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
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
| #!/usr/bin/env perl
| # Copyright 2017-2020 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/.
| # ====================================================================
| #
| # Keccak-1600 for PPC64.
| #
| # June 2017.
| #
| # This is straightforward KECCAK_1X_ALT implementation that works on
| # *any* PPC64. Then PowerISA 2.07 adds 2x64-bit vector rotate, and
| # it's possible to achieve performance better than below, but that is
| # naturally option only for POWER8 and successors...
| #
| ######################################################################
| # Numbers are cycles per processed byte.
| #
| # r=1088(*)
| #
| # PPC970/G5 14.6/+120%
| # POWER7 10.3/+100%
| # POWER8 11.5/+85%
| # POWER9 9.4/+45%
| #
| # (*) Corresponds to SHA3-256. Percentage after slash is improvement
| # over gcc-4.x-generated KECCAK_1X_ALT code. Newer compilers do
| # much better (but watch out for them generating code specific
| # to processor they execute on).
|
| $flavour = shift;
|
| if ($flavour =~ /64/) {
| $SIZE_T =8;
| $LRSAVE =2*$SIZE_T;
| $UCMP ="cmpld";
| $STU ="stdu";
| $POP ="ld";
| $PUSH ="std";
| } else { die "nonsense $flavour"; }
|
| $0 =~ m/(.*[\/\\])[^\/\\]+$/; $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: $!";
|
| $FRAME=24*$SIZE_T+6*$SIZE_T+32;
| $LOCALS=6*$SIZE_T;
| $TEMP=$LOCALS+6*$SIZE_T;
|
| my $sp ="r1";
|
| my @A = map([ "r$_", "r".($_+1), "r".($_+2), "r".($_+3), "r".($_+4) ],
| (7, 12, 17, 22, 27));
| $A[1][1] = "r6"; # r13 is reserved
|
| my @C = map("r$_", (0,3,4,5));
|
| my @rhotates = ([ 0, 1, 62, 28, 27 ],
| [ 36, 44, 6, 55, 20 ],
| [ 3, 10, 43, 25, 39 ],
| [ 41, 45, 15, 21, 8 ],
| [ 18, 2, 61, 56, 14 ]);
|
| $code.=<<___;
| .text
|
| .type KeccakF1600_int,\@function
| .align 5
| KeccakF1600_int:
| li r0,24
| mtctr r0
| b .Loop
| .align 4
| .Loop:
| xor $C[0],$A[0][0],$A[1][0] ; Theta
| std $A[0][4],`$TEMP+0`($sp)
| xor $C[1],$A[0][1],$A[1][1]
| std $A[1][4],`$TEMP+8`($sp)
| xor $C[2],$A[0][2],$A[1][2]
| std $A[2][4],`$TEMP+16`($sp)
| xor $C[3],$A[0][3],$A[1][3]
| std $A[3][4],`$TEMP+24`($sp)
| ___
| $C[4]=$A[0][4];
| $C[5]=$A[1][4];
| $C[6]=$A[2][4];
| $C[7]=$A[3][4];
| $code.=<<___;
| xor $C[4],$A[0][4],$A[1][4]
| xor $C[0],$C[0],$A[2][0]
| xor $C[1],$C[1],$A[2][1]
| xor $C[2],$C[2],$A[2][2]
| xor $C[3],$C[3],$A[2][3]
| xor $C[4],$C[4],$A[2][4]
| xor $C[0],$C[0],$A[3][0]
