aboutsummaryrefslogblamecommitdiffstatshomepage
path: root/src/crypto/chacha20-mips.S
blob: 77da2c2fb240ea516ac981df0b0c1d3f2f5df1ee (plain) (tree)
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
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474

























































































































































































































































































































































































































































































                                                                                        
/* SPDX-License-Identifier: GPL-2.0
 *
 * Copyright (C) 2016-2018 René van Dorst <opensource@vdorst.com>. All Rights Reserved.
 * Copyright (C) 2015-2018 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
 */

#define MASK_U32	0x3c
#define MASK_BYTES	0x03
#define CHACHA20_BLOCK_SIZE 64
#define STACK_SIZE	4*16

#define X0  $t0
#define X1  $t1
#define X2  $t2
#define X3  $t3
#define X4  $t4
#define X5  $t5
#define X6  $t6
#define X7  $t7
#define X8  $v1
#define X9  $fp
#define X10 $s7
#define X11 $s6
#define X12 $s5
#define X13 $s4
#define X14 $s3
#define X15 $s2
/* Use regs which are overwritten on exit for Tx so we don't leak clear data. */
#define T0  $s1
#define T1  $s0
#define T(n) T ## n
#define X(n) X ## n

/* Input arguments */
#define OUT		$a0
#define IN		$a1
#define BYTES		$a2
/* KEY and NONCE argument must be u32 aligned */
#define KEY		$a3
/* NONCE pointer is given via stack */
#define NONCE		$t9

/* Output argument */
/* NONCE[0] is kept in a register and not in memory.
 * We don't want to touch original value in memory.
 * Must be incremented every loop iteration.
 */
#define NONCE_0		$v0

/* SAVED_X and SAVED_CA are set in the jump table.
 * Use regs which are overwritten on exit else we don't leak clear data.
 * They are used to handling the last bytes which are not multiple of 4.
 */
#define SAVED_X		X15
#define SAVED_CA	$ra

#define PTR_LAST_ROUND	$t8

/* ChaCha20 constants and stack location */
#define CONSTANT_OFS_SP	48
#define UNALIGNED_OFS_SP 40

#define CONSTANT_1	0x61707865
#define CONSTANT_2	0x3320646e
#define CONSTANT_3	0x79622d32
#define CONSTANT_4	0x6b206574

#if __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__
#define MSB 0
#define LSB 3
#define ROTx rotl
#define ROTR(n) rotr n, 24
#define	CPU_TO_LE32(n) \
	wsbh	n; \
	rotr	n, 16;
#else
#define MSB 3
#define LSB 0
#define ROTx rotr
#define CPU_TO_LE32(n)
#define ROTR(n)
#endif

#define STORE_UNALIGNED(x, a, s, o) \
.Lchacha20_mips_xor_unaligned_ ## x ## _b: ; \
	.if ((s != NONCE) || (o != 0)); \
		lw	T0, o(s); \
	.endif; \
	lwl	T1, x-4+MSB ## (IN); \
	lwr	T1, x-4+LSB ## (IN); \
	.if ((s == NONCE) && (o == 0)); \
		addu	X ## a, NONCE_0; \
	.else; \
		addu	X ## a, T0; \
	.endif; \
	CPU_TO_LE32(X ## a); \
	xor	X ## a, T1; \
	swl	X ## a, x-4+MSB ## (OUT); \
	swr	X ## a, x-4+LSB ## (OUT);

#define STORE_ALIGNED(x, a, s, o) \
.Lchacha20_mips_xor_aligned_ ## x ## _b: ; \
	.if ((s != NONCE) || (o != 0)); \
		lw	T0, o(s); \
	.endif; \
	lw	T1, x-4 ## (IN); \
	.if ((s == NONCE) && (o == 0)); \
		addu	X ## a, NONCE_0; \
	.else; \
		addu	X ## a, T0; \
	.endif; \
	CPU_TO_LE32(X ## a); \
	xor	X ## a, T1; \
	sw	X ## a, x-4 ## (OUT);

