aboutsummaryrefslogtreecommitdiffstats
path: root/lib/zinc/poly1305/poly1305-mips64.pl
blob: d30a03d79177f86fce4d54e8be9b3bdb9b7fa851 (plain) (blame)
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
#!/usr/bin/env perl
# SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
#
# This code is taken from the OpenSSL project but the author, Andy Polyakov,
# has relicensed it under the licenses specified in the SPDX header above.
# The original headers, including the original license headers, are
# included below for completeness.
#
# ====================================================================
# Written by Andy Polyakov <appro@openssl.org> for the OpenSSL
# project. The module is, however, dual licensed under OpenSSL and
# CRYPTOGAMS licenses depending on where you obtain it. For further
# details see http://www.openssl.org/~appro/cryptogams/.
# ====================================================================
#
# Poly1305 hash for MIPS64.
#
# May 2016
#
# Numbers are cycles per processed byte with poly1305_blocks alone.
#
#		IALU/gcc
# R1x000	5.64/+120%	(big-endian)
# Octeon II	3.80/+280%	(little-endian)

######################################################################
# There is a number of MIPS ABI in use, O32 and N32/64 are most
# widely used. Then there is a new contender: NUBI. It appears that if
# one picks the latter, it's possible to arrange code in ABI neutral
# manner. Therefore let's stick to NUBI register layout:
#
($zero,$at,$t0,$t1,$t2)=map("\$$_",(0..2,24,25));
($a0,$a1,$a2,$a3,$a4,$a5,$a6,$a7)=map("\$$_",(4..11));
($s0,$s1,$s2,$s3,$s4,$s5,$s6,$s7,$s8,$s9,$s10,$s11)=map("\$$_",(12..23));
($gp,$tp,$sp,$fp,$ra)=map("\$$_",(3,28..31));
#
# The return value is placed in $a0. Following coding rules facilitate
# interoperability:
#
# - never ever touch $tp, "thread pointer", former $gp [o32 can be
#   excluded from the rule, because it's specified volatile];
# - copy return value to $t0, former $v0 [or to $a0 if you're adapting
#   old code];
# - on O32 populate $a4-$a7 with 'lw $aN,4*N($sp)' if necessary;
#
# For reference here is register layout for N32/64 MIPS ABIs:
#
# ($zero,$at,$v0,$v1)=map("\$$_",(0..3));
# ($a0,$a1,$a2,$a3,$a4,$a5,$a6,$a7)=map("\$$_",(4..11));
# ($t0,$t1,$t2,$t3,$t8,$t9)=map("\$$_",(12..15,24,25));
# ($s0,$s1,$s2,$s3,$s4,$s5,$s6,$s7)=map("\$$_",(16..23));
# ($gp,$sp,$fp,$ra)=map("\$$_",(28..31));
#
# <appro@openssl.org>
#
######################################################################

$flavour = shift || "64"; # supported flavours are o32,n32,64,nubi32,nubi64

die "MIPS64 only" unless ($flavour =~ /64|n32/i);

$v0 = ($flavour =~ /nubi/i) ? $a0 : $t0;
$SAVED_REGS_MASK = ($flavour =~ /nubi/i) ? "0x0003f000" : "0x00030000";

($ctx,$inp,$len,$padbit) = ($a0,$a1,$a2,$a3);
($in0,$in1,$tmp0,$tmp1,$tmp2,$tmp3,$tmp4) = ($a4,$a5,$a6,$a7,$at,$t0,$t1);

$code.=<<___;
#if (defined(_MIPS_ARCH_MIPS64R3) || defined(_MIPS_ARCH_MIPS64R5) || \\
     defined(_MIPS_ARCH_MIPS64R6)) \\
     && !defined(_MIPS_ARCH_MIPS64R2)
# define _MIPS_ARCH_MIPS64R2
#endif

#if defined(_MIPS_ARCH_MIPS64R6)
# define dmultu(rs,rt)
# define mflo(rd,rs,rt)	dmulu	rd,rs,rt
# define mfhi(rd,rs,rt)	dmuhu	rd,rs,rt
#else
# define dmultu(rs,rt)		dmultu	rs,rt
# define mflo(rd,rs,rt)	mflo	rd
# define mfhi(rd,rs,rt)	mfhi	rd
#endif

