riscv: Provide vector crypto implementation of AES-CTR mode.

Support zvbb-zvkned based rvv AES-128/192/256-CTR encryption.

Signed-off-by: Phoebe Chen <phoebe.chen@sifive.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/21923)

diff --git a/crypto/aes/asm/aes-riscv64-zvkb-zvkned.pl b/crypto/aes/asm/aes-riscv64-zvkb-zvkned.pl
new file mode 100644 (file)
index 0000000..8b87b21
--- /dev/null
+++ b/crypto/aes/asm/aes-riscv64-zvkb-zvkned.pl
@@ -0,0 +1,376 @@
+#! /usr/bin/env perl
+# This file is dual-licensed, meaning that you can use it under your
+# choice of either of the following two licenses:
+#
+# Copyright 2023 The OpenSSL Project Authors. All Rights Reserved.
+#
+# Licensed under the Apache License 2.0 (the "License"). You can obtain
+# a copy in the file LICENSE in the source distribution or at
+# https://www.openssl.org/source/license.html
+#
+# or
+#
+# Copyright (c) 2023, Jerry Shih <jerry.shih@sifive.com>
+# All rights reserved.
+#
+# Redistribution and use in source and binary forms, with or without
+# modification, are permitted provided that the following conditions
+# are met:
+# 1. Redistributions of source code must retain the above copyright
+#    notice, this list of conditions and the following disclaimer.
+# 2. Redistributions in binary form must reproduce the above copyright
+#    notice, this list of conditions and the following disclaimer in the
+#    documentation and/or other materials provided with the distribution.
+#
+# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+# - RV64I
+# - RISC-V Vector ('V') with VLEN >= 128
+# - RISC-V Vector Cryptography Bit-manipulation extension ('Zvkb')
+# - RISC-V Vector AES block cipher extension ('Zvkned')
+# - RISC-V Zicclsm(Main memory supports misaligned loads/stores)
+
+use strict;
+use warnings;
+
+use FindBin qw($Bin);
+use lib "$Bin";
+use lib "$Bin/../../perlasm";
+use riscv;
+
+# $output is the last argument if it looks like a file (it has an extension)
+# $flavour is the first argument if it doesn't look like a file
+my $output = $#ARGV >= 0 && $ARGV[$#ARGV] =~ m|\.\w+$| ? pop : undef;
+my $flavour = $#ARGV >= 0 && $ARGV[0] !~ m|\.| ? shift : undef;
+
+$output and open STDOUT,">$output";
+
+my $code=<<___;
+.text
+___
+
+################################################################################
+# void rv64i_zvkb_zvkned_ctr32_encrypt_blocks(const unsigned char *in,
+#                                             unsigned char *out, size_t blocks,
+#                                             const void *key,
+#                                             const unsigned char ivec[16]);
+{
+my ($INP, $OUTP, $BLOCK_NUM, $KEYP, $IVP) = ("a0", "a1", "a2", "a3", "a4");
+my ($T0, $T1, $T2, $T3) = ("t0", "t1", "t2", "t3");
+my ($VL) = ("t4");
+my ($LEN32) = ("t5");
+my ($CTR) = ("t6");
+my ($MASK) = ("v0");
+my ($V0, $V1, $V2, $V3, $V4, $V5, $V6, $V7,
+    $V8, $V9, $V10, $V11, $V12, $V13, $V14, $V15,
+    $V16, $V17, $V18, $V19, $V20, $V21, $V22, $V23,
+    $V24, $V25, $V26, $V27, $V28, $V29, $V30, $V31,
+) = map("v$_",(0..31));
+
+# Prepare the AES ctr input data into v16.
+sub init_aes_ctr_input {
+    my $code=<<___;
+    # Setup mask into v0
+    # The mask pattern for 4*N-th elements
+    # mask v0: [000100010001....]
+    # Note:
+    #   We could setup the mask just for the maximum element length instead of
+    #   the VLMAX.
+    li $T0, 0b10001000
+    @{[vsetvli $T2, "zero", "e8", "m1", "ta", "ma"]}
+    @{[vmv_v_x $MASK, $T0]}
+    # Load IV.
+    # v31:[IV0, IV1, IV2, big-endian count]
+    @{[vsetivli "zero", 4, "e32", "m1", "ta", "ma"]}
+    @{[vle32_v $V31, $IVP]}