| xor $C[1],$C[1],$A[3][1]
| xor $C[2],$C[2],$A[3][2]
| xor $C[3],$C[3],$A[3][3]
| xor $C[4],$C[4],$A[3][4]
| xor $C[0],$C[0],$A[4][0]
| xor $C[2],$C[2],$A[4][2]
| xor $C[1],$C[1],$A[4][1]
| xor $C[3],$C[3],$A[4][3]
| rotldi $C[5],$C[2],1
| xor $C[4],$C[4],$A[4][4]
| rotldi $C[6],$C[3],1
| xor $C[5],$C[5],$C[0]
| rotldi $C[7],$C[4],1
|
| xor $A[0][1],$A[0][1],$C[5]
| xor $A[1][1],$A[1][1],$C[5]
| xor $A[2][1],$A[2][1],$C[5]
| xor $A[3][1],$A[3][1],$C[5]
| xor $A[4][1],$A[4][1],$C[5]
|
| rotldi $C[5],$C[0],1
| xor $C[6],$C[6],$C[1]
| xor $C[2],$C[2],$C[7]
| rotldi $C[7],$C[1],1
| xor $C[3],$C[3],$C[5]
| xor $C[4],$C[4],$C[7]
|
| xor $C[1], $A[0][2],$C[6] ;mr $C[1],$A[0][2]
| xor $A[1][2],$A[1][2],$C[6]
| xor $A[2][2],$A[2][2],$C[6]
| xor $A[3][2],$A[3][2],$C[6]
| xor $A[4][2],$A[4][2],$C[6]
|
| xor $A[0][0],$A[0][0],$C[4]
| xor $A[1][0],$A[1][0],$C[4]
| xor $A[2][0],$A[2][0],$C[4]
| xor $A[3][0],$A[3][0],$C[4]
| xor $A[4][0],$A[4][0],$C[4]
| ___
| $C[4]=undef;
| $C[5]=undef;
| $C[6]=undef;
| $C[7]=undef;
| $code.=<<___;
| ld $A[0][4],`$TEMP+0`($sp)
| xor $C[0], $A[0][3],$C[2] ;mr $C[0],$A[0][3]
| ld $A[1][4],`$TEMP+8`($sp)
| xor $A[1][3],$A[1][3],$C[2]
| ld $A[2][4],`$TEMP+16`($sp)
| xor $A[2][3],$A[2][3],$C[2]
| ld $A[3][4],`$TEMP+24`($sp)
| xor $A[3][3],$A[3][3],$C[2]
| xor $A[4][3],$A[4][3],$C[2]
|
| xor $C[2], $A[0][4],$C[3] ;mr $C[2],$A[0][4]
| xor $A[1][4],$A[1][4],$C[3]
| xor $A[2][4],$A[2][4],$C[3]
| xor $A[3][4],$A[3][4],$C[3]
| xor $A[4][4],$A[4][4],$C[3]
|
| mr $C[3],$A[0][1] ; Rho+Pi
| rotldi $A[0][1],$A[1][1],$rhotates[1][1]
| ;mr $C[1],$A[0][2]
| rotldi $A[0][2],$A[2][2],$rhotates[2][2]
| ;mr $C[0],$A[0][3]
| rotldi $A[0][3],$A[3][3],$rhotates[3][3]
| ;mr $C[2],$A[0][4]
| rotldi $A[0][4],$A[4][4],$rhotates[4][4]
|
| rotldi $A[1][1],$A[1][4],$rhotates[1][4]
| rotldi $A[2][2],$A[2][3],$rhotates[2][3]
| rotldi $A[3][3],$A[3][2],$rhotates[3][2]
| rotldi $A[4][4],$A[4][1],$rhotates[4][1]
|
| rotldi $A[1][4],$A[4][2],$rhotates[4][2]
| rotldi $A[2][3],$A[3][4],$rhotates[3][4]
| rotldi $A[3][2],$A[2][1],$rhotates[2][1]
| rotldi $A[4][1],$A[1][3],$rhotates[1][3]
|
| rotldi $A[4][2],$A[2][4],$rhotates[2][4]
| rotldi $A[3][4],$A[4][3],$rhotates[4][3]
| rotldi $A[2][1],$A[1][2],$rhotates[1][2]
| rotldi $A[1][3],$A[3][1],$rhotates[3][1]
|
| rotldi $A[2][4],$A[4][0],$rhotates[4][0]
| rotldi $A[4][3],$A[3][0],$rhotates[3][0]
| rotldi $A[1][2],$A[2][0],$rhotates[2][0]
| rotldi $A[3][1],$A[1][0],$rhotates[1][0]
|
| rotldi $A[1][0],$C[0],$rhotates[0][3]
