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
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
|
#! /usr/bin/env perl
# Copyright 2016 The OpenSSL Project Authors. All Rights Reserved.
#
# Licensed under the OpenSSL license (the "License"). You may not use
# this file except in compliance with the License. You can obtain a copy
# in the file LICENSE in the source distribution or at
# https://www.openssl.org/source/license.html
#
# ====================================================================
# 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/.
# ====================================================================
#
# This module implements Poly1305 hash for PowerPC FPU.
#
# June 2015
#
# Numbers are cycles per processed byte with poly1305_blocks alone,
# and improvement coefficients relative to gcc-generated code.
#
# Freescale e300 9.78/+30%
# PPC74x0 6.92/+50%
# PPC970 6.03/+80%
# POWER7 3.50/+30%
# POWER8 3.75/+10%
$flavour = shift;
if ($flavour =~ /64/) {
$SIZE_T =8;
$LRSAVE =2*$SIZE_T;
$UCMP ="cmpld";
$STU ="stdu";
$POP ="ld";
$PUSH ="std";
} elsif ($flavour =~ /32/) {
$SIZE_T =4;
$LRSAVE =$SIZE_T;
$UCMP ="cmplw";
$STU ="stwu";
$POP ="lwz";
$PUSH ="stw";
} else { die "nonsense $flavour"; }
$LITTLE_ENDIAN = ($flavour=~/le$/) ? 4 : 0;
$LWXLE = $LITTLE_ENDIAN ? "lwzx" : "lwbrx";
$0 =~ m/(.*[\/\\])[^\/\\]+$/; $dir=$1;
( $xlate="${dir}ppc-xlate.pl" and -f $xlate ) or
( $xlate="${dir}../../perlasm/ppc-xlate.pl" and -f $xlate) or
die "can't locate ppc-xlate.pl";
open STDOUT,"| $^X $xlate $flavour ".shift || die "can't call $xlate: $!";
$LOCALS=6*$SIZE_T;
$FRAME=$LOCALS+6*8+18*8;
my $sp="r1";
my ($ctx,$inp,$len,$padbit) = map("r$_",(3..6));
my ($in0,$in1,$in2,$in3,$i1,$i2,$i3) = map("r$_",(7..12,6));
my ($h0lo,$h0hi,$h1lo,$h1hi,$h2lo,$h2hi,$h3lo,$h3hi,
$two0,$two32,$two64,$two96,$two130,$five_two130,
$r0lo,$r0hi,$r1lo,$r1hi,$r2lo,$r2hi,
$s2lo,$s2hi,$s3lo,$s3hi,
$c0lo,$c0hi,$c1lo,$c1hi,$c2lo,$c2hi,$c3lo,$c3hi) = map("f$_",(0..31));
# borrowings
my ($r3lo,$r3hi,$s1lo,$s1hi) = ($c0lo,$c0hi,$c1lo,$c1hi);
my ($x0,$x1,$x2,$x3) = ($c2lo,$c2hi,$c3lo,$c3hi);
my ($y0,$y1,$y2,$y3) = ($c3lo,$c3hi,$c1lo,$c1hi);
$code.=<<___;
.machine "any"
.text
.globl .poly1305_init_fpu
.align 6
.poly1305_init_fpu:
$STU $sp,-$LOCALS($sp) # minimal frame
mflr $padbit
$PUSH $padbit,`$LOCALS+$LRSAVE`($sp)
bl LPICmeup
xor r0,r0,r0
mtlr $padbit # restore lr
lfd $two0,8*0($len) # load constants
lfd $two32,8*1($len)
lfd $two64,8*2($len)
lfd $two96,8*3($len)
lfd $two130,8*4($len)
lfd $five_two130,8*5($len)
stfd $two0,8*0($ctx) # initial hash value, biased 0
stfd $two32,8*1($ctx)
stfd $two64,8*2($ctx)
