Implement the SHA-512 using the new special instructions that have
been introduced as an optional extension in ARMv8.2.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@xxxxxxxxxx>
---
arch/arm64/crypto/Kconfig | 6 ++
arch/arm64/crypto/Makefile | 3 +
arch/arm64/crypto/sha512-ce-core.S | 207 +++++++++++++++++++++++++++++++++++++
arch/arm64/crypto/sha512-ce-glue.c | 119 +++++++++++++++++++++
4 files changed, 335 insertions(+)
create mode 100644 arch/arm64/crypto/sha512-ce-core.S
create mode 100644 arch/arm64/crypto/sha512-ce-glue.c
diff --git a/arch/arm64/crypto/Kconfig b/arch/arm64/crypto/Kconfig
index 70c517aa4501..aad288f4b9de 100644
--- a/arch/arm64/crypto/Kconfig
+++ b/arch/arm64/crypto/Kconfig
@@ -29,6 +29,12 @@ config CRYPTO_SHA2_ARM64_CE
select CRYPTO_HASH
select CRYPTO_SHA256_ARM64
+config CRYPTO_SHA512_ARM64_CE
+ tristate "SHA-384/SHA-512 digest algorithm (ARMv8 Crypto Extensions)"
+ depends on KERNEL_MODE_NEON
+ select CRYPTO_HASH
+ select CRYPTO_SHA512_ARM64
+
config CRYPTO_GHASH_ARM64_CE
tristate "GHASH/AES-GCM using ARMv8 Crypto Extensions"
depends on KERNEL_MODE_NEON
diff --git a/arch/arm64/crypto/Makefile b/arch/arm64/crypto/Makefile
index b5edc5918c28..d7573d31d397 100644
--- a/arch/arm64/crypto/Makefile
+++ b/arch/arm64/crypto/Makefile
@@ -14,6 +14,9 @@ sha1-ce-y := sha1-ce-glue.o sha1-ce-core.o
obj-$(CONFIG_CRYPTO_SHA2_ARM64_CE) += sha2-ce.o
sha2-ce-y := sha2-ce-glue.o sha2-ce-core.o
+obj-$(CONFIG_CRYPTO_SHA512_ARM64_CE) += sha512-ce.o
+sha512-ce-y := sha512-ce-glue.o sha512-ce-core.o
+
obj-$(CONFIG_CRYPTO_GHASH_ARM64_CE) += ghash-ce.o
ghash-ce-y := ghash-ce-glue.o ghash-ce-core.o
diff --git a/arch/arm64/crypto/sha512-ce-core.S b/arch/arm64/crypto/sha512-ce-core.S
new file mode 100644
index 000000000000..6c562f8df0b0
--- /dev/null
+++ b/arch/arm64/crypto/sha512-ce-core.S
@@ -0,0 +1,207 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * sha512-ce-core.S - core SHA-384/SHA-512 transform using v8 Crypto Extensions
+ *
+ * Copyright (C) 2018 Linaro Ltd <ard.biesheuvel@xxxxxxxxxx>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+ //
+ // Temporary - for testing only. binutils has no support for these yet
+ //
+ .irp b,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31
+ .set .Lq\b, \b
+ .set .Lv\b\().2d, \b
+ .endr
+
+ .macro sha512h, rd, rn, rm
+ .inst 0xce608000 | .L\rd | (.L\rn << 5) | (.L\rm << 16)
+ .endm
+
+ .macro sha512h2, rd, rn, rm
+ .inst 0xce608400 | .L\rd | (.L\rn << 5) | (.L\rm << 16)
+ .endm
+
+ .macro sha512su0, rd, rn
+ .inst 0xcec08000 | .L\rd | (.L\rn << 5)
+ .endm
+
+ .macro sha512su1, rd, rn, rm
+ .inst 0xce608800 | .L\rd | (.L\rn << 5) | (.L\rm << 16)
+ .endm
+
+ .text
+ .arch armv8-a+crypto
+
+ /*
+ * The SHA-512 round constants
+ */
+ .align 4
+.Lsha512_rcon:
+ .quad 0x428a2f98d728ae22, 0x7137449123ef65cd
+ .quad 0xb5c0fbcfec4d3b2f, 0xe9b5dba58189dbbc
+ .quad 0x3956c25bf348b538, 0x59f111f1b605d019
+ .quad 0x923f82a4af194f9b, 0xab1c5ed5da6d8118