+    # Convert the big-endian counter into little-endian.
+    @{[vsetivli "zero", 4, "e32", "m1", "ta", "mu"]}
+    @{[vrev8_v $V31, $V31, $MASK]}
+    # Splat the IV to v16
+    @{[vsetvli "zero", $LEN32, "e32", "m4", "ta", "ma"]}
+    @{[vmv_v_i $V16, 0]}
+    @{[vaesz_vs $V16, $V31]}
+    # Prepare the ctr pattern into v20
+    # v20: [x, x, x, 0, x, x, x, 1, x, x, x, 2, ...]
+    @{[viota_m $V20, $MASK, $MASK]}
+    # v16:[IV0, IV1, IV2, count+0, IV0, IV1, IV2, count+1, ...]
+    @{[vsetvli $VL, $LEN32, "e32", "m4", "ta", "mu"]}
+    @{[vadd_vv $V16, $V16, $V20, $MASK]}
+___
+
+    return $code;
+}
+
+$code .= <<___;
+.p2align 3
+.globl rv64i_zvkb_zvkned_ctr32_encrypt_blocks
+.type rv64i_zvkb_zvkned_ctr32_encrypt_blocks,\@function
+rv64i_zvkb_zvkned_ctr32_encrypt_blocks:
+    beqz $BLOCK_NUM, 1f
+
+    # Load number of rounds
+    lwu $T0, 240($KEYP)
+    li $T1, 14
+    li $T2, 12
+    li $T3, 10
+
+    slli $LEN32, $BLOCK_NUM, 2
+
+    beq $T0, $T1, ctr32_encrypt_blocks_256
+    beq $T0, $T2, ctr32_encrypt_blocks_192
+    beq $T0, $T3, ctr32_encrypt_blocks_128
+
+1:
+    ret
+
+.size rv64i_zvkb_zvkned_ctr32_encrypt_blocks,.-rv64i_zvkb_zvkned_ctr32_encrypt_blocks
+___
+
+$code .= <<___;
+.p2align 3
+ctr32_encrypt_blocks_128:
+    # Load all 11 round keys to v1-v11 registers.
+    @{[vsetivli "zero", 4, "e32", "m1", "ta", "ma"]}
+    @{[vle32_v $V1, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V2, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V3, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V4, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V5, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V6, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V7, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V8, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V9, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V10, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V11, $KEYP]}
+
+    @{[init_aes_ctr_input]}
+
+    ##### AES body
+    j 2f
+1:
+    @{[vsetvli $VL, $LEN32, "e32", "m4", "ta", "mu"]}
+    # Increase ctr in v16.
+    @{[vadd_vx $V16, $V16, $CTR, $MASK]}
+2:
+    # Load plaintext into v20
+    @{[vle32_v $V20, $INP]}
+    slli $T0, $VL, 2
+    srli $CTR, $VL, 2
+    sub $LEN32, $LEN32, $VL
+    add $INP, $INP, $T0
+    # Prepare the AES ctr input into v24.
+    # The ctr data uses big-endian form.
+    @{[vmv_v_v $V24, $V16]}
+    @{[vrev8_v $V24, $V24, $MASK]}
+
+    @{[vaesz_vs $V24, $V1]}
+    @{[vaesem_vs $V24, $V2]}
+    @{[vaesem_vs $V24, $V3]}
+    @{[vaesem_vs $V24, $V4]}
+    @{[vaesem_vs $V24, $V5]}
+    @{[vaesem_vs $V24, $V6]}
+    @{[vaesem_vs $V24, $V7]}
+    @{[vaesem_vs $V24, $V8]}
+    @{[vaesem_vs $V24, $V9]}
+    @{[vaesem_vs $V24, $V10]}
+    @{[vaesef_vs $V24, $V11]}
+
+    # ciphertext
+    @{[vxor_vv $V24, $V24, $V20]}
+
+    # Store the ciphertext.
+    @{[vse32_v $V24, $OUTP]}
+    add $OUTP, $OUTP, $T0
+
+    bnez $LEN32, 1b
+
+    ret
+.size ctr32_encrypt_blocks_128,.-ctr32_encrypt_blocks_128
+___
+
+$code .= <<___;
+.p2align 3
+ctr32_encrypt_blocks_192:
+    # Load all 13 round keys to v1-v13 registers.
+    @{[vsetivli "zero", 4, "e32", "m1", "ta", "ma"]}
+    @{[vle32_v $V1, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V2, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V3, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V4, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V5, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V6, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V7, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V8, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V9, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V10, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V11, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V12, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V13, $KEYP]}