| rotldi $A[2][0],$C[3],$rhotates[0][1]
| rotldi $A[3][0],$C[2],$rhotates[0][4]
| rotldi $A[4][0],$C[1],$rhotates[0][2]
|
| andc $C[0],$A[0][2],$A[0][1] ; Chi+Iota
| andc $C[1],$A[0][3],$A[0][2]
| andc $C[2],$A[0][0],$A[0][4]
| andc $C[3],$A[0][1],$A[0][0]
| xor $A[0][0],$A[0][0],$C[0]
| andc $C[0],$A[0][4],$A[0][3]
| xor $A[0][1],$A[0][1],$C[1]
| ld $C[1],`$LOCALS+4*$SIZE_T`($sp)
| xor $A[0][3],$A[0][3],$C[2]
| xor $A[0][4],$A[0][4],$C[3]
| xor $A[0][2],$A[0][2],$C[0]
| ldu $C[3],8($C[1]) ; Iota[i++]
|
| andc $C[0],$A[1][2],$A[1][1]
| std $C[1],`$LOCALS+4*$SIZE_T`($sp)
| andc $C[1],$A[1][3],$A[1][2]
| andc $C[2],$A[1][0],$A[1][4]
| xor $A[0][0],$A[0][0],$C[3] ; A[0][0] ^= Iota
| andc $C[3],$A[1][1],$A[1][0]
| xor $A[1][0],$A[1][0],$C[0]
| andc $C[0],$A[1][4],$A[1][3]
| xor $A[1][1],$A[1][1],$C[1]
| xor $A[1][3],$A[1][3],$C[2]
| xor $A[1][4],$A[1][4],$C[3]
| xor $A[1][2],$A[1][2],$C[0]
|
| andc $C[0],$A[2][2],$A[2][1]
| andc $C[1],$A[2][3],$A[2][2]
| andc $C[2],$A[2][0],$A[2][4]
| andc $C[3],$A[2][1],$A[2][0]
| xor $A[2][0],$A[2][0],$C[0]
| andc $C[0],$A[2][4],$A[2][3]
| xor $A[2][1],$A[2][1],$C[1]
| xor $A[2][3],$A[2][3],$C[2]
| xor $A[2][4],$A[2][4],$C[3]
| xor $A[2][2],$A[2][2],$C[0]
|
| andc $C[0],$A[3][2],$A[3][1]
| andc $C[1],$A[3][3],$A[3][2]
| andc $C[2],$A[3][0],$A[3][4]
| andc $C[3],$A[3][1],$A[3][0]
| xor $A[3][0],$A[3][0],$C[0]
| andc $C[0],$A[3][4],$A[3][3]
| xor $A[3][1],$A[3][1],$C[1]
| xor $A[3][3],$A[3][3],$C[2]
| xor $A[3][4],$A[3][4],$C[3]
| xor $A[3][2],$A[3][2],$C[0]
|
| andc $C[0],$A[4][2],$A[4][1]
| andc $C[1],$A[4][3],$A[4][2]
| andc $C[2],$A[4][0],$A[4][4]
| andc $C[3],$A[4][1],$A[4][0]
| xor $A[4][0],$A[4][0],$C[0]
| andc $C[0],$A[4][4],$A[4][3]
| xor $A[4][1],$A[4][1],$C[1]
| xor $A[4][3],$A[4][3],$C[2]
| xor $A[4][4],$A[4][4],$C[3]
| xor $A[4][2],$A[4][2],$C[0]
|
| bdnz .Loop
|
| blr
| .long 0
| .byte 0,12,0x14,0,0,0,0,0
| .size KeccakF1600_int,.-KeccakF1600_int
|
| .type KeccakF1600,\@function
| .align 5
| KeccakF1600:
| $STU $sp,-$FRAME($sp)
| mflr r0
| $PUSH r14,`$FRAME-$SIZE_T*18`($sp)
| $PUSH r15,`$FRAME-$SIZE_T*17`($sp)
| $PUSH r16,`$FRAME-$SIZE_T*16`($sp)
| $PUSH r17,`$FRAME-$SIZE_T*15`($sp)
| $PUSH r18,`$FRAME-$SIZE_T*14`($sp)
| $PUSH r19,`$FRAME-$SIZE_T*13`($sp)
| $PUSH r20,`$FRAME-$SIZE_T*12`($sp)
| $PUSH r21,`$FRAME-$SIZE_T*11`($sp)
| $PUSH r22,`$FRAME-$SIZE_T*10`($sp)
| $PUSH r23,`$FRAME-$SIZE_T*9`($sp)
| $PUSH r24,`$FRAME-$SIZE_T*8`($sp)
| $PUSH r25,`$FRAME-$SIZE_T*7`($sp)