/* Jump table macro.
 * Used for setup and handling the last bytes, which are not multiple of 4.
 * X15 is free to store Xn
 * Every jumptable entry must be equal in size.
 */
#define JMPTBL_ALIGNED(x, a, s, o) \
.Lchacha20_mips_jmptbl_aligned_ ## a: ; \
	.if ((s == NONCE) && (o == 0)); \
		move	SAVED_CA, NONCE_0; \
	.else; \
		lw	SAVED_CA, o(s);\
	.endif; \
	b	.Lchacha20_mips_xor_aligned_ ## x ## _b; \
	move	SAVED_X, X ## a;

#define JMPTBL_UNALIGNED(x, a, s, o) \
.Lchacha20_mips_jmptbl_unaligned_ ## a: ; \
	.if ((s == NONCE) && (o == 0)); \
		move	SAVED_CA, NONCE_0; \
	.else; \
		lw	SAVED_CA, o(s);\
	.endif; \
	b	.Lchacha20_mips_xor_unaligned_ ## x ## _b; \
	move	SAVED_X, X ## a;

#define AXR(A, B, C, D,  K, L, M, N,  V, W, Y, Z,  S) \
	addu	X(A), X(K); \
	addu	X(B), X(L); \
	addu	X(C), X(M); \
	addu	X(D), X(N); \
	xor	X(V), X(A); \
	xor	X(W), X(B); \
	xor	X(Y), X(C); \
	xor	X(Z), X(D); \
	rotl	X(V), S;    \
	rotl	X(W), S;    \
	rotl	X(Y), S;    \
	rotl	X(Z), S;

.text
.set reorder
.set noat
.globl chacha20_mips
.ent   chacha20_mips
chacha20_mips:
	.frame $sp, STACK_SIZE, $ra
	/* This is in the fifth argument */
	lw	NONCE, 16($sp)

	/* Return bytes = 0. */
	.set noreorder
	beqz	BYTES, .Lchacha20_mips_end
	addiu	$sp, -STACK_SIZE
	.set reorder

	/* Calculate PTR_LAST_ROUND */
	addiu	PTR_LAST_ROUND, BYTES, -1
	ins	PTR_LAST_ROUND, $zero, 0, 6
	addu	PTR_LAST_ROUND, OUT

	/* Save s0-s7, fp, ra. */
	sw	$ra,  0($sp)
	sw	$fp,  4($sp)
	sw	$s0,  8($sp)
	sw	$s1, 12($sp)
	sw	$s2, 16($sp)
	sw	$s3, 20($sp)
	sw	$s4, 24($sp)
	sw	$s5, 28($sp)
	sw	$s6, 32($sp)
	sw	$s7, 36($sp)

	lw	NONCE_0, 0(NONCE)
	/* Test IN or OUT is unaligned.
	 * UNALIGNED (T1) = ( IN | OUT ) & 0x00000003
	 */
	or	T1, IN, OUT
	andi	T1, 0x3

	/* Load constant */
	lui	X0, %hi(CONSTANT_1)
	lui	X1, %hi(CONSTANT_2)
	lui	X2, %hi(CONSTANT_3)
	lui	X3, %hi(CONSTANT_4)
	ori	X0, %lo(CONSTANT_1)
	ori	X1, %lo(CONSTANT_2)
	ori	X2, %lo(CONSTANT_3)
	ori	X3, %lo(CONSTANT_4)

	/* Store constant on stack. */
	sw	X0,  0+CONSTANT_OFS_SP($sp)
	sw	X1,  4+CONSTANT_OFS_SP($sp)
	sw	X2,  8+CONSTANT_OFS_SP($sp)
	sw	X3, 12+CONSTANT_OFS_SP($sp)

	sw	T1, UNALIGNED_OFS_SP($sp)

	.set	noreorder
	b	.Lchacha20_rounds_start
	andi	BYTES, (CHACHA20_BLOCK_SIZE-1)
	.set	reorder