#ifdef	__KERNEL__
# define poly1305_init   poly1305_init_mips
# define poly1305_blocks poly1305_blocks_mips
# define poly1305_emit   poly1305_emit_mips
#endif

#if defined(__MIPSEB__) && !defined(MIPSEB)
# define MIPSEB
#endif

#ifdef MIPSEB
# define MSB 0
# define LSB 7
#else
# define MSB 7
# define LSB 0
#endif

.text
.set	noat
.set	noreorder

.align	5
.globl	poly1305_init
.ent	poly1305_init
poly1305_init:
	.frame	$sp,0,$ra
	.set	reorder

	sd	$zero,0($ctx)
	sd	$zero,8($ctx)
	sd	$zero,16($ctx)

	beqz	$inp,.Lno_key

#if defined(_MIPS_ARCH_MIPS64R6)
	ld	$in0,0($inp)
	ld	$in1,8($inp)
#else
	ldl	$in0,0+MSB($inp)
	ldl	$in1,8+MSB($inp)
	ldr	$in0,0+LSB($inp)
	ldr	$in1,8+LSB($inp)
#endif
#ifdef	MIPSEB
# if defined(_MIPS_ARCH_MIPS64R2)
	dsbh	$in0,$in0		# byte swap
	 dsbh	$in1,$in1
	dshd	$in0,$in0
	 dshd	$in1,$in1
# else
	ori	$tmp0,$zero,0xFF
	dsll	$tmp2,$tmp0,32
	or	$tmp0,$tmp2		# 0x000000FF000000FF

	and	$tmp1,$in0,$tmp0	# byte swap
	 and	$tmp3,$in1,$tmp0
	dsrl	$tmp2,$in0,24
	 dsrl	$tmp4,$in1,24
	dsll	$tmp1,24
	 dsll	$tmp3,24
	and	$tmp2,$tmp0
	 and	$tmp4,$tmp0
	dsll	$tmp0,8			# 0x0000FF000000FF00
	or	$tmp1,$tmp2
	 or	$tmp3,$tmp4
	and	$tmp2,$in0,$tmp0
	 and	$tmp4,$in1,$tmp0
	dsrl	$in0,8
	 dsrl	$in1,8
	dsll	$tmp2,8
	 dsll	$tmp4,8
	and	$in0,$tmp0
	 and	$in1,$tmp0
	or	$tmp1,$tmp2
	 or	$tmp3,$tmp4
	or	$in0,$tmp1
	 or	$in1,$tmp3
	dsrl	$tmp1,$in0,32
	 dsrl	$tmp3,$in1,32
	dsll	$in0,32
	 dsll	$in1,32
	or	$in0,$tmp1
	 or	$in1,$tmp3
# endif
#endif
	li	$tmp0,1
	dsll	$tmp0,32
	daddiu	$tmp0,-63
	dsll	$tmp0,28
	daddiu	$tmp0,-1		# 0ffffffc0fffffff

	and	$in0,$tmp0
	daddiu	$tmp0,-3		# 0ffffffc0ffffffc
	and	$in1,$tmp0

	sd	$in0,24($ctx)
	dsrl	$tmp0,$in1,2
	sd	$in1,32($ctx)
	daddu	$tmp0,$in1		# s1 = r1 + (r1 >> 2)
	sd	$tmp0,40($ctx)

.Lno_key:
	li	$v0,0			# return 0
	jr	$ra
.end	poly1305_init
___
{
my ($h0,$h1,$h2,$r0,$r1,$s1,$d0,$d1,$d2) =
   ($s0,$s1,$s2,$s3,$s4,$s5,$in0,$in1,$t2);

$code.=<<___;
.align	5
.globl	poly1305_blocks
.ent	poly1305_blocks
poly1305_blocks:
	.set	noreorder
	dsrl	$len,4			# number of complete blocks
	bnez	$len,poly1305_blocks_internal
	nop
	jr	$ra
	nop
.end	poly1305_blocks