| $PUSH r26,`$FRAME-$SIZE_T*6`($sp)
| $PUSH r27,`$FRAME-$SIZE_T*5`($sp)
| $PUSH r28,`$FRAME-$SIZE_T*4`($sp)
| $PUSH r29,`$FRAME-$SIZE_T*3`($sp)
| $PUSH r30,`$FRAME-$SIZE_T*2`($sp)
| $PUSH r31,`$FRAME-$SIZE_T*1`($sp)
| $PUSH r0,`$FRAME+$LRSAVE`($sp)
|
| bl PICmeup
| subi r12,r12,8 ; prepare for ldu
|
| $PUSH r3,`$LOCALS+0*$SIZE_T`($sp)
| ;$PUSH r4,`$LOCALS+1*$SIZE_T`($sp)
| ;$PUSH r5,`$LOCALS+2*$SIZE_T`($sp)
| ;$PUSH r6,`$LOCALS+3*$SIZE_T`($sp)
| $PUSH r12,`$LOCALS+4*$SIZE_T`($sp)
|
| ld $A[0][0],`8*0`(r3) ; load A[5][5]
| ld $A[0][1],`8*1`(r3)
| ld $A[0][2],`8*2`(r3)
| ld $A[0][3],`8*3`(r3)
| ld $A[0][4],`8*4`(r3)
| ld $A[1][0],`8*5`(r3)
| ld $A[1][1],`8*6`(r3)
| ld $A[1][2],`8*7`(r3)
| ld $A[1][3],`8*8`(r3)
| ld $A[1][4],`8*9`(r3)
| ld $A[2][0],`8*10`(r3)
| ld $A[2][1],`8*11`(r3)
| ld $A[2][2],`8*12`(r3)
| ld $A[2][3],`8*13`(r3)
| ld $A[2][4],`8*14`(r3)
| ld $A[3][0],`8*15`(r3)
| ld $A[3][1],`8*16`(r3)
| ld $A[3][2],`8*17`(r3)
| ld $A[3][3],`8*18`(r3)
| ld $A[3][4],`8*19`(r3)
| ld $A[4][0],`8*20`(r3)
| ld $A[4][1],`8*21`(r3)
| ld $A[4][2],`8*22`(r3)
| ld $A[4][3],`8*23`(r3)
| ld $A[4][4],`8*24`(r3)
|
| bl KeccakF1600_int
|
| $POP r3,`$LOCALS+0*$SIZE_T`($sp)
| std $A[0][0],`8*0`(r3) ; return A[5][5]
| std $A[0][1],`8*1`(r3)
| std $A[0][2],`8*2`(r3)
| std $A[0][3],`8*3`(r3)
| std $A[0][4],`8*4`(r3)
| std $A[1][0],`8*5`(r3)
| std $A[1][1],`8*6`(r3)
| std $A[1][2],`8*7`(r3)
| std $A[1][3],`8*8`(r3)
| std $A[1][4],`8*9`(r3)
| std $A[2][0],`8*10`(r3)
| std $A[2][1],`8*11`(r3)
| std $A[2][2],`8*12`(r3)
| std $A[2][3],`8*13`(r3)
| std $A[2][4],`8*14`(r3)
| std $A[3][0],`8*15`(r3)
| std $A[3][1],`8*16`(r3)
| std $A[3][2],`8*17`(r3)
| std $A[3][3],`8*18`(r3)
| std $A[3][4],`8*19`(r3)
| std $A[4][0],`8*20`(r3)
| std $A[4][1],`8*21`(r3)
| std $A[4][2],`8*22`(r3)
| std $A[4][3],`8*23`(r3)
| std $A[4][4],`8*24`(r3)
|
| $POP r0,`$FRAME+$LRSAVE`($sp)
| $POP r14,`$FRAME-$SIZE_T*18`($sp)
| $POP r15,`$FRAME-$SIZE_T*17`($sp)
| $POP r16,`$FRAME-$SIZE_T*16`($sp)
| $POP r17,`$FRAME-$SIZE_T*15`($sp)
| $POP r18,`$FRAME-$SIZE_T*14`($sp)
| $POP r19,`$FRAME-$SIZE_T*13`($sp)
| $POP r20,`$FRAME-$SIZE_T*12`($sp)
| $POP r21,`$FRAME-$SIZE_T*11`($sp)
| $POP r22,`$FRAME-$SIZE_T*10`($sp)
| $POP r23,`$FRAME-$SIZE_T*9`($sp)
| $POP r24,`$FRAME-$SIZE_T*8`($sp)
| $POP r25,`$FRAME-$SIZE_T*7`($sp)
| $POP r26,`$FRAME-$SIZE_T*6`($sp)
| $POP r27,`$FRAME-$SIZE_T*5`($sp)
| $POP r28,`$FRAME-$SIZE_T*4`($sp)
| $POP r29,`$FRAME-$SIZE_T*3`($sp)