.align 4
.Loop_chacha20_rounds:
	addiu	IN,  CHACHA20_BLOCK_SIZE
	addiu	OUT, CHACHA20_BLOCK_SIZE
	addiu	NONCE_0, 1

	lw	X0,  0+CONSTANT_OFS_SP($sp)
	lw	X1,  4+CONSTANT_OFS_SP($sp)
	lw	X2,  8+CONSTANT_OFS_SP($sp)
	lw	X3, 12+CONSTANT_OFS_SP($sp)
	lw	T1,   UNALIGNED_OFS_SP($sp)

.Lchacha20_rounds_start:
	lw	X4,   0(KEY)
	lw	X5,   4(KEY)
	lw	X6,   8(KEY)
	lw	X7,  12(KEY)
	lw	X8,  16(KEY)
	lw	X9,  20(KEY)
	lw	X10, 24(KEY)
	lw	X11, 28(KEY)

	move	X12, NONCE_0
	lw	X13,  4(NONCE)
	lw	X14,  8(NONCE)
	lw	X15, 12(NONCE)

	li	$at, 9
.Loop_chacha20_xor_rounds:
	AXR( 0, 1, 2, 3,  4, 5, 6, 7, 12,13,14,15, 16);
	AXR( 8, 9,10,11, 12,13,14,15,  4, 5, 6, 7, 12);
	AXR( 0, 1, 2, 3,  4, 5, 6, 7, 12,13,14,15,  8);
	AXR( 8, 9,10,11, 12,13,14,15,  4, 5, 6, 7,  7);
	AXR( 0, 1, 2, 3,  5, 6, 7, 4, 15,12,13,14, 16);
	AXR(10,11, 8, 9, 15,12,13,14,  5, 6, 7, 4, 12);
	AXR( 0, 1, 2, 3,  5, 6, 7, 4, 15,12,13,14,  8);
	AXR(10,11, 8, 9, 15,12,13,14,  5, 6, 7, 4,  7);
	.set noreorder
	bnez	$at, .Loop_chacha20_xor_rounds
	addiu	$at, -1

	/* Unaligned? Jump */
	bnez	T1, .Loop_chacha20_unaligned
	andi	$at, BYTES, MASK_U32

	/* Last round? No jump */
	bne	OUT, PTR_LAST_ROUND, .Lchacha20_mips_xor_aligned_64_b
	/* Load upper half of jump table addr */
	lui	T0, %hi(.Lchacha20_mips_jmptbl_aligned_0)

	/* Full block? Jump */
	beqz	BYTES, .Lchacha20_mips_xor_aligned_64_b
	/* Calculate lower half jump table addr and offset */
	ins	T0, $at, 2, 6

	subu	T0, $at
	addiu	T0, %lo(.Lchacha20_mips_jmptbl_aligned_0)

	jr	T0
	/* Delay slot */
	nop

	.set	reorder

.Loop_chacha20_unaligned:
	.set noreorder

	/* Last round? no jump */
	bne	OUT, PTR_LAST_ROUND, .Lchacha20_mips_xor_unaligned_64_b
	/* Load upper half of jump table addr */
	lui	T0, %hi(.Lchacha20_mips_jmptbl_unaligned_0)

	/* Full block? Jump */
	beqz	BYTES, .Lchacha20_mips_xor_unaligned_64_b

	/* Calculate lower half jump table addr and offset */
	ins     T0, $at, 2, 6
	subu	T0, $at
	addiu	T0, %lo(.Lchacha20_mips_jmptbl_unaligned_0)