+ .quad 0xd807aa98a3030242, 0x12835b0145706fbe
+ .quad 0x243185be4ee4b28c, 0x550c7dc3d5ffb4e2
+ .quad 0x72be5d74f27b896f, 0x80deb1fe3b1696b1
+ .quad 0x9bdc06a725c71235, 0xc19bf174cf692694
+ .quad 0xe49b69c19ef14ad2, 0xefbe4786384f25e3
+ .quad 0x0fc19dc68b8cd5b5, 0x240ca1cc77ac9c65
+ .quad 0x2de92c6f592b0275, 0x4a7484aa6ea6e483
+ .quad 0x5cb0a9dcbd41fbd4, 0x76f988da831153b5
+ .quad 0x983e5152ee66dfab, 0xa831c66d2db43210
+ .quad 0xb00327c898fb213f, 0xbf597fc7beef0ee4
+ .quad 0xc6e00bf33da88fc2, 0xd5a79147930aa725
+ .quad 0x06ca6351e003826f, 0x142929670a0e6e70
+ .quad 0x27b70a8546d22ffc, 0x2e1b21385c26c926
+ .quad 0x4d2c6dfc5ac42aed, 0x53380d139d95b3df
+ .quad 0x650a73548baf63de, 0x766a0abb3c77b2a8
+ .quad 0x81c2c92e47edaee6, 0x92722c851482353b
+ .quad 0xa2bfe8a14cf10364, 0xa81a664bbc423001
+ .quad 0xc24b8b70d0f89791, 0xc76c51a30654be30
+ .quad 0xd192e819d6ef5218, 0xd69906245565a910
+ .quad 0xf40e35855771202a, 0x106aa07032bbd1b8
+ .quad 0x19a4c116b8d2d0c8, 0x1e376c085141ab53
+ .quad 0x2748774cdf8eeb99, 0x34b0bcb5e19b48a8
+ .quad 0x391c0cb3c5c95a63, 0x4ed8aa4ae3418acb
+ .quad 0x5b9cca4f7763e373, 0x682e6ff3d6b2b8a3
+ .quad 0x748f82ee5defb2fc, 0x78a5636f43172f60
+ .quad 0x84c87814a1f0ab72, 0x8cc702081a6439ec
+ .quad 0x90befffa23631e28, 0xa4506cebde82bde9
+ .quad 0xbef9a3f7b2c67915, 0xc67178f2e372532b
+ .quad 0xca273eceea26619c, 0xd186b8c721c0c207
+ .quad 0xeada7dd6cde0eb1e, 0xf57d4f7fee6ed178
+ .quad 0x06f067aa72176fba, 0x0a637dc5a2c898a6
+ .quad 0x113f9804bef90dae, 0x1b710b35131c471b
+ .quad 0x28db77f523047d84, 0x32caab7b40c72493
+ .quad 0x3c9ebe0a15c9bebc, 0x431d67c49c100d4c
+ .quad 0x4cc5d4becb3e42b6, 0x597f299cfc657e2a
+ .quad 0x5fcb6fab3ad6faec, 0x6c44198c4a475817
+
+ .macro dround, i0, i1, i2, i3, i4, rc0, rc1, in0, in1, in2, in3, in4
+ .ifnb \rc1
+ ld1 {v\rc1\().2d}, [x3], #16
+ .endif
+ add v\rc0\().2d, v\rc0\().2d, v\in0\().2d
+ ext v6.16b, v\i2\().16b, v\i3\().16b, #8
+ ext v\rc0\().16b, v\rc0\().16b, v\rc0\().16b, #8
+ ext v7.16b, v\i1\().16b, v\i2\().16b, #8
+ add v\i3\().2d, v\i3\().2d, v\rc0\().2d
+ .ifnb \in1
+ ext v10.16b, v\in3\().16b, v\in4\().16b, #8
+ sha512su0 v\in0\().2d, v\in1\().2d
+ .endif
+ sha512h q\i3, q6, v7.2d
+ .ifnb \in1
+ sha512su1 v\in0\().2d, v\in2\().2d, v10.2d
+ .endif
+ add v\i4\().2d, v\i1\().2d, v\i3\().2d
+ sha512h2 q\i3, q\i1, v\i0\().2d
+ .endm
+
+ /*
+ * void sha512_ce_transform(struct sha512_state *sst, u8 const *src,
+ * int blocks)
+ */
+ENTRY(sha512_ce_transform)
+ /* load state */
+ ld1 {v20.2d-v23.2d}, [x0]
+
+ /* load input */
+0: ld1 {v12.2d-v15.2d}, [x1], #64
+ ld1 {v16.2d-v19.2d}, [x1], #64
+ sub w2, w2, #1
+
+ /* load round constants */
+ adr x3, .Lsha512_rcon
+
+CPU_LE( rev64 v12.16b, v12.16b )
+CPU_LE( rev64 v13.16b, v13.16b )
+CPU_LE( rev64 v14.16b, v14.16b )
+CPU_LE( rev64 v15.16b, v15.16b )
+CPU_LE( rev64 v16.16b, v16.16b )
+CPU_LE( rev64 v17.16b, v17.16b )
+CPU_LE( rev64 v18.16b, v18.16b )
+CPU_LE( rev64 v19.16b, v19.16b )
+
+ ld1 {v8.2d}, [x3], #16
