From f1e866b10676faf8b9092cb821a9ac8acf31dbd8 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Tue, 10 Mar 2015 09:47:48 +0100 Subject: [PATCH] crypto: arm - add support for GHASH using ARMv8 Crypto Extensions This implements the GHASH hash algorithm (as used by the GCM AEAD chaining mode) using the AArch32 version of the 64x64 to 128 bit polynomial multiplication instruction (vmull.p64) that is part of the ARMv8 Crypto Extensions. Signed-off-by: Ard Biesheuvel Signed-off-by: Herbert Xu --- arch/arm/crypto/Kconfig | 10 ++ arch/arm/crypto/Makefile | 2 + arch/arm/crypto/ghash-ce-core.S | 94 ++++++++++++ arch/arm/crypto/ghash-ce-glue.c | 318 ++++++++++++++++++++++++++++++++++++++++ 4 files changed, 424 insertions(+) create mode 100644 arch/arm/crypto/ghash-ce-core.S create mode 100644 arch/arm/crypto/ghash-ce-glue.c diff --git a/arch/arm/crypto/Kconfig b/arch/arm/crypto/Kconfig index 63588bdf3b5d..d63f319924d2 100644 --- a/arch/arm/crypto/Kconfig +++ b/arch/arm/crypto/Kconfig @@ -110,4 +110,14 @@ config CRYPTO_AES_ARM_CE Use an implementation of AES in CBC, CTR and XTS modes that uses ARMv8 Crypto Extensions +config CRYPTO_GHASH_ARM_CE + tristate "PMULL-accelerated GHASH using ARMv8 Crypto Extensions" + depends on KERNEL_MODE_NEON + select CRYPTO_HASH + select CRYPTO_CRYPTD + help + Use an implementation of GHASH (used by the GCM AEAD chaining mode) + that uses the 64x64 to 128 bit polynomial multiplication (vmull.p64) + that is part of the ARMv8 Crypto Extensions + endif diff --git a/arch/arm/crypto/Makefile b/arch/arm/crypto/Makefile index 2514c420e8d3..9a273bd7dffd 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_SHA512_ARM_NEON) += sha512-arm-neon.o obj-$(CONFIG_CRYPTO_SHA1_ARM_CE) += sha1-arm-ce.o obj-$(CONFIG_CRYPTO_SHA2_ARM_CE) += sha2-arm-ce.o +obj-$(CONFIG_CRYPTO_GHASH_ARM_CE) += ghash-arm-ce.o aes-arm-y := aes-armv4.o aes_glue.o aes-arm-bs-y := aesbs-core.o aesbs-glue.o @@ -19,6 +20,7 @@ sha512-arm-neon-y := sha512-armv7-neon.o sha512_neon_glue.o sha1-arm-ce-y := sha1-ce-core.o sha1-ce-glue.o sha2-arm-ce-y := sha2-ce-core.o sha2-ce-glue.o aes-arm-ce-y := aes-ce-core.o aes-ce-glue.o +ghash-arm-ce-y := ghash-ce-core.o ghash-ce-glue.o quiet_cmd_perl = PERL $@ cmd_perl = $(PERL) $(<) > $(@) diff --git a/arch/arm/crypto/ghash-ce-core.S b/arch/arm/crypto/ghash-ce-core.S new file mode 100644 index 000000000000..e643a15eadf2 --- /dev/null +++ b/arch/arm/crypto/ghash-ce-core.S @@ -0,0 +1,94 @@ +/* + * Accelerated GHASH implementation with ARMv8 vmull.p64 instructions. + * + * Copyright (C) 2015 Linaro Ltd. + * + * 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 +#include + + SHASH .req q0 + SHASH2 .req q1 + T1 .req q2 + T2 .req q3 + MASK .req q4 + XL .req q5 + XM .req q6 + XH .req q7 + IN1 .req q7 + + SHASH_L .req d0 + SHASH_H .req d1 + SHASH2_L .req d2 + T1_L .req d4 + MASK_L .req d8 + XL_L .req d10 + XL_H .req d11 + XM_L .req d12 + XM_H .req d13 + XH_L .req d14 + + .text + .fpu crypto-neon-fp-armv8 + + /* + * void pmull_ghash_update(int blocks, u64 dg[], const char *src, + * struct ghash_key const *k, const char *head) + */ +ENTRY(pmull_ghash_update) + vld1.8 {SHASH}, [r3] + vld1.64 {XL}, [r1] + vmov.i8 MASK, #0xe1 + vext.8 SHASH2, SHASH, SHASH, #8 + vshl.u64 MASK, MASK, #57 + veor SHASH2, SHASH2, SHASH + + /* do the head block first, if supplied */ + ldr ip, [sp] + teq ip, #0 + beq 0f + vld1.64 {T1}, [ip] + teq r0, #0 + b 1f + +0: vld1.64 {T1}, [r2]! + subs r0, r0, #1 + +1: /* multiply XL by SHASH in GF(2^128) */ +#ifndef CONFIG_CPU_BIG_ENDIAN + vrev64.8 T1, T1 +#endif + vext.8 T2, XL, XL, #8 + vext.8 IN1, T1, T1, #8 + veor T1, T1, T2 + veor XL, XL, IN1 + + vmull.p64 XH, SHASH_H, XL_H @ a1 * b1 + veor T1, T1, XL + vmull.p64 XL, SHASH_L, XL_L @ a0 * b0 + vmull.p64 XM, SHASH2_L, T1_L @ (a1 + a0)(b1 + b0) + + vext.8 T1, XL, XH, #8 + veor T2, XL, XH + veor XM, XM, T1 + veor XM, XM, T2 + vmull.p64 T2, XL_L, MASK_L + + vmov XH_L, XM_H + vmov XM_H, XL_L + + veor XL, XM, T2 + vext.8 T2, XL, XL, #8 + vmull.p64 XL, XL_L, MASK_L + veor T2, T2, XH + veor XL, XL, T2 + + bne 0b + + vst1.64 {XL}, [r1] + bx lr +ENDPROC(pmull_ghash_update) diff --git a/arch/arm/crypto/ghash-ce-glue.c b/arch/arm/crypto/ghash-ce-glue.c new file mode 100644 index 000000000000..8c959d128065 --- /dev/null +++ b/arch/arm/crypto/ghash-ce-glue.c @@ -0,0 +1,318 @@ +/* + * Accelerated GHASH implementation with ARMv8 vmull.p64 instructions. + * + * Copyright (C) 2015 Linaro Ltd. + * + * 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 +#include +#include +#include +#include +#include +#include +#include +#include + +MODULE_DESCRIPTION("GHASH secure hash using ARMv8 Crypto Extensions"); +MODULE_AUTHOR("Ard Biesheuvel "); +MODULE_LICENSE("GPL v2"); + +#define GHASH_BLOCK_SIZE 16 +#define GHASH_DIGEST_SIZE 16 + +struct ghash_key { + u64 a; + u64 b; +}; + +struct ghash_desc_ctx { + u64 digest[GHASH_DIGEST_SIZE/sizeof(u64)]; + u8 buf[GHASH_BLOCK_SIZE]; + u32 count; +}; + +struct ghash_async_ctx { + struct cryptd_ahash *cryptd_tfm; +}; + +asmlinkage void pmull_ghash_update(int blocks, u64 dg[], const char *src, + struct ghash_key const *k, const char *head); + +static int ghash_init(struct shash_desc *desc) +{ + struct ghash_desc_ctx *ctx = shash_desc_ctx(desc); + + *ctx = (struct ghash_desc_ctx){}; + return 0; +} + +static int ghash_update(struct shash_desc *desc, const u8 *src, + unsigned int len) +{ + struct ghash_desc_ctx *ctx = shash_desc_ctx(desc); + unsigned int partial = ctx->count % GHASH_BLOCK_SIZE; + + ctx->count += len; + + if ((partial + len) >= GHASH_BLOCK_SIZE) { + struct ghash_key *key = crypto_shash_ctx(desc->tfm); + int blocks; + + if (partial) { + int p = GHASH_BLOCK_SIZE - partial; + + memcpy(ctx->buf + partial, src, p); + src += p; + len -= p; + } + + blocks = len / GHASH_BLOCK_SIZE; + len %= GHASH_BLOCK_SIZE; + + kernel_neon_begin(); + pmull_ghash_update(blocks, ctx->digest, src, key, + partial ? ctx->buf : NULL); + kernel_neon_end(); + src += blocks * GHASH_BLOCK_SIZE; + partial = 0; + } + if (len) + memcpy(ctx->buf + partial, src, len); + return 0; +} + +static int ghash_final(struct shash_desc *desc, u8 *dst) +{ + struct ghash_desc_ctx *ctx = shash_desc_ctx(desc); + unsigned int partial = ctx->count % GHASH_BLOCK_SIZE; + + if (partial) { + struct ghash_key *key = crypto_shash_ctx(desc->tfm); + + memset(ctx->buf + partial, 0, GHASH_BLOCK_SIZE - partial); + kernel_neon_begin(); + pmull_ghash_update(1, ctx->digest, ctx->buf, key, NULL); + kernel_neon_end(); + } + put_unaligned_be64(ctx->digest[1], dst); + put_unaligned_be64(ctx->digest[0], dst + 8); + + *ctx = (struct ghash_desc_ctx){}; + return 0; +} + +static int ghash_setkey(struct crypto_shash *tfm, + const u8 *inkey, unsigned int keylen) +{ + struct ghash_key *key = crypto_shash_ctx(tfm); + u64 a, b; + + if (keylen != GHASH_BLOCK_SIZE) { + crypto_shash_set_flags(tfm, CRYPTO_TFM_RES_BAD_KEY_LEN); + return -EINVAL; + } + + /* perform multiplication by 'x' in GF(2^128) */ + b = get_unaligned_be64(inkey); + a = get_unaligned_be64(inkey + 8); + + key->a = (a << 1) | (b >> 63); + key->b = (b << 1) | (a >> 63); + + if (b >> 63) + key->b ^= 0xc200000000000000UL; + + return 0; +} + +static struct shash_alg ghash_alg = { + .digestsize = GHASH_DIGEST_SIZE, + .init = ghash_init, + .update = ghash_update, + .final = ghash_final, + .setkey = ghash_setkey, + .descsize = sizeof(struct ghash_desc_ctx), + .base = { + .cra_name = "ghash", + .cra_driver_name = "__driver-ghash-ce", + .cra_priority = 0, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = GHASH_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct ghash_key), + .cra_module = THIS_MODULE, + }, +}; + +static int ghash_async_init(struct ahash_request *req) +{ + struct crypto_ahash *tfm = crypto_ahash_reqtfm(req); + struct ghash_async_ctx *ctx = crypto_ahash_ctx(tfm); + struct ahash_request *cryptd_req = ahash_request_ctx(req); + struct cryptd_ahash *cryptd_tfm = ctx->cryptd_tfm; + + if (!may_use_simd()) { + memcpy(cryptd_req, req, sizeof(*req)); + ahash_request_set_tfm(cryptd_req, &cryptd_tfm->base); + return crypto_ahash_init(cryptd_req); + } else { + struct shash_desc *desc = cryptd_shash_desc(cryptd_req); + struct crypto_shash *child = cryptd_ahash_child(cryptd_tfm); + + desc->tfm = child; + desc->flags = req->base.flags; + return crypto_shash_init(desc); + } +} + +static int ghash_async_update(struct ahash_request *req) +{ + struct ahash_request *cryptd_req = ahash_request_ctx(req); + + if (!may_use_simd()) { + struct crypto_ahash *tfm = crypto_ahash_reqtfm(req); + struct ghash_async_ctx *ctx = crypto_ahash_ctx(tfm); + struct cryptd_ahash *cryptd_tfm = ctx->cryptd_tfm; + + memcpy(cryptd_req, req, sizeof(*req)); + ahash_request_set_tfm(cryptd_req, &cryptd_tfm->base); + return crypto_ahash_update(cryptd_req); + } else { + struct shash_desc *desc = cryptd_shash_desc(cryptd_req); + return shash_ahash_update(req, desc); + } +} + +static int ghash_async_final(struct ahash_request *req) +{ + struct ahash_request *cryptd_req = ahash_request_ctx(req); + + if (!may_use_simd()) { + struct crypto_ahash *tfm = crypto_ahash_reqtfm(req); + struct ghash_async_ctx *ctx = crypto_ahash_ctx(tfm); + struct cryptd_ahash *cryptd_tfm = ctx->cryptd_tfm; + + memcpy(cryptd_req, req, sizeof(*req)); + ahash_request_set_tfm(cryptd_req, &cryptd_tfm->base); + return crypto_ahash_final(cryptd_req); + } else { + struct shash_desc *desc = cryptd_shash_desc(cryptd_req); + return crypto_shash_final(desc, req->result); + } +} + +static int ghash_async_digest(struct ahash_request *req) +{ + struct crypto_ahash *tfm = crypto_ahash_reqtfm(req); + struct ghash_async_ctx *ctx = crypto_ahash_ctx(tfm); + struct ahash_request *cryptd_req = ahash_request_ctx(req); + struct cryptd_ahash *cryptd_tfm = ctx->cryptd_tfm; + + if (!may_use_simd()) { + memcpy(cryptd_req, req, sizeof(*req)); + ahash_request_set_tfm(cryptd_req, &cryptd_tfm->base); + return crypto_ahash_digest(cryptd_req); + } else { + struct shash_desc *desc = cryptd_shash_desc(cryptd_req); + struct crypto_shash *child = cryptd_ahash_child(cryptd_tfm); + + desc->tfm = child; + desc->flags = req->base.flags; + return shash_ahash_digest(req, desc); + } +} + +static int ghash_async_setkey(struct crypto_ahash *tfm, const u8 *key, + unsigned int keylen) +{ + struct ghash_async_ctx *ctx = crypto_ahash_ctx(tfm); + struct crypto_ahash *child = &ctx->cryptd_tfm->base; + int err; + + crypto_ahash_clear_flags(child, CRYPTO_TFM_REQ_MASK); + crypto_ahash_set_flags(child, crypto_ahash_get_flags(tfm) + & CRYPTO_TFM_REQ_MASK); + err = crypto_ahash_setkey(child, key, keylen); + crypto_ahash_set_flags(tfm, crypto_ahash_get_flags(child) + & CRYPTO_TFM_RES_MASK); + + return err; +} + +static int ghash_async_init_tfm(struct crypto_tfm *tfm) +{ + struct cryptd_ahash *cryptd_tfm; + struct ghash_async_ctx *ctx = crypto_tfm_ctx(tfm); + + cryptd_tfm = cryptd_alloc_ahash("__driver-ghash-ce", 0, 0); + if (IS_ERR(cryptd_tfm)) + return PTR_ERR(cryptd_tfm); + ctx->cryptd_tfm = cryptd_tfm; + crypto_ahash_set_reqsize(__crypto_ahash_cast(tfm), + sizeof(struct ahash_request) + + crypto_ahash_reqsize(&cryptd_tfm->base)); + + return 0; +} + +static void ghash_async_exit_tfm(struct crypto_tfm *tfm) +{ + struct ghash_async_ctx *ctx = crypto_tfm_ctx(tfm); + + cryptd_free_ahash(ctx->cryptd_tfm); +} + +static struct ahash_alg ghash_async_alg = { + .init = ghash_async_init, + .update = ghash_async_update, + .final = ghash_async_final, + .setkey = ghash_async_setkey, + .digest = ghash_async_digest, + .halg.digestsize = GHASH_DIGEST_SIZE, + .halg.base = { + .cra_name = "ghash", + .cra_driver_name = "ghash-ce", + .cra_priority = 300, + .cra_flags = CRYPTO_ALG_TYPE_AHASH | CRYPTO_ALG_ASYNC, + .cra_blocksize = GHASH_BLOCK_SIZE, + .cra_type = &crypto_ahash_type, + .cra_ctxsize = sizeof(struct ghash_async_ctx), + .cra_module = THIS_MODULE, + .cra_init = ghash_async_init_tfm, + .cra_exit = ghash_async_exit_tfm, + }, +}; + +static int __init ghash_ce_mod_init(void) +{ + int err; + + if (!(elf_hwcap2 & HWCAP2_PMULL)) + return -ENODEV; + + err = crypto_register_shash(&ghash_alg); + if (err) + return err; + err = crypto_register_ahash(&ghash_async_alg); + if (err) + goto err_shash; + + return 0; + +err_shash: + crypto_unregister_shash(&ghash_alg); + return err; +} + +static void __exit ghash_ce_mod_exit(void) +{ + crypto_unregister_ahash(&ghash_async_alg); + crypto_unregister_shash(&ghash_alg); +} + +module_init(ghash_ce_mod_init); +module_exit(ghash_ce_mod_exit); -- 2.11.0