode 100644 crypto/aria.c
 create mode 100644 include/crypto/aria.h

Signature · 1 line

2.17.1

Taehee Yoo

author

2022-06-20 16:41 UTC #1 lore raw

PATCH 1/2 crypto: Implement ARIA symmetric cipher algorithm

ARIA(RFC 5794) is a symmetric block cipher algorithm.
This algorithm is being used widely in South Korea as a standard cipher algorithm.
This code is written based on the ARIA implementation of OpenSSL. The OpenSSL code is based on the distributed source code[1] by KISA.

ARIA has three key sizes and corresponding rounds.
ARIA128: 12 rounds.
ARIA192: 14 rounds.
ARIA245: 16 rounds.

[1] https://seed.kisa.or.kr/kisa/Board/19/detailView.do (Korean)

Signed-off-by Taehee Yoo <[email protected]>

---
 crypto/Kconfig        |  15 ++
 crypto/Makefile       |   1 +
 crypto/aria.c         | 288 ++++++++++++++++++++++++++
 include/crypto/aria.h | 461 ++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 765 insertions(+)
 create mode 100644 crypto/aria.c
 create mode 100644 include/crypto/aria.h

crypto/Kconfig +15 -0

index 1d44893a997b..03cc8a9e8908 100644
@@ -1489,6 +1489,21 @@ config CRYPTO_SEED
 	  See also:
 	  <http://www.kisa.or.kr/kisa/seed/jsp/seed_eng.jsp>
 
+config CRYPTO_ARIA
+	tristate "ARIA cipher algorithm"
+	select CRYPTO_ALGAPI
+	help
+	  ARIA cipher algorithm (RFC5794).
+
+	  ARIA is a standard encryption algorithm of the Republic of Korea.
+	  The ARIA specifies three key sizes and rounds.
+	  128-bit: 12 rounds.
+	  192-bit: 14 rounds.
+	  256-bit: 16 rounds.
+
+	  See also:
+	  <https://seed.kisa.or.kr/kisa/algorithm/EgovAriaInfo.do>
+
 config CRYPTO_SERPENT
 	tristate "Serpent cipher algorithm"
 	select CRYPTO_ALGAPI

crypto/Makefile +1 -0

index ceaaa9f34145..dfa3e1cdd30c 100644
@@ -147,6 +147,7 @@ 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-$(CONFIG_CRYPTO_ARIA) += aria.o
 obj-$(CONFIG_CRYPTO_CHACHA20) += chacha_generic.o
 obj-$(CONFIG_CRYPTO_POLY1305) += poly1305_generic.o
 obj-$(CONFIG_CRYPTO_DEFLATE) += deflate.o

crypto/aria.c new file +288 -0

288 lines not on this page · raw message

include/crypto/aria.h new file +461 -0

461 lines not on this page · raw message

Signature · 1 line

2.17.1

Taehee Yoo

author

2022-06-20 16:41 UTC #2 lore raw

PATCH 2/2 crypto: add ARIA testmgr tests

In reply to Taehee Yoo

It contains ARIA ecb(aria), cbc(aria), cfb(aria), ctr(aria), and gcm(aria). ecb testvector is from RFC standard.
cbc, cfb, and ctr testvectors are from KISA[1], who developed ARIA algorithm.
gcm(aria) is from openssl test vector.

[1] https://seed.kisa.or.kr/kisa/kcmvp/EgovVerification.do (Korean)

Signed-off-by Taehee Yoo <[email protected]>

---
 crypto/tcrypt.c  |   38 +-
 crypto/testmgr.c |   31 +
 crypto/testmgr.h | 2860 ++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 2928 insertions(+), 1 deletion(-)

crypto/tcrypt.c +37 -1

index 2bacf8384f59..835159555e69 100644
@@ -71,7 +71,7 @@ static const char *check[] = {
 	"blowfish", "twofish", "serpent", "sha384", "sha512", "md4", "aes",
 	"cast6", "arc4", "michael_mic", "deflate", "crc32c", "tea", "xtea",
 	"khazad", "wp512", "wp384", "wp256", "xeta",  "fcrypt",
-	"camellia", "seed", "rmd160",
+	"camellia", "seed", "rmd160", "aria",
 	"lzo", "lzo-rle", "cts", "sha3-224", "sha3-256", "sha3-384",
 	"sha3-512", "streebog256", "streebog512",
 	NULL