stfd $two96,8*3($ctx)
$UCMP $inp,r0
beq- Lno_key
lfd $h3lo,8*13($len) # new fpscr
mffs $h3hi # old fpscr
stfd $two0,8*4($ctx) # key "template"
stfd $two32,8*5($ctx)
stfd $two64,8*6($ctx)
stfd $two96,8*7($ctx)
li $in1,4
li $in2,8
li $in3,12
$LWXLE $in0,0,$inp # load key
$LWXLE $in1,$in1,$inp
$LWXLE $in2,$in2,$inp
$LWXLE $in3,$in3,$inp
lis $i1,0xf000 # 0xf0000000
ori $i2,$i1,3 # 0xf0000003
andc $in0,$in0,$i1 # &=0x0fffffff
andc $in1,$in1,$i2 # &=0x0ffffffc
andc $in2,$in2,$i2
andc $in3,$in3,$i2
stw $in0,`8*4+(4^$LITTLE_ENDIAN)`($ctx) # fill "template"
stw $in1,`8*5+(4^$LITTLE_ENDIAN)`($ctx)
stw $in2,`8*6+(4^$LITTLE_ENDIAN)`($ctx)
stw $in3,`8*7+(4^$LITTLE_ENDIAN)`($ctx)
mtfsf 255,$h3lo # fpscr
stfd $two0,8*18($ctx) # copy constants to context
stfd $two32,8*19($ctx)
stfd $two64,8*20($ctx)
stfd $two96,8*21($ctx)
stfd $two130,8*22($ctx)
stfd $five_two130,8*23($ctx)
lfd $h0lo,8*4($ctx) # load [biased] key
lfd $h1lo,8*5($ctx)
lfd $h2lo,8*6($ctx)
lfd $h3lo,8*7($ctx)
fsub $h0lo,$h0lo,$two0 # r0
fsub $h1lo,$h1lo,$two32 # r1
fsub $h2lo,$h2lo,$two64 # r2
fsub $h3lo,$h3lo,$two96 # r3
lfd $two0,8*6($len) # more constants
lfd $two32,8*7($len)
lfd $two64,8*8($len)
lfd $two96,8*9($len)
fmul $h1hi,$h1lo,$five_two130 # s1
fmul $h2hi,$h2lo,$five_two130 # s2
stfd $h3hi,8*15($ctx) # borrow slot for original fpscr
fmul $h3hi,$h3lo,$five_two130 # s3
fadd $h0hi,$h0lo,$two0
stfd $h1hi,8*12($ctx) # put aside for now
fadd $h1hi,$h1lo,$two32
stfd $h2hi,8*13($ctx)
fadd $h2hi,$h2lo,$two64
stfd $h3hi,8*14($ctx)
fadd $h3hi,$h3lo,$two96
fsub $h0hi,$h0hi,$two0
fsub $h1hi,$h1hi,$two32
fsub $h2hi,$h2hi,$two64
fsub $h3hi,$h3hi,$two96
lfd $two0,8*10($len) # more constants
lfd $two32,8*11($len)
lfd $two64,8*12($len)
fsub $h0lo,$h0lo,$h0hi
fsub $h1lo,$h1lo,$h1hi
fsub $h2lo,$h2lo,$h2hi
fsub $h3lo,$h3lo,$h3hi
stfd $h0hi,8*5($ctx) # r0hi
stfd $h1hi,8*7($ctx) # r1hi
stfd $h2hi,8*9($ctx) # r2hi
stfd $h3hi,8*11($ctx) # r3hi
stfd $h0lo,8*4($ctx) # r0lo
stfd $h1lo,8*6($ctx) # r1lo
stfd $h2lo,8*8($ctx) # r2lo
stfd $h3lo,8*10($ctx) # r3lo
lfd $h1lo,8*12($ctx) # s1
lfd $h2lo,8*13($ctx) # s2
lfd $h3lo,8*14($ctx) # s3
lfd $h0lo,8*15($ctx) # pull original fpscr
fadd $h1hi,$h1lo,$two0
fadd $h2hi,$h2lo,$two32
fadd $h3hi,$h3lo,$two64
fsub $h1hi,$h1hi,$two0
fsub $h2hi,$h2hi,$two32
fsub $h3hi,$h3hi,$two64
fsub $h1lo,$h1lo,$h1hi
fsub $h2lo,$h2lo,$h2hi
fsub $h3lo,$h3lo,$h3hi
stfd $h1hi,8*13($ctx) # s1hi
stfd $h2hi,8*15($ctx) # s2hi
stfd $h3hi,8*17($ctx) # s3hi
stfd $h1lo,8*12($ctx) # s1lo
stfd $h2lo,8*14($ctx) # s2lo
stfd $h3lo,8*16($ctx) # s3lo
mtfsf 255,$h0lo # restore fpscr
Lno_key:
xor r3,r3,r3
addi $sp,$sp,$LOCALS
blr
.long 0