	jr	T0
	/* Delay slot */
	nop

	.set	reorder

/* Aligned code path
 */
.align 4
	STORE_ALIGNED(64, 15, NONCE,12)
	STORE_ALIGNED(60, 14, NONCE, 8)
	STORE_ALIGNED(56, 13, NONCE, 4)
	STORE_ALIGNED(52, 12, NONCE, 0)
	STORE_ALIGNED(48, 11, KEY, 28)
	STORE_ALIGNED(44, 10, KEY, 24)
	STORE_ALIGNED(40,  9, KEY, 20)
	STORE_ALIGNED(36,  8, KEY, 16)
	STORE_ALIGNED(32,  7, KEY, 12)
	STORE_ALIGNED(28,  6, KEY,  8)
	STORE_ALIGNED(24,  5, KEY,  4)
	STORE_ALIGNED(20,  4, KEY,  0)
	STORE_ALIGNED(16,  3, $sp, 12+CONSTANT_OFS_SP)
	STORE_ALIGNED(12,  2, $sp,  8+CONSTANT_OFS_SP)
	STORE_ALIGNED( 8,  1, $sp,  4+CONSTANT_OFS_SP)
.Lchacha20_mips_xor_aligned_4_b:
	/* STORE_ALIGNED( 4,  0, $sp, 0+CONSTANT_OFS_SP) */
	lw	T0, 0+CONSTANT_OFS_SP($sp)
	lw	T1, 0(IN)
	addu	X0, T0
	CPU_TO_LE32(X0)
	xor	X0, T1
	.set noreorder
	bne	OUT, PTR_LAST_ROUND, .Loop_chacha20_rounds
	sw	X0, 0(OUT)
	.set reorder

	.set noreorder
	bne	$at, BYTES, .Lchacha20_mips_xor_bytes
	/* Empty delayslot, Increase NONCE_0, return NONCE_0 value */
	addiu	NONCE_0, 1
	.set noreorder

.Lchacha20_mips_xor_done:
	/* Restore used registers */
	lw	$ra,  0($sp)
	lw	$fp,  4($sp)
	lw	$s0,  8($sp)
	lw	$s1, 12($sp)
	lw	$s2, 16($sp)
	lw	$s3, 20($sp)
	lw	$s4, 24($sp)
	lw	$s5, 28($sp)
	lw	$s6, 32($sp)
	lw	$s7, 36($sp)
.Lchacha20_mips_end:
	.set noreorder
	jr	$ra
	addiu	$sp, STACK_SIZE
	.set reorder

	.set noreorder
	/* Start jump table */
	JMPTBL_ALIGNED( 0,  0, $sp,  0+CONSTANT_OFS_SP)
	JMPTBL_ALIGNED( 4,  1, $sp,  4+CONSTANT_OFS_SP)
	JMPTBL_ALIGNED( 8,  2, $sp,  8+CONSTANT_OFS_SP)
	JMPTBL_ALIGNED(12,  3, $sp, 12+CONSTANT_OFS_SP)
	JMPTBL_ALIGNED(16,  4, KEY,  0)
	JMPTBL_ALIGNED(20,  5, KEY,  4)
	JMPTBL_ALIGNED(24,  6, KEY,  8)
	JMPTBL_ALIGNED(28,  7, KEY, 12)
	JMPTBL_ALIGNED(32,  8, KEY, 16)
	JMPTBL_ALIGNED(36,  9, KEY, 20)
	JMPTBL_ALIGNED(40, 10, KEY, 24)
	JMPTBL_ALIGNED(44, 11, KEY, 28)
	JMPTBL_ALIGNED(48, 12, NONCE, 0)
	JMPTBL_ALIGNED(52, 13, NONCE, 4)
	JMPTBL_ALIGNED(56, 14, NONCE, 8)
	JMPTBL_ALIGNED(60, 15, NONCE,12)
	/* End jump table */
	.set reorder

