15fb8ef25SArd Biesheuvel // SPDX-License-Identifier: GPL-2.0-or-later
25fb8ef25SArd Biesheuvel /*
35fb8ef25SArd Biesheuvel * The "hash function" used as the core of the ChaCha stream cipher (RFC7539)
45fb8ef25SArd Biesheuvel *
55fb8ef25SArd Biesheuvel * Copyright (C) 2015 Martin Willi
65fb8ef25SArd Biesheuvel */
75fb8ef25SArd Biesheuvel
85fb8ef25SArd Biesheuvel #include <linux/bug.h>
95fb8ef25SArd Biesheuvel #include <linux/kernel.h>
105fb8ef25SArd Biesheuvel #include <linux/export.h>
115fb8ef25SArd Biesheuvel #include <linux/bitops.h>
125fb8ef25SArd Biesheuvel #include <linux/string.h>
135fb8ef25SArd Biesheuvel #include <asm/unaligned.h>
145fb8ef25SArd Biesheuvel #include <crypto/chacha.h>
155fb8ef25SArd Biesheuvel
chacha_permute(u32 * x,int nrounds)165fb8ef25SArd Biesheuvel static void chacha_permute(u32 *x, int nrounds)
175fb8ef25SArd Biesheuvel {
185fb8ef25SArd Biesheuvel int i;
195fb8ef25SArd Biesheuvel
205fb8ef25SArd Biesheuvel /* whitelist the allowed round counts */
215fb8ef25SArd Biesheuvel WARN_ON_ONCE(nrounds != 20 && nrounds != 12);
225fb8ef25SArd Biesheuvel
235fb8ef25SArd Biesheuvel for (i = 0; i < nrounds; i += 2) {
245fb8ef25SArd Biesheuvel x[0] += x[4]; x[12] = rol32(x[12] ^ x[0], 16);
255fb8ef25SArd Biesheuvel x[1] += x[5]; x[13] = rol32(x[13] ^ x[1], 16);
265fb8ef25SArd Biesheuvel x[2] += x[6]; x[14] = rol32(x[14] ^ x[2], 16);
275fb8ef25SArd Biesheuvel x[3] += x[7]; x[15] = rol32(x[15] ^ x[3], 16);
285fb8ef25SArd Biesheuvel
295fb8ef25SArd Biesheuvel x[8] += x[12]; x[4] = rol32(x[4] ^ x[8], 12);
305fb8ef25SArd Biesheuvel x[9] += x[13]; x[5] = rol32(x[5] ^ x[9], 12);
315fb8ef25SArd Biesheuvel x[10] += x[14]; x[6] = rol32(x[6] ^ x[10], 12);
325fb8ef25SArd Biesheuvel x[11] += x[15]; x[7] = rol32(x[7] ^ x[11], 12);
335fb8ef25SArd Biesheuvel
345fb8ef25SArd Biesheuvel x[0] += x[4]; x[12] = rol32(x[12] ^ x[0], 8);
355fb8ef25SArd Biesheuvel x[1] += x[5]; x[13] = rol32(x[13] ^ x[1], 8);
365fb8ef25SArd Biesheuvel x[2] += x[6]; x[14] = rol32(x[14] ^ x[2], 8);
375fb8ef25SArd Biesheuvel x[3] += x[7]; x[15] = rol32(x[15] ^ x[3], 8);
385fb8ef25SArd Biesheuvel
395fb8ef25SArd Biesheuvel x[8] += x[12]; x[4] = rol32(x[4] ^ x[8], 7);
405fb8ef25SArd Biesheuvel x[9] += x[13]; x[5] = rol32(x[5] ^ x[9], 7);
415fb8ef25SArd Biesheuvel x[10] += x[14]; x[6] = rol32(x[6] ^ x[10], 7);
425fb8ef25SArd Biesheuvel x[11] += x[15]; x[7] = rol32(x[7] ^ x[11], 7);
435fb8ef25SArd Biesheuvel
445fb8ef25SArd Biesheuvel x[0] += x[5]; x[15] = rol32(x[15] ^ x[0], 16);
455fb8ef25SArd Biesheuvel x[1] += x[6]; x[12] = rol32(x[12] ^ x[1], 16);
465fb8ef25SArd Biesheuvel x[2] += x[7]; x[13] = rol32(x[13] ^ x[2], 16);
475fb8ef25SArd Biesheuvel x[3] += x[4]; x[14] = rol32(x[14] ^ x[3], 16);
485fb8ef25SArd Biesheuvel
495fb8ef25SArd Biesheuvel x[10] += x[15]; x[5] = rol32(x[5] ^ x[10], 12);
505fb8ef25SArd Biesheuvel x[11] += x[12]; x[6] = rol32(x[6] ^ x[11], 12);
515fb8ef25SArd Biesheuvel x[8] += x[13]; x[7] = rol32(x[7] ^ x[8], 12);
525fb8ef25SArd Biesheuvel x[9] += x[14]; x[4] = rol32(x[4] ^ x[9], 12);
535fb8ef25SArd Biesheuvel
545fb8ef25SArd Biesheuvel x[0] += x[5]; x[15] = rol32(x[15] ^ x[0], 8);
555fb8ef25SArd Biesheuvel x[1] += x[6]; x[12] = rol32(x[12] ^ x[1], 8);
565fb8ef25SArd Biesheuvel x[2] += x[7]; x[13] = rol32(x[13] ^ x[2], 8);
575fb8ef25SArd Biesheuvel x[3] += x[4]; x[14] = rol32(x[14] ^ x[3], 8);