| $POP r30,`$FRAME-$SIZE_T*2`($sp)
| $POP r31,`$FRAME-$SIZE_T*1`($sp)
| mtlr r0
| addi $sp,$sp,$FRAME
| blr
| .long 0
| .byte 0,12,4,1,0x80,18,1,0
| .long 0
| .size KeccakF1600,.-KeccakF1600
|
| .type dword_le_load,\@function
| .align 5
| dword_le_load:
| lbzu r0,1(r3)
| lbzu r4,1(r3)
| lbzu r5,1(r3)
| insrdi r0,r4,8,48
| lbzu r4,1(r3)
| insrdi r0,r5,8,40
| lbzu r5,1(r3)
| insrdi r0,r4,8,32
| lbzu r4,1(r3)
| insrdi r0,r5,8,24
| lbzu r5,1(r3)
| insrdi r0,r4,8,16
| lbzu r4,1(r3)
| insrdi r0,r5,8,8
| insrdi r0,r4,8,0
| blr
| .long 0
| .byte 0,12,0x14,0,0,0,1,0
| .long 0
| .size dword_le_load,.-dword_le_load
|
| .globl SHA3_absorb
| .type SHA3_absorb,\@function
| .align 5
| SHA3_absorb:
| $STU $sp,-$FRAME($sp)
| mflr r0
| $PUSH r14,`$FRAME-$SIZE_T*18`($sp)
| $PUSH r15,`$FRAME-$SIZE_T*17`($sp)
| $PUSH r16,`$FRAME-$SIZE_T*16`($sp)
| $PUSH r17,`$FRAME-$SIZE_T*15`($sp)
| $PUSH r18,`$FRAME-$SIZE_T*14`($sp)
| $PUSH r19,`$FRAME-$SIZE_T*13`($sp)
| $PUSH r20,`$FRAME-$SIZE_T*12`($sp)
| $PUSH r21,`$FRAME-$SIZE_T*11`($sp)
| $PUSH r22,`$FRAME-$SIZE_T*10`($sp)
| $PUSH r23,`$FRAME-$SIZE_T*9`($sp)
| $PUSH r24,`$FRAME-$SIZE_T*8`($sp)
| $PUSH r25,`$FRAME-$SIZE_T*7`($sp)
| $PUSH r26,`$FRAME-$SIZE_T*6`($sp)
| $PUSH r27,`$FRAME-$SIZE_T*5`($sp)
| $PUSH r28,`$FRAME-$SIZE_T*4`($sp)
| $PUSH r29,`$FRAME-$SIZE_T*3`($sp)
| $PUSH r30,`$FRAME-$SIZE_T*2`($sp)
| $PUSH r31,`$FRAME-$SIZE_T*1`($sp)
| $PUSH r0,`$FRAME+$LRSAVE`($sp)
|
| bl PICmeup
| subi r4,r4,1 ; prepare for lbzu
| subi r12,r12,8 ; prepare for ldu
|
| $PUSH r3,`$LOCALS+0*$SIZE_T`($sp) ; save A[][]
| $PUSH r4,`$LOCALS+1*$SIZE_T`($sp) ; save inp
| $PUSH r5,`$LOCALS+2*$SIZE_T`($sp) ; save len
| $PUSH r6,`$LOCALS+3*$SIZE_T`($sp) ; save bsz
| mr r0,r6
| $PUSH r12,`$LOCALS+4*$SIZE_T`($sp)
|
| ld $A[0][0],`8*0`(r3) ; load A[5][5]
| ld $A[0][1],`8*1`(r3)
| ld $A[0][2],`8*2`(r3)
| ld $A[0][3],`8*3`(r3)
| ld $A[0][4],`8*4`(r3)
| ld $A[1][0],`8*5`(r3)
| ld $A[1][1],`8*6`(r3)
| ld $A[1][2],`8*7`(r3)
| ld $A[1][3],`8*8`(r3)
| ld $A[1][4],`8*9`(r3)
| ld $A[2][0],`8*10`(r3)
| ld $A[2][1],`8*11`(r3)
| ld $A[2][2],`8*12`(r3)
| ld $A[2][3],`8*13`(r3)
| ld $A[2][4],`8*14`(r3)
| ld $A[3][0],`8*15`(r3)
| ld $A[3][1],`8*16`(r3)
| ld $A[3][2],`8*17`(r3)
| ld $A[3][3],`8*18`(r3)
| ld $A[3][4],`8*19`(r3)
| ld $A[4][0],`8*20`(r3)
| ld $A[4][1],`8*21`(r3)
| ld $A[4][2],`8*22`(r3)
| ld $A[4][3],`8*23`(r3)
| ld $A[4][4],`8*24`(r3)
|
| mr r3,r4
| mr r4,r5
| mr r5,r0
|
| b .Loop_absorb
|
| .align 4
| .Loop_absorb:
| $UCMP r4,r5 ; len < bsz?