@@ -1729,6 +1729,10 @@ static int do_test(const char *alg, u32 type, u32 mask, int m, u32 num_mb)
 		ret += tcrypt_test("ccm(sm4)");
 		break;
 
+	case 57:
+		ret += tcrypt_test("gcm(aria)");
+		break;
+
 	case 100:
 		ret += tcrypt_test("hmac(md5)");
 		break;

@@ -1865,6 +1869,12 @@ static int do_test(const char *alg, u32 type, u32 mask, int m, u32 num_mb)
 		ret += tcrypt_test("cfb(sm4)");
 		ret += tcrypt_test("ctr(sm4)");
 		break;
+	case 192:
+		ret += tcrypt_test("ecb(aria)");
+		ret += tcrypt_test("cbc(aria)");
+		ret += tcrypt_test("cfb(aria)");
+		ret += tcrypt_test("ctr(aria)");
+		break;
 	case 200:
 		test_cipher_speed("ecb(aes)", ENCRYPT, sec, NULL, 0,
 				speed_template_16_24_32);

@@ -2186,6 +2196,32 @@ static int do_test(const char *alg, u32 type, u32 mask, int m, u32 num_mb)
 				   16, 16, aead_speed_template_19, num_mb);
 		break;
 
+	case 226:
+		test_cipher_speed("ecb(aria)", ENCRYPT, sec, NULL, 0,
+				  speed_template_16_24_32);
+		test_cipher_speed("ecb(aria)", DECRYPT, sec, NULL, 0,
+				  speed_template_16_24_32);
+		test_cipher_speed("cbc(aria)", ENCRYPT, sec, NULL, 0,
+				  speed_template_16_24_32);
+		test_cipher_speed("cbc(aria)", DECRYPT, sec, NULL, 0,
+				  speed_template_16_24_32);
+		test_cipher_speed("cfb(aria)", ENCRYPT, sec, NULL, 0,
+				  speed_template_16_24_32);
+		test_cipher_speed("cfb(aria)", DECRYPT, sec, NULL, 0,
+				  speed_template_16_24_32);
+		test_cipher_speed("ctr(aria)", ENCRYPT, sec, NULL, 0,
+				  speed_template_16_24_32);
+		test_cipher_speed("ctr(aria)", DECRYPT, sec, NULL, 0,
+				  speed_template_16_24_32);
+		break;
+
+	case 227:
+		test_aead_speed("gcm(aria)", ENCRYPT, sec,
+				NULL, 0, 16, 8, speed_template_16_24_32);
+		test_aead_speed("gcm(aria)", DECRYPT, sec,
+				NULL, 0, 16, 8, speed_template_16_24_32);
+		break;
+
 	case 300:
 		if (alg) {
 			test_hash_speed(alg, sec, generic_hash_speed_template);

crypto/testmgr.c +31 -0

index 5801a8f9f713..14bb5fa6adc5 100644
@@ -4412,6 +4412,12 @@ static const struct alg_test_desc alg_test_descs[] = {
 		.suite = {
 			.cipher = __VECS(anubis_cbc_tv_template)
 		},
+	}, {
+		.alg = "cbc(aria)",
+		.test = alg_test_skcipher,
+		.suite = {
+			.cipher = __VECS(aria_cbc_tv_template)
+		},
 	}, {
 		.alg = "cbc(blowfish)",
 		.test = alg_test_skcipher,

@@ -4529,6 +4535,12 @@ static const struct alg_test_desc alg_test_descs[] = {
 		.suite = {
 			.cipher = __VECS(aes_cfb_tv_template)
 		},
+	}, {
+		.alg = "cfb(aria)",
+		.test = alg_test_skcipher,
+		.suite = {
+			.cipher = __VECS(aria_cfb_tv_template)
+		},
 	}, {
 		.alg = "cfb(sm4)",
 		.test = alg_test_skcipher,

@@ -4598,6 +4610,12 @@ static const struct alg_test_desc alg_test_descs[] = {
 		.suite = {
 			.cipher = __VECS(aes_ctr_tv_template)
 		}
+	}, {
+		.alg = "ctr(aria)",
+		.test = alg_test_skcipher,
+		.suite = {
+			.cipher = __VECS(aria_ctr_tv_template)
+		}
 	}, {
 		.alg = "ctr(blowfish)",
 		.test = alg_test_skcipher,