.byte 0,12,4,1,0x80,0,2,0
.size .poly1305_init_fpu,.-.poly1305_init_fpu
.globl .poly1305_blocks_fpu
.align 4
.poly1305_blocks_fpu:
srwi. $len,$len,4
beq- Labort
$STU $sp,-$FRAME($sp)
mflr r0
stfd f14,`$FRAME-8*18`($sp)
stfd f15,`$FRAME-8*17`($sp)
stfd f16,`$FRAME-8*16`($sp)
stfd f17,`$FRAME-8*15`($sp)
stfd f18,`$FRAME-8*14`($sp)
stfd f19,`$FRAME-8*13`($sp)
stfd f20,`$FRAME-8*12`($sp)
stfd f21,`$FRAME-8*11`($sp)
stfd f22,`$FRAME-8*10`($sp)
stfd f23,`$FRAME-8*9`($sp)
stfd f24,`$FRAME-8*8`($sp)
stfd f25,`$FRAME-8*7`($sp)
stfd f26,`$FRAME-8*6`($sp)
stfd f27,`$FRAME-8*5`($sp)
stfd f28,`$FRAME-8*4`($sp)
stfd f29,`$FRAME-8*3`($sp)
stfd f30,`$FRAME-8*2`($sp)
stfd f31,`$FRAME-8*1`($sp)
$PUSH r0,`$FRAME+$LRSAVE`($sp)
xor r0,r0,r0
li $in3,1
mtctr $len
neg $len,$len
stw r0,`$LOCALS+8*4+(0^$LITTLE_ENDIAN)`($sp)
stw $in3,`$LOCALS+8*4+(4^$LITTLE_ENDIAN)`($sp)
lfd $two0,8*18($ctx) # load constants
lfd $two32,8*19($ctx)
lfd $two64,8*20($ctx)
lfd $two96,8*21($ctx)
lfd $two130,8*22($ctx)
lfd $five_two130,8*23($ctx)
lfd $h0lo,8*0($ctx) # load [biased] hash value
lfd $h1lo,8*1($ctx)
lfd $h2lo,8*2($ctx)
lfd $h3lo,8*3($ctx)
stfd $two0,`$LOCALS+8*0`($sp) # input "template"
oris $in3,$padbit,`(1023+52+96)<<4`
stfd $two32,`$LOCALS+8*1`($sp)
stfd $two64,`$LOCALS+8*2`($sp)
stw $in3,`$LOCALS+8*3+(0^$LITTLE_ENDIAN)`($sp)
li $i1,4
li $i2,8
li $i3,12
$LWXLE $in0,0,$inp # load input
$LWXLE $in1,$i1,$inp
$LWXLE $in2,$i2,$inp
$LWXLE $in3,$i3,$inp
addi $inp,$inp,16
stw $in0,`$LOCALS+8*0+(4^$LITTLE_ENDIAN)`($sp) # fill "template"
stw $in1,`$LOCALS+8*1+(4^$LITTLE_ENDIAN)`($sp)
stw $in2,`$LOCALS+8*2+(4^$LITTLE_ENDIAN)`($sp)
stw $in3,`$LOCALS+8*3+(4^$LITTLE_ENDIAN)`($sp)
mffs $x0 # original fpscr
lfd $x1,`$LOCALS+8*4`($sp) # new fpscr
lfd $r0lo,8*4($ctx) # load key
lfd $r0hi,8*5($ctx)
lfd $r1lo,8*6($ctx)
lfd $r1hi,8*7($ctx)
lfd $r2lo,8*8($ctx)
lfd $r2hi,8*9($ctx)
lfd $r3lo,8*10($ctx)
lfd $r3hi,8*11($ctx)
lfd $s1lo,8*12($ctx)
lfd $s1hi,8*13($ctx)
lfd $s2lo,8*14($ctx)
lfd $s2hi,8*15($ctx)
lfd $s3lo,8*16($ctx)
lfd $s3hi,8*17($ctx)
stfd $x0,`$LOCALS+8*4`($sp) # save original fpscr
mtfsf 255,$x1
addic $len,$len,1
addze r0,r0
slwi. r0,r0,4
sub $inp,$inp,r0 # conditional rewind
lfd $x0,`$LOCALS+8*0`($sp)
lfd $x1,`$LOCALS+8*1`($sp)
lfd $x2,`$LOCALS+8*2`($sp)
lfd $x3,`$LOCALS+8*3`($sp)
fsub $h0lo,$h0lo,$two0 # de-bias hash value
$LWXLE $in0,0,$inp # modulo-scheduled input load
fsub $h1lo,$h1lo,$two32
$LWXLE $in1,$i1,$inp
fsub $h2lo,$h2lo,$two64
$LWXLE $in2,$i2,$inp
fsub $h3lo,$h3lo,$two96
$LWXLE $in3,$i3,$inp
fsub $x0,$x0,$two0 # de-bias input
addi $inp,$inp,16
fsub $x1,$x1,$two32
fsub $x2,$x2,$two64
fsub $x3,$x3,$two96