| blt .Labsorbed
|
| sub r4,r4,r5 ; len -= bsz
| srwi r5,r5,3
| $PUSH r4,`$LOCALS+2*$SIZE_T`($sp) ; save len
| mtctr r5
| bl dword_le_load ; *inp++
| xor $A[0][0],$A[0][0],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[0][1],$A[0][1],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[0][2],$A[0][2],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[0][3],$A[0][3],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[0][4],$A[0][4],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[1][0],$A[1][0],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[1][1],$A[1][1],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[1][2],$A[1][2],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[1][3],$A[1][3],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[1][4],$A[1][4],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[2][0],$A[2][0],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[2][1],$A[2][1],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[2][2],$A[2][2],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[2][3],$A[2][3],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[2][4],$A[2][4],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[3][0],$A[3][0],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[3][1],$A[3][1],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[3][2],$A[3][2],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[3][3],$A[3][3],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[3][4],$A[3][4],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[4][0],$A[4][0],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[4][1],$A[4][1],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[4][2],$A[4][2],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[4][3],$A[4][3],r0
| bdz .Lprocess_block
| bl dword_le_load ; *inp++
| xor $A[4][4],$A[4][4],r0
|
| .Lprocess_block:
| $PUSH r3,`$LOCALS+1*$SIZE_T`($sp) ; save inp
|
| bl KeccakF1600_int
|
| $POP r0,`$LOCALS+4*$SIZE_T`($sp) ; pull iotas[24]
| $POP r5,`$LOCALS+3*$SIZE_T`($sp) ; restore bsz
| $POP r4,`$LOCALS+2*$SIZE_T`($sp) ; restore len
| $POP r3,`$LOCALS+1*$SIZE_T`($sp) ; restore inp
| addic r0,r0,`-8*24` ; rewind iotas
| $PUSH r0,`$LOCALS+4*$SIZE_T`($sp)
|
| b .Loop_absorb
|
| .align 4
| .Labsorbed:
| $POP r3,`$LOCALS+0*$SIZE_T`($sp)
| std $A[0][0],`8*0`(r3) ; return A[5][5]
| std $A[0][1],`8*1`(r3)
| std $A[0][2],`8*2`(r3)
| std $A[0][3],`8*3`(r3)
| std $A[0][4],`8*4`(r3)
| std $A[1][0],`8*5`(r3)
| std $A[1][1],`8*6`(r3)
| std $A[1][2],`8*7`(r3)
| std $A[1][3],`8*8`(r3)
| std $A[1][4],`8*9`(r3)
| std $A[2][0],`8*10`(r3)
| std $A[2][1],`8*11`(r3)
| std $A[2][2],`8*12`(r3)
| std $A[2][3],`8*13`(r3)
| std $A[2][4],`8*14`(r3)
| std $A[3][0],`8*15`(r3)
| std $A[3][1],`8*16`(r3)
| std $A[3][2],`8*17`(r3)
| std $A[3][3],`8*18`(r3)