+
+ mov v0.16b, v20.16b
+ mov v1.16b, v21.16b
+ mov v2.16b, v22.16b
+ mov v3.16b, v23.16b
+
+ // v0 ab cd -- ef gh ab
+ // v1 cd -- ef gh ab cd
+ // v2 ef gh ab cd -- ef
+ // v3 gh ab cd -- ef gh
+ // v4 -- ef gh ab cd --
+
+ dround 0, 1, 2, 3, 4, 8, 9, 12, 13, 19, 16, 17
+ dround 3, 0, 4, 2, 1, 9, 8, 13, 14, 12, 17, 18
+ dround 2, 3, 1, 4, 0, 8, 9, 14, 15, 13, 18, 19
+ dround 4, 2, 0, 1, 3, 9, 8, 15, 16, 14, 19, 12
+ dround 1, 4, 3, 0, 2, 8, 9, 16, 17, 15, 12, 13
+
+ dround 0, 1, 2, 3, 4, 9, 8, 17, 18, 16, 13, 14
+ dround 3, 0, 4, 2, 1, 8, 9, 18, 19, 17, 14, 15
+ dround 2, 3, 1, 4, 0, 9, 8, 19, 12, 18, 15, 16
+ dround 4, 2, 0, 1, 3, 8, 9, 12, 13, 19, 16, 17
+ dround 1, 4, 3, 0, 2, 9, 8, 13, 14, 12, 17, 18
+
+ dround 0, 1, 2, 3, 4, 8, 9, 14, 15, 13, 18, 19
+ dround 3, 0, 4, 2, 1, 9, 8, 15, 16, 14, 19, 12
+ dround 2, 3, 1, 4, 0, 8, 9, 16, 17, 15, 12, 13
+ dround 4, 2, 0, 1, 3, 9, 8, 17, 18, 16, 13, 14
+ dround 1, 4, 3, 0, 2, 8, 9, 18, 19, 17, 14, 15
+
+ dround 0, 1, 2, 3, 4, 9, 8, 19, 12, 18, 15, 16
+ dround 3, 0, 4, 2, 1, 8, 9, 12, 13, 19, 16, 17
+ dround 2, 3, 1, 4, 0, 9, 8, 13, 14, 12, 17, 18
+ dround 4, 2, 0, 1, 3, 8, 9, 14, 15, 13, 18, 19
+ dround 1, 4, 3, 0, 2, 9, 8, 15, 16, 14, 19, 12
+
+ dround 0, 1, 2, 3, 4, 8, 9, 16, 17, 15, 12, 13
+ dround 3, 0, 4, 2, 1, 9, 8, 17, 18, 16, 13, 14
+ dround 2, 3, 1, 4, 0, 8, 9, 18, 19, 17, 14, 15
+ dround 4, 2, 0, 1, 3, 9, 8, 19, 12, 18, 15, 16
+ dround 1, 4, 3, 0, 2, 8, 9, 12, 13, 19, 16, 17
+
+ dround 0, 1, 2, 3, 4, 9, 8, 13, 14, 12, 17, 18
+ dround 3, 0, 4, 2, 1, 8, 9, 14, 15, 13, 18, 19
+ dround 2, 3, 1, 4, 0, 9, 8, 15, 16, 14, 19, 12
+ dround 4, 2, 0, 1, 3, 8, 9, 16, 17, 15, 12, 13
+ dround 1, 4, 3, 0, 2, 9, 8, 17, 18, 16, 13, 14
+
+ dround 0, 1, 2, 3, 4, 8, 9, 18, 19, 17, 14, 15
+ dround 3, 0, 4, 2, 1, 9, 8, 19, 12, 18, 15, 16
+ dround 2, 3, 1, 4, 0, 8, 9, 12
+ dround 4, 2, 0, 1, 3, 9, 8, 13
+ dround 1, 4, 3, 0, 2, 8, 9, 14
+
+ dround 0, 1, 2, 3, 4, 9, 8, 15
+ dround 3, 0, 4, 2, 1, 8, 9, 16
+ dround 2, 3, 1, 4, 0, 9, 8, 17
+ dround 4, 2, 0, 1, 3, 8, 9, 18
+ dround 1, 4, 3, 0, 2, 9, , 19
+
+ /* update state */
+ add v20.2d, v20.2d, v0.2d
+ add v21.2d, v21.2d, v1.2d
+ add v22.2d, v22.2d, v2.2d
+ add v23.2d, v23.2d, v3.2d
+
+ /* handled all input blocks? */
+ cbnz w2, 0b
+
+ /* store new state */
+3: st1 {v20.2d-v23.2d}, [x0]
+ ret
+ENDPROC(sha512_ce_transform)
diff --git a/arch/arm64/crypto/sha512-ce-glue.c b/arch/arm64/crypto/sha512-ce-glue.c
new file mode 100644
index 000000000000..a77c8632a589
--- /dev/null
+++ b/arch/arm64/crypto/sha512-ce-glue.c
@@ -0,0 +1,119 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * sha512-ce-glue.c - SHA-384/SHA-512 using ARMv8 Crypto Extensions
+ *
+ * Copyright (C) 2018 Linaro Ltd <ard.biesheuvel@xxxxxxxxxx>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <asm/neon.h>
+#include <asm/simd.h>
+#include <asm/unaligned.h>
+#include <crypto/internal/hash.h>
+#include <crypto/sha.h>
+#include <crypto/sha512_base.h>