fadd $x0,$x0,$h0lo # accumulate input
stw $in0,`$LOCALS+8*0+(4^$LITTLE_ENDIAN)`($sp)
fadd $x1,$x1,$h1lo
stw $in1,`$LOCALS+8*1+(4^$LITTLE_ENDIAN)`($sp)
fadd $x2,$x2,$h2lo
stw $in2,`$LOCALS+8*2+(4^$LITTLE_ENDIAN)`($sp)
fadd $x3,$x3,$h3lo
stw $in3,`$LOCALS+8*3+(4^$LITTLE_ENDIAN)`($sp)
b Lentry
.align 4
Loop:
fsub $y0,$y0,$two0 # de-bias input
addic $len,$len,1
fsub $y1,$y1,$two32
addze r0,r0
fsub $y2,$y2,$two64
slwi. r0,r0,4
fsub $y3,$y3,$two96
sub $inp,$inp,r0 # conditional rewind
fadd $h0lo,$h0lo,$y0 # accumulate input
fadd $h0hi,$h0hi,$y1
fadd $h2lo,$h2lo,$y2
fadd $h2hi,$h2hi,$y3
######################################### base 2^48 -> base 2^32
fadd $c1lo,$h1lo,$two64
$LWXLE $in0,0,$inp # modulo-scheduled input load
fadd $c1hi,$h1hi,$two64
$LWXLE $in1,$i1,$inp
fadd $c3lo,$h3lo,$two130
$LWXLE $in2,$i2,$inp
fadd $c3hi,$h3hi,$two130
$LWXLE $in3,$i3,$inp
fadd $c0lo,$h0lo,$two32
addi $inp,$inp,16
fadd $c0hi,$h0hi,$two32
fadd $c2lo,$h2lo,$two96
fadd $c2hi,$h2hi,$two96
fsub $c1lo,$c1lo,$two64
stw $in0,`$LOCALS+8*0+(4^$LITTLE_ENDIAN)`($sp) # fill "template"
fsub $c1hi,$c1hi,$two64
stw $in1,`$LOCALS+8*1+(4^$LITTLE_ENDIAN)`($sp)
fsub $c3lo,$c3lo,$two130
stw $in2,`$LOCALS+8*2+(4^$LITTLE_ENDIAN)`($sp)
fsub $c3hi,$c3hi,$two130
stw $in3,`$LOCALS+8*3+(4^$LITTLE_ENDIAN)`($sp)
fsub $c0lo,$c0lo,$two32
fsub $c0hi,$c0hi,$two32
fsub $c2lo,$c2lo,$two96
fsub $c2hi,$c2hi,$two96
fsub $h1lo,$h1lo,$c1lo
fsub $h1hi,$h1hi,$c1hi
fsub $h3lo,$h3lo,$c3lo
fsub $h3hi,$h3hi,$c3hi
fsub $h2lo,$h2lo,$c2lo
fsub $h2hi,$h2hi,$c2hi
fsub $h0lo,$h0lo,$c0lo
fsub $h0hi,$h0hi,$c0hi
fadd $h1lo,$h1lo,$c0lo
fadd $h1hi,$h1hi,$c0hi
fadd $h3lo,$h3lo,$c2lo
fadd $h3hi,$h3hi,$c2hi
fadd $h2lo,$h2lo,$c1lo
fadd $h2hi,$h2hi,$c1hi
fmadd $h0lo,$c3lo,$five_two130,$h0lo
fmadd $h0hi,$c3hi,$five_two130,$h0hi
fadd $x1,$h1lo,$h1hi
lfd $s1lo,8*12($ctx) # reload constants
fadd $x3,$h3lo,$h3hi
lfd $s1hi,8*13($ctx)
fadd $x2,$h2lo,$h2hi
lfd $r3lo,8*10($ctx)
fadd $x0,$h0lo,$h0hi
lfd $r3hi,8*11($ctx)
Lentry:
fmul $h0lo,$s3lo,$x1
fmul $h0hi,$s3hi,$x1
fmul $h2lo,$r1lo,$x1
fmul $h2hi,$r1hi,$x1
fmul $h1lo,$r0lo,$x1
fmul $h1hi,$r0hi,$x1
fmul $h3lo,$r2lo,$x1
fmul $h3hi,$r2hi,$x1
fmadd $h0lo,$s1lo,$x3,$h0lo
fmadd $h0hi,$s1hi,$x3,$h0hi
fmadd $h2lo,$s3lo,$x3,$h2lo
fmadd $h2hi,$s3hi,$x3,$h2hi
fmadd $h1lo,$s2lo,$x3,$h1lo
fmadd $h1hi,$s2hi,$x3,$h1hi
fmadd $h3lo,$r0lo,$x3,$h3lo
fmadd $h3hi,$r0hi,$x3,$h3hi
fmadd $h0lo,$s2lo,$x2,$h0lo
fmadd $h0hi,$s2hi,$x2,$h0hi
fmadd $h2lo,$r0lo,$x2,$h2lo
fmadd $h2hi,$r0hi,$x2,$h2hi
fmadd $h1lo,$s3lo,$x2,$h1lo
fmadd $h1hi,$s3hi,$x2,$h1hi
fmadd $h3lo,$r1lo,$x2,$h3lo
fmadd $h3hi,$r1hi,$x2,$h3hi
fmadd $h0lo,$r0lo,$x0,$h0lo
lfd $y0,`$LOCALS+8
|