/* Unaligned code path
 */
	STORE_UNALIGNED(64, 15, NONCE,12)
	STORE_UNALIGNED(60, 14, NONCE, 8)
	STORE_UNALIGNED(56, 13, NONCE, 4)
	STORE_UNALIGNED(52, 12, NONCE, 0)
	STORE_UNALIGNED(48, 11, KEY, 28)
	STORE_UNALIGNED(44, 10, KEY, 24)
	STORE_UNALIGNED(40,  9, KEY, 20)
	STORE_UNALIGNED(36,  8, KEY, 16)
	STORE_UNALIGNED(32,  7, KEY, 12)
	STORE_UNALIGNED(28,  6, KEY,  8)
	STORE_UNALIGNED(24,  5, KEY,  4)
	STORE_UNALIGNED(20,  4, KEY,  0)
	STORE_UNALIGNED(16,  3, $sp, 12+CONSTANT_OFS_SP)
	STORE_UNALIGNED(12,  2, $sp,  8+CONSTANT_OFS_SP)
	STORE_UNALIGNED( 8,  1, $sp,  4+CONSTANT_OFS_SP)
.Lchacha20_mips_xor_unaligned_4_b:
	/* STORE_UNALIGNED( 4,  0, $sp, 0+CONSTANT_OFS_SP) */
	lw	T0, 0+CONSTANT_OFS_SP($sp)
	lwl	T1, 0+MSB(IN)
	lwr	T1, 0+LSB(IN)
	addu	X0, T0
	CPU_TO_LE32(X0)
	xor	X0, T1
	swl	X0, 0+MSB(OUT)
	.set noreorder
	bne	OUT, PTR_LAST_ROUND, .Loop_chacha20_rounds
	swr	X0, 0+LSB(OUT)
	.set reorder

	/* Fall through to byte handling */
	.set noreorder
	beq	$at, BYTES, .Lchacha20_mips_xor_done
	/* Empty delayslot, increase NONCE_0, return NONCE_0 value */
.Lchacha20_mips_xor_unaligned_0_b:
.Lchacha20_mips_xor_aligned_0_b:
	addiu	NONCE_0, 1
	.set reorder

.Lchacha20_mips_xor_bytes:
	addu	OUT, $at
	addu	IN, $at
	addu	SAVED_X, SAVED_CA
	/* First byte */
	lbu	T1, 0(IN)
	andi	$at, BYTES, 2
	CPU_TO_LE32(SAVED_X)
	ROTR(SAVED_X)
	xor	T1, SAVED_X
	.set noreorder
	beqz	$at, .Lchacha20_mips_xor_done
	sb	T1, 0(OUT)
	.set reorder
	/* Second byte */
	lbu	T1, 1(IN)
	andi	$at, BYTES, 1
	ROTx	SAVED_X, 8
	xor	T1, SAVED_X
	.set noreorder
	beqz	$at, .Lchacha20_mips_xor_done
	sb	T1, 1(OUT)
	.set reorder
	/* Third byte */
	lbu	T1, 2(IN)
	ROTx	SAVED_X, 8
	xor	T1, SAVED_X
	.set noreorder
	b	.Lchacha20_mips_xor_done
	sb	T1, 2(OUT)
	.set reorder
.set noreorder

.Lchacha20_mips_jmptbl_unaligned:
	/* Start jump table */
	JMPTBL_UNALIGNED( 0,  0, $sp,  0+CONSTANT_OFS_SP)
	JMPTBL_UNALIGNED( 4,  1, $sp,  4+CONSTANT_OFS_SP)
	JMPTBL_UNALIGNED( 8,  2, $sp,  8+CONSTANT_OFS_SP)
	JMPTBL_UNALIGNED(12,  3, $sp, 12+CONSTANT_OFS_SP)
	JMPTBL_UNALIGNED(16,  4, KEY,  0)
	JMPTBL_UNALIGNED(20,  5, KEY,  4)
	JMPTBL_UNALIGNED(24,  6, KEY,  8)
	JMPTBL_UNALIGNED(28,  7, KEY, 12)
	JMPTBL_UNALIGNED(32,  8, KEY, 16)
	JMPTBL_UNALIGNED(36,  9, KEY, 20)
	JMPTBL_UNALIGNED(40, 10, KEY, 24)
	JMPTBL_UNALIGNED(44, 11, KEY, 28)
	JMPTBL_UNALIGNED(48, 12, NONCE, 0)
	JMPTBL_UNALIGNED(52, 13, NONCE, 4)
	JMPTBL_UNALIGNED(56, 14, NONCE, 8)
	JMPTBL_UNALIGNED(60, 15, NONCE,12)
	/* End jump table */
.set reorder

.end chacha20_mips
.set at