+
+    @{[init_aes_ctr_input]}
+
+    ##### AES body
+    j 2f
+1:
+    @{[vsetvli $VL, $LEN32, "e32", "m4", "ta", "mu"]}
+    # Increase ctr in v16.
+    @{[vadd_vx $V16, $V16, $CTR, $MASK]}
+2:
+    # Load plaintext into v20
+    @{[vle32_v $V20, $INP]}
+    slli $T0, $VL, 2
+    srli $CTR, $VL, 2
+    sub $LEN32, $LEN32, $VL
+    add $INP, $INP, $T0
+    # Prepare the AES ctr input into v24.
+    # The ctr data uses big-endian form.
+    @{[vmv_v_v $V24, $V16]}
+    @{[vrev8_v $V24, $V24, $MASK]}
+
+    @{[vaesz_vs $V24, $V1]}
+    @{[vaesem_vs $V24, $V2]}
+    @{[vaesem_vs $V24, $V3]}
+    @{[vaesem_vs $V24, $V4]}
+    @{[vaesem_vs $V24, $V5]}
+    @{[vaesem_vs $V24, $V6]}
+    @{[vaesem_vs $V24, $V7]}
+    @{[vaesem_vs $V24, $V8]}
+    @{[vaesem_vs $V24, $V9]}
+    @{[vaesem_vs $V24, $V10]}
+    @{[vaesem_vs $V24, $V11]}
+    @{[vaesem_vs $V24, $V12]}
+    @{[vaesef_vs $V24, $V13]}
+
+    # ciphertext
+    @{[vxor_vv $V24, $V24, $V20]}
+
+    # Store the ciphertext.
+    @{[vse32_v $V24, $OUTP]}
+    add $OUTP, $OUTP, $T0
+
+    bnez $LEN32, 1b
+
+    ret
+.size ctr32_encrypt_blocks_192,.-ctr32_encrypt_blocks_192
+___
+
+$code .= <<___;
+.p2align 3
+ctr32_encrypt_blocks_256:
+    # Load all 15 round keys to v1-v15 registers.
+    @{[vsetivli "zero", 4, "e32", "m1", "ta", "ma"]}
+    @{[vle32_v $V1, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V2, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V3, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V4, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V5, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V6, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V7, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V8, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V9, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V10, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V11, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V12, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V13, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V14, $KEYP]}
+    addi $KEYP, $KEYP, 16
+    @{[vle32_v $V15, $KEYP]}
+
+    @{[init_aes_ctr_input]}
+
+    ##### AES body
+    j 2f
+1:
+    @{[vsetvli $VL, $LEN32, "e32", "m4", "ta", "mu"]}
+    # Increase ctr in v16.
+    @{[vadd_vx $V16, $V16, $CTR, $MASK]}
+2:
+    # Load plaintext into v20
+    @{[vle32_v $V20, $INP]}
+    slli $T0, $VL, 2
+    srli $CTR, $VL, 2
+    sub $LEN32, $LEN32, $VL
+    add $INP, $INP, $T0
+    # Prepare the AES ctr input into v24.
+    # The ctr data uses big-endian form.
+    @{[vmv_v_v $V24, $V16]}
+    @{[vrev8_v $V24, $V24, $MASK]}
+
+    @{[vaesz_vs $V24, $V1]}
+    @{[vaesem_vs $V24, $V2]}
+    @{[vaesem_vs $V24, $V3]}
+    @{[vaesem_vs $V24, $V4]}
+    @{[vaesem_vs $V24, $V5]}
+    @{[vaesem_vs $V24, $V6]}
+    @{[vaesem_vs $V24, $V7]}
+    @{[vaesem_vs $V24, $V8]}
+    @{[vaesem_vs $V24, $V9]}
+    @{[vaesem_vs $V24, $V10]}
+    @{[vaesem_vs $V24, $V11]}
+    @{[vaesem_vs $V24, $V12]}
+    @{[vaesem_vs $V24, $V13]}
+    @{[vaesem_vs $V24, $V14]}
+    @{[vaesef_vs $V24, $V15]}
+
+    # ciphertext
+    @{[vxor_vv $V24, $V24, $V20]}
+
+    # Store the ciphertext.
+    @{[vse32_v $V24, $OUTP]}
+    add $OUTP, $OUTP, $T0
+
+    bnez $LEN32, 1b
+
+    ret
+.size ctr32_encrypt_blocks_256,.-ctr32_encrypt_blocks_256
+___
+}
+
+print $code;
+
+close STDOUT or die "error closing STDOUT: $!";