.align	5
.ent	poly1305_blocks_internal
poly1305_blocks_internal:
	.frame	$sp,6*8,$ra
	.mask	$SAVED_REGS_MASK,-8
	.set	noreorder
	dsubu	$sp,6*8
	sd	$s5,40($sp)
	sd	$s4,32($sp)
___
$code.=<<___ if ($flavour =~ /nubi/i);	# optimize non-nubi prologue
	sd	$s3,24($sp)
	sd	$s2,16($sp)
	sd	$s1,8($sp)
	sd	$s0,0($sp)
___
$code.=<<___;
	.set	reorder

	ld	$h0,0($ctx)		# load hash value
	ld	$h1,8($ctx)
	ld	$h2,16($ctx)

	ld	$r0,24($ctx)		# load key
	ld	$r1,32($ctx)
	ld	$s1,40($ctx)

.Loop:
#if defined(_MIPS_ARCH_MIPS64R6)
	ld	$in0,0($inp)		# load input
	ld	$in1,8($inp)
#else
	ldl	$in0,0+MSB($inp)	# load input
	ldl	$in1,8+MSB($inp)
	ldr	$in0,0+LSB($inp)
	ldr	$in1,8+LSB($inp)
#endif
	daddiu	$len,-1
	daddiu	$inp,16
#ifdef	MIPSEB
# if defined(_MIPS_ARCH_MIPS64R2)
	dsbh	$in0,$in0		# byte swap
	 dsbh	$in1,$in1
	dshd	$in0,$in0
	 dshd	$in1,$in1
# else
	ori	$tmp0,$zero,0xFF
	dsll	$tmp2,$tmp0,32
	or	$tmp0,$tmp2		# 0x000000FF000000FF

	and	$tmp1,$in0,$tmp0	# byte swap
	 and	$tmp3,$in1,$tmp0
	dsrl	$tmp2,$in0,24
	 dsrl	$tmp4,$in1,24
	dsll	$tmp1,24
	 dsll	$tmp3,24
	and	$tmp2,$tmp0
	 and	$tmp4,$tmp0
	dsll	$tmp0,8			# 0x0000FF000000FF00
	or	$tmp1,$tmp2
	 or	$tmp3,$tmp4
	and	$tmp2,$in0,$tmp0
	 and	$tmp4,$in1,$tmp0
	dsrl	$in0,8
	 dsrl	$in1,8
	dsll	$tmp2,8
	 dsll	$tmp4,8
	and	$in0,$tmp0
	 and	$in1,$tmp0
	or	$tmp1,$tmp2
	 or	$tmp3,$tmp4
	or	$in0,$tmp1
	 or	$in1,$tmp3
	dsrl	$tmp1,$in0,32
	 dsrl	$tmp3,$in1,32
	dsll	$in0,32
	 dsll	$in1,32
	or	$in0,$tmp1
	 or	$in1,$tmp3
# endif
#endif
	daddu	$h0,$in0		# accumulate input
	daddu	$h1,$in1
	sltu	$tmp0,$h0,$in0
	sltu	$tmp1,$h1,$in1
	daddu	$h1,$tmp0

	dmultu	($r0,$h0)		# h0*r0
	 daddu	$h2,$padbit
	 sltu	$tmp0,$h1,$tmp0
	mflo	($d0,$r0,$h0)
	mfhi	($d1,$r0,$h0)

	dmultu	($s1,$h1)		# h1*5*r1
	 daddu	$tmp0,$tmp1
	 daddu	$h2,$tmp0
	mflo	($tmp0,$s1,$h1)
	mfhi	($tmp1,$s1,$h1)

	dmultu	($r1,$h0)		# h0*r1
	 daddu	$d0,$tmp0
	 daddu	$d1,$tmp1
	mflo	($tmp2,$r1,$h0)
	mfhi	($d2,$r1,$h0)
	 sltu	$tmp0,$d0,$tmp0
	 daddu	$d1,$tmp0

	dmultu	($r0,$h1)		# h1*r0
	 daddu	$d1,$tmp2
	 sltu	$tmp2,$d1,$tmp2
	mflo	($tmp0,$r0,$h1)
	mfhi	($tmp1,$r0,$h1)
	 daddu	$d2,$tmp2

	dmultu	($s1,$h2)		# h2*5*r1
	 daddu	$d1,$tmp0
	 daddu	$d2,$tmp1
	mflo	($tmp2,$s1,$h2)

	dmultu	($r0,$h2)		# h2*r0
	 sltu	$tmp0,$d1,$tmp0
	 daddu	$d2,$tmp0
	mflo	($tmp3,$r0,$h2)