+#include <linux/cpufeature.h>
+#include <linux/crypto.h>
+#include <linux/module.h>
+
+MODULE_DESCRIPTION("SHA-384/SHA-512 secure hash using ARMv8 Crypto Extensions");
+MODULE_AUTHOR("Ard Biesheuvel <ard.biesheuvel@xxxxxxxxxx>");
+MODULE_LICENSE("GPL v2");
+
+asmlinkage void sha512_ce_transform(struct sha512_state *sst, u8 const *src,
+ int blocks);
+
+asmlinkage void sha512_block_data_order(u64 *digest, u8 const *src, int blocks);
+
+static int sha512_ce_update(struct shash_desc *desc, const u8 *data,
+ unsigned int len)
+{
+ if (!may_use_simd())
+ return sha512_base_do_update(desc, data, len,
+ (sha512_block_fn *)sha512_block_data_order);
+
+ kernel_neon_begin();
+ sha512_base_do_update(desc, data, len,
+ (sha512_block_fn *)sha512_ce_transform);
+ kernel_neon_end();
+
+ return 0;
+}
+
+static int sha512_ce_finup(struct shash_desc *desc, const u8 *data,
+ unsigned int len, u8 *out)
+{
+ if (!may_use_simd()) {
+ if (len)
+ sha512_base_do_update(desc, data, len,
+ (sha512_block_fn *)sha512_block_data_order);
+ sha512_base_do_finalize(desc,
+ (sha512_block_fn *)sha512_block_data_order);
+ return sha512_base_finish(desc, out);
+ }
+
+ kernel_neon_begin();
+ sha512_base_do_update(desc, data, len,
+ (sha512_block_fn *)sha512_ce_transform);
+ sha512_base_do_finalize(desc, (sha512_block_fn *)sha512_ce_transform);
+ kernel_neon_end();
+ return sha512_base_finish(desc, out);
+}
+
+static int sha512_ce_final(struct shash_desc *desc, u8 *out)
+{
+ if (!may_use_simd()) {
+ sha512_base_do_finalize(desc,
+ (sha512_block_fn *)sha512_block_data_order);
+ return sha512_base_finish(desc, out);
+ }
+
+ kernel_neon_begin();
+ sha512_base_do_finalize(desc, (sha512_block_fn *)sha512_ce_transform);
+ kernel_neon_end();
+ return sha512_base_finish(desc, out);
+}
+
+static struct shash_alg algs[] = { {
+ .init = sha384_base_init,
+ .update = sha512_ce_update,
+ .final = sha512_ce_final,
+ .finup = sha512_ce_finup,
+ .descsize = sizeof(struct sha512_state),
+ .digestsize = SHA384_DIGEST_SIZE,
+ .base.cra_name = "sha384",
+ .base.cra_driver_name = "sha384-ce",
+ .base.cra_priority = 200,
+ .base.cra_flags = CRYPTO_ALG_TYPE_SHASH,
+ .base.cra_blocksize = SHA512_BLOCK_SIZE,
+ .base.cra_module = THIS_MODULE,
+}, {
+ .init = sha512_base_init,
+ .update = sha512_ce_update,
+ .final = sha512_ce_final,
+ .finup = sha512_ce_finup,
+ .descsize = sizeof(struct sha512_state),
+ .digestsize = SHA512_DIGEST_SIZE,
+ .base.cra_name = "sha512",
+ .base.cra_driver_name = "sha512-ce",
+ .base.cra_priority = 200,
+ .base.cra_flags = CRYPTO_ALG_TYPE_SHASH,
+ .base.cra_blocksize = SHA512_BLOCK_SIZE,
+ .base.cra_module = THIS_MODULE,
+} };
+
+static int __init sha512_ce_mod_init(void)
+{
+ return crypto_register_shashes(algs, ARRAY_SIZE(algs));
+}
+
+static void __exit sha512_ce_mod_fini(void)
+{
+ crypto_unregister_shashes(algs, ARRAY_SIZE(algs));
+}
+
+module_cpu_feature_match(SHA512, sha512_ce_mod_init);
+module_exit(sha512_ce_mod_fini);
--
2.11.0