diff --git a/crypto/aes/build.info b/crypto/aes/build.info
index 7865231212f846a6c60377dac1c899bdaa83891d..c7fab5720d38c521339b21a6410e0656bbb64746 100644 (file)
--- a/crypto/aes/build.info
+++ b/crypto/aes/build.info
@@ -47,7 +47,7 @@ IF[{- !$disabled{asm} -}]
   # aes-c64xplus.s implements AES_ctr32_encrypt
   $AESDEF_c64xplus=AES_ASM AES_CTR_ASM
 
-  $AESASM_riscv64=aes_cbc.c aes-riscv64.s aes-riscv64-zkn.s aes-riscv64-zvkned.s
+  $AESASM_riscv64=aes_cbc.c aes-riscv64.s aes-riscv64-zkn.s aes-riscv64-zvkb-zvkned.s aes-riscv64-zvkned.s aes-riscv64-zvbb-zvkg-zvkned.s
   $AESDEF_riscv64=AES_ASM
   $AESASM_riscv32=aes_core.c aes_cbc.c aes-riscv32-zkn.s
 
@@ -124,6 +124,7 @@ INCLUDE[aes-mips.o]=..
 GENERATE[aes-riscv64.s]=asm/aes-riscv64.pl
 GENERATE[aes-riscv64-zkn.s]=asm/aes-riscv64-zkn.pl
 GENERATE[aes-riscv32-zkn.s]=asm/aes-riscv32-zkn.pl
+GENERATE[aes-riscv64-zvkb-zvkned.s]=asm/aes-riscv64-zvkb-zvkned.pl
 GENERATE[aes-riscv64-zvkned.s]=asm/aes-riscv64-zvkned.pl
 
 GENERATE[aesv8-armx.S]=asm/aesv8-armx.pl

diff --git a/include/crypto/aes_platform.h b/include/crypto/aes_platform.h
index d76ea937bbcb101290ad264d026a58d53ca6f318..13d94436289fa5a1a6540d2adc6648548e891716 100644 (file)
--- a/include/crypto/aes_platform.h
+++ b/include/crypto/aes_platform.h
@@ -461,6 +461,20 @@ void rv64i_zvkned_cbc_encrypt(const unsigned char *in, unsigned char *out,
 void rv64i_zvkned_cbc_decrypt(const unsigned char *in, unsigned char *out,
                               size_t length, const AES_KEY *key,
                               unsigned char *ivec, const int enc);
+
+void rv64i_zvkned_ecb_encrypt(const unsigned char *in, unsigned char *out,
+                              size_t length, const AES_KEY *key,
+                              const int enc);
+
+void rv64i_zvkned_ecb_decrypt(const unsigned char *in, unsigned char *out,
+                              size_t length, const AES_KEY *key,
+                              const int enc);
+
+void rv64i_zvkb_zvkned_ctr32_encrypt_blocks(const unsigned char *in,
+                                            unsigned char *out, size_t blocks,
+                                            const void *key,
+                                            const unsigned char ivec[16]);
+
 # elif defined(OPENSSL_CPUID_OBJ) && defined(__riscv) && __riscv_xlen == 32
 /* RISC-V 32 support */
 #  include "riscv_arch.h"

diff --git a/providers/implementations/ciphers/cipher_aes_hw_rv64i.inc b/providers/implementations/ciphers/cipher_aes_hw_rv64i.inc
index c7e1c035c71b2f9bcfb9e0312ee249bb20c8f40a..7a3337a357fc2214dbcc367e48bcf5963e06d261 100644 (file)
--- a/providers/implementations/ciphers/cipher_aes_hw_rv64i.inc
+++ b/providers/implementations/ciphers/cipher_aes_hw_rv64i.inc
@@ -89,6 +89,10 @@ static int cipher_hw_rv64i_zvkned_initkey(PROV_CIPHER_CTX *dat,
         } else {
             dat->stream.cbc = (cbc128_f) rv64i_zvkned_cbc_decrypt;
         }
+    } else if (dat->mode == EVP_CIPH_CTR_MODE) {
+        if (RISCV_HAS_ZVKB()) {
+            dat->stream.ctr = (ctr128_f) rv64i_zvkb_zvkned_ctr32_encrypt_blocks;
+        }
     }
 
     /* Zvkned supports aes-128/192/256 encryption and decryption. */