585fb8ef25SArd Biesheuvel
595fb8ef25SArd Biesheuvel x[10] += x[15]; x[5] = rol32(x[5] ^ x[10], 7);
605fb8ef25SArd Biesheuvel x[11] += x[12]; x[6] = rol32(x[6] ^ x[11], 7);
615fb8ef25SArd Biesheuvel x[8] += x[13]; x[7] = rol32(x[7] ^ x[8], 7);
625fb8ef25SArd Biesheuvel x[9] += x[14]; x[4] = rol32(x[4] ^ x[9], 7);
635fb8ef25SArd Biesheuvel }
645fb8ef25SArd Biesheuvel }
655fb8ef25SArd Biesheuvel
665fb8ef25SArd Biesheuvel /**
67*b2a4411aSRandy Dunlap * chacha_block_generic - generate one keystream block and increment block counter
685fb8ef25SArd Biesheuvel * @state: input state matrix (16 32-bit words)
695fb8ef25SArd Biesheuvel * @stream: output keystream block (64 bytes)
705fb8ef25SArd Biesheuvel * @nrounds: number of rounds (20 or 12; 20 is recommended)
715fb8ef25SArd Biesheuvel *
725fb8ef25SArd Biesheuvel * This is the ChaCha core, a function from 64-byte strings to 64-byte strings.
735fb8ef25SArd Biesheuvel * The caller has already converted the endianness of the input. This function
745fb8ef25SArd Biesheuvel * also handles incrementing the block counter in the input matrix.
755fb8ef25SArd Biesheuvel */
chacha_block_generic(u32 * state,u8 * stream,int nrounds)765fb8ef25SArd Biesheuvel void chacha_block_generic(u32 *state, u8 *stream, int nrounds)
775fb8ef25SArd Biesheuvel {
785fb8ef25SArd Biesheuvel u32 x[16];
795fb8ef25SArd Biesheuvel int i;
805fb8ef25SArd Biesheuvel
815fb8ef25SArd Biesheuvel memcpy(x, state, 64);
825fb8ef25SArd Biesheuvel
835fb8ef25SArd Biesheuvel chacha_permute(x, nrounds);
845fb8ef25SArd Biesheuvel
855fb8ef25SArd Biesheuvel for (i = 0; i < ARRAY_SIZE(x); i++)
865fb8ef25SArd Biesheuvel put_unaligned_le32(x[i] + state[i], &stream[i * sizeof(u32)]);
875fb8ef25SArd Biesheuvel
885fb8ef25SArd Biesheuvel state[12]++;
895fb8ef25SArd Biesheuvel }
905fb8ef25SArd Biesheuvel EXPORT_SYMBOL(chacha_block_generic);
915fb8ef25SArd Biesheuvel
925fb8ef25SArd Biesheuvel /**
935fb8ef25SArd Biesheuvel * hchacha_block_generic - abbreviated ChaCha core, for XChaCha
945fb8ef25SArd Biesheuvel * @state: input state matrix (16 32-bit words)
95*b2a4411aSRandy Dunlap * @stream: output (8 32-bit words)
965fb8ef25SArd Biesheuvel * @nrounds: number of rounds (20 or 12; 20 is recommended)
975fb8ef25SArd Biesheuvel *
985fb8ef25SArd Biesheuvel * HChaCha is the ChaCha equivalent of HSalsa and is an intermediate step
995fb8ef25SArd Biesheuvel * towards XChaCha (see https://cr.yp.to/snuffle/xsalsa-20081128.pdf). HChaCha
1005fb8ef25SArd Biesheuvel * skips the final addition of the initial state, and outputs only certain words
1015fb8ef25SArd Biesheuvel * of the state. It should not be used for streaming directly.
1025fb8ef25SArd Biesheuvel */
hchacha_block_generic(const u32 * state,u32 * stream,int nrounds)1035fb8ef25SArd Biesheuvel void hchacha_block_generic(const u32 *state, u32 *stream, int nrounds)
1045fb8ef25SArd Biesheuvel {
1055fb8ef25SArd Biesheuvel u32 x[16];
1065fb8ef25SArd Biesheuvel
1075fb8ef25SArd Biesheuvel memcpy(x, state, 64);
1085fb8ef25SArd Biesheuvel
1095fb8ef25SArd Biesheuvel chacha_permute(x, nrounds);
1105fb8ef25SArd Biesheuvel
1115fb8ef25SArd Biesheuvel memcpy(&stream[0], &x[0], 16);
1125fb8ef25SArd Biesheuvel memcpy(&stream[4], &x[12], 16);
1135fb8ef25SArd Biesheuvel }
1145fb8ef25SArd Biesheuvel EXPORT_SYMBOL(hchacha_block_generic);
115