	daddu	$d1,$tmp2
	daddu	$d2,$tmp3
	sltu	$tmp2,$d1,$tmp2
	daddu	$d2,$tmp2

	li	$tmp0,-4		# final reduction
	and	$tmp0,$d2
	dsrl	$tmp1,$d2,2
	andi	$h2,$d2,3
	daddu	$tmp0,$tmp1
	daddu	$h0,$d0,$tmp0
	sltu	$tmp0,$h0,$tmp0
	daddu	$h1,$d1,$tmp0
	sltu	$tmp0,$h1,$tmp0
	daddu	$h2,$h2,$tmp0

	bnez	$len,.Loop

	sd	$h0,0($ctx)		# store hash value
	sd	$h1,8($ctx)
	sd	$h2,16($ctx)

	.set	noreorder
	ld	$s5,40($sp)		# epilogue
	ld	$s4,32($sp)
___
$code.=<<___ if ($flavour =~ /nubi/i);	# optimize non-nubi epilogue
	ld	$s3,24($sp)
	ld	$s2,16($sp)
	ld	$s1,8($sp)
	ld	$s0,0($sp)
___
$code.=<<___;
	jr	$ra
	daddu	$sp,6*8
.end	poly1305_blocks_internal
___
}
{
my ($ctx,$mac,$nonce) = ($a0,$a1,$a2);

$code.=<<___;
.align	5
.globl	poly1305_emit
.ent	poly1305_emit
poly1305_emit:
	.frame	$sp,0,$ra
	.set	reorder

	ld	$tmp0,0($ctx)
	ld	$tmp1,8($ctx)
	ld	$tmp2,16($ctx)

	daddiu	$in0,$tmp0,5		# compare to modulus
	sltiu	$tmp3,$in0,5
	daddu	$in1,$tmp1,$tmp3
	sltu	$tmp3,$in1,$tmp3
	daddu	$tmp2,$tmp2,$tmp3

	dsrl	$tmp2,2			# see if it carried/borrowed
	dsubu	$tmp2,$zero,$tmp2
	nor	$tmp3,$zero,$tmp2

	and	$in0,$tmp2
	and	$tmp0,$tmp3
	and	$in1,$tmp2
	and	$tmp1,$tmp3
	or	$in0,$tmp0
	or	$in1,$tmp1

	lwu	$tmp0,0($nonce)		# load nonce
	lwu	$tmp1,4($nonce)
	lwu	$tmp2,8($nonce)
	lwu	$tmp3,12($nonce)
	dsll	$tmp1,32
	dsll	$tmp3,32
	or	$tmp0,$tmp1
	or	$tmp2,$tmp3

	daddu	$in0,$tmp0		# accumulate nonce
	daddu	$in1,$tmp2
	sltu	$tmp0,$in0,$tmp0
	daddu	$in1,$tmp0

	dsrl	$tmp0,$in0,8		# write mac value
	dsrl	$tmp1,$in0,16
	dsrl	$tmp2,$in0,24
	sb	$in0,0($mac)
	dsrl	$tmp3,$in0,32
	sb	$tmp0,1($mac)
	dsrl	$tmp0,$in0,40
	sb	$tmp1,2($mac)
	dsrl	$tmp1,$in0,48
	sb	$tmp2,3($mac)
	dsrl	$tmp2,$in0,56
	sb	$tmp3,4($mac)
	dsrl	$tmp3,$in1,8
	sb	$tmp0,5($mac)
	dsrl	$tmp0,$in1,16
	sb	$tmp1,6($mac)
	dsrl	$tmp1,$in1,24
	sb	$tmp2,7($mac)

	sb	$in1,8($mac)
	dsrl	$tmp2,$in1,32
	sb	$tmp3,9($mac)
	dsrl	$tmp3,$in1,40
	sb	$tmp0,10($mac)
	dsrl	$tmp0,$in1,48
	sb	$tmp1,11($mac)
	dsrl	$tmp1,$in1,56
	sb	$tmp2,12($mac)
	sb	$tmp3,13($mac)
	sb	$tmp0,14($mac)
	sb	$tmp1,15($mac)

	jr	$ra
.end	poly1305_emit
.rdata
.align	2
___
}

open SELF,$0;
while(<SELF>) {
	next if (/^#!/);
	last if (!s/^#/\/\// and !/^$/);
	print;
}
close SELF;

$output=pop and open STDOUT,">$output";
print $code;
close STDOUT;