@@ -4858,6 +4876,12 @@ static const struct alg_test_desc alg_test_descs[] = {
 		.suite = {
 			.cipher = __VECS(arc4_tv_template)
 		}
+	}, {
+		.alg = "ecb(aria)",
+		.test = alg_test_skcipher,
+		.suite = {
+			.cipher = __VECS(aria_tv_template)
+		}
 	}, {
 		.alg = "ecb(blowfish)",
 		.test = alg_test_skcipher,

@@ -5074,6 +5098,13 @@ static const struct alg_test_desc alg_test_descs[] = {
 		.suite = {
 			.aead = __VECS(aes_gcm_tv_template)
 		}
+	}, {
+		.alg = "gcm(aria)",
+		.generic_driver = "gcm_base(ctr(aria-generic),ghash-generic)",
+		.test = alg_test_aead,
+		.suite = {
+			.aead = __VECS(aria_gcm_tv_template)
+		}
 	}, {
 		.alg = "gcm(sm4)",
 		.generic_driver = "gcm_base(ctr(sm4-generic),ghash-generic)",

crypto/testmgr.h +2860 -0

2,866 lines not on this page · raw message

Signature · 1 line

2.17.1

Herbert Xu

maintainer

2022-06-30 06:39 UTC #3 lore raw

In reply to Taehee Yoo

Quoted 8 lines · Taehee Yoo

This patchset adds a new ARIA(RFC 5794) symmetric cipher algorithm.
Like SEED, the ARIA is a standard cipher algorithm in South Korea. Especially Government and Banking industry have been using this algorithm. So the implementation of ARIA will be useful for them and network vendors.
Usecases of this algorithm are TLS[1], and IPSec.
It would be very useful for them if it implements kTLS for ARIA.

You haven't added any glue to use this for IPsec. Unless there is an in-kernel user we won't add any new algorithms.

Thanks,

Signature · 3 lines

Email: Herbert Xu <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

Taehee Yoo

author

2022-06-30 07:13 UTC #4 lore raw

On 6/30/22 15:39, Herbert Xu wrote:

Hi Herbert,
Thank you so much for your review!

Quoted 7 lines

Quoted 6 lines · Taehee Yoo
This patchset adds a new ARIA(RFC 5794) symmetric cipher algorithm.
Like SEED, the ARIA is a standard cipher algorithm in South Korea. Especially Government and Banking industry have been using this algorithm. So the implementation of ARIA will be useful for them and network vendors.

Quoted 7 lines

Quoted 2 lines
Usecases of this algorithm are TLS[1], and IPSec.
It would be very useful for them if it implements kTLS for ARIA.
You haven't added any glue to use this for IPsec. Unless there is an in-kernel user we won't add any new algorithms.
Thanks,

Unfortunately, IPsec draft[1] is expired so I think adding the aria-IPsec feature is not our rule as far as I know even if there are already users in South Korea. So, my plan is to add the kTLS feature because aria-TLS[2] is now standard.
So, Is it okay to add the aria-kTLS feature into the v2 patch, instead of aria-IPsec?

[1] https://datatracker.ietf.org/doc/draft-nsri-ipsecme-aria-ipsec/
[2] https://datatracker.ietf.org/doc/html/rfc6209

Thanks,
Taehee Yoo

Herbert Xu

maintainer

2022-06-30 07:16 UTC #5 lore raw

Quoted 4 lines · Taehee Yoo

So, my plan is to add the kTLS feature because aria-TLS[2] is now standard. So, Is it okay to add the aria-kTLS feature into the v2 patch, instead of aria-IPsec?

Sure. As long as the kTLS people are happy to add this then it's good to go.

Cheers,

Signature · 3 lines

Email: Herbert Xu <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

Taehee Yoo

author

2022-06-30 07:22 UTC #6 lore raw

Quoted 10 lines · Herbert Xu

Quoted 4 lines · Taehee Yoo
So, my plan is to add the kTLS feature because aria-TLS[2] is now standard. So, Is it okay to add the aria-kTLS feature into the v2 patch, instead of aria-IPsec?
Sure. As long as the kTLS people are happy to add this then it's good to go.
Cheers,

I will add aria-kTLS feature into the v2 patch and I will send it to both the crypto and netdev mailing list after some tests.

Thanks a lot!
Taehee Yoo