| std $A[3][4],`8*19`(r3)
| std $A[4][0],`8*20`(r3)
| std $A[4][1],`8*21`(r3)
| std $A[4][2],`8*22`(r3)
| std $A[4][3],`8*23`(r3)
| std $A[4][4],`8*24`(r3)
|
| mr r3,r4 ; return value
| $POP r0,`$FRAME+$LRSAVE`($sp)
| $POP r14,`$FRAME-$SIZE_T*18`($sp)
| $POP r15,`$FRAME-$SIZE_T*17`($sp)
| $POP r16,`$FRAME-$SIZE_T*16`($sp)
| $POP r17,`$FRAME-$SIZE_T*15`($sp)
| $POP r18,`$FRAME-$SIZE_T*14`($sp)
| $POP r19,`$FRAME-$SIZE_T*13`($sp)
| $POP r20,`$FRAME-$SIZE_T*12`($sp)
| $POP r21,`$FRAME-$SIZE_T*11`($sp)
| $POP r22,`$FRAME-$SIZE_T*10`($sp)
| $POP r23,`$FRAME-$SIZE_T*9`($sp)
| $POP r24,`$FRAME-$SIZE_T*8`($sp)
| $POP r25,`$FRAME-$SIZE_T*7`($sp)
| $POP r26,`$FRAME-$SIZE_T*6`($sp)
| $POP r27,`$FRAME-$SIZE_T*5`($sp)
| $POP r28,`$FRAME-$SIZE_T*4`($sp)
| $POP r29,`$FRAME-$SIZE_T*3`($sp)
| $POP r30,`$FRAME-$SIZE_T*2`($sp)
| $POP r31,`$FRAME-$SIZE_T*1`($sp)
| mtlr r0
| addi $sp,$sp,$FRAME
| blr
| .long 0
| .byte 0,12,4,1,0x80,18,4,0
| .long 0
| .size SHA3_absorb,.-SHA3_absorb
| ___
| {
| my ($A_flat,$out,$len,$bsz) = map("r$_",(28..31));
| $code.=<<___;
| .globl SHA3_squeeze
| .type SHA3_squeeze,\@function
| .align 5
| SHA3_squeeze:
| $STU $sp,`-10*$SIZE_T`($sp)
| mflr r0
| $PUSH r28,`6*$SIZE_T`($sp)
| $PUSH r29,`7*$SIZE_T`($sp)
| $PUSH r30,`8*$SIZE_T`($sp)
| $PUSH r31,`9*$SIZE_T`($sp)
| $PUSH r0,`10*$SIZE_T+$LRSAVE`($sp)
|
| mr $A_flat,r3
| subi r3,r3,8 ; prepare for ldu
| subi $out,r4,1 ; prepare for stbu
| mr $len,r5
| mr $bsz,r6
| b .Loop_squeeze
|
| .align 4
| .Loop_squeeze:
| ldu r0,8(r3)
| ${UCMP}i $len,8
| blt .Lsqueeze_tail
|
| stbu r0,1($out)
| srdi r0,r0,8
| stbu r0,1($out)
| srdi r0,r0,8
| stbu r0,1($out)
| srdi r0,r0,8
| stbu r0,1($out)
| srdi r0,r0,8
| stbu r0,1($out)
| srdi r0,r0,8
| stbu r0,1($out)
| srdi r0,r0,8
| stbu r0,1($out)
| srdi r0,r0,8
| stbu r0,1($out)
|
| subic. $len,$len,8
| beq .Lsqueeze_done
|
| subic. r6,r6,8
| bgt .Loop_squeeze
|
| mr r3,$A_flat
| bl KeccakF1600
| subi r3,$A_flat,8 ; prepare for ldu
| mr r6,$bsz
| b .Loop_squeeze
|
| .align 4
| .Lsqueeze_tail:
| mtctr $len
| .Loop_tail:
| stbu r0,1($out)
| srdi r0,r0,8
| bdnz .Loop_tail
|
| .Lsqueeze_done:
| $POP r0,`10*$SIZE_T+$LRSAVE`($sp)
| $POP r28,`6*$SIZE_T`($sp)
| $POP r29,`7*$SIZE_T`($sp)
| $POP r30,`8*$SIZE_T`($sp)
| $POP r31,`9*$SIZE_T`($sp)
| mtlr r0
| addi $sp,$sp,`10*$SIZE_T`
| blr
| .long 0
| .byte 0,12,4,1,0x80,4,4,0
| .long 0
| .size SHA3_squeeze,.-SHA3_squeeze
| ___
| }
|
| # Ugly hack here, because PPC assembler syntax seem to vary too
| # much from platforms to platform...
| $code.=<<___;
| .align 6
| PICmeup:
| mflr r0
| bcl 20,31,\$+4
| mflr r12 ; vvvvvv "distance" between . and 1st data entry
| addi r12,r12,`64-8`
| mtlr r0
| blr
| .long 0
| .byte 0,12,0x14,0,0,0,0,0
| .space `64-9*4`
| .type iotas,\@object
| iotas:
| .quad 0x0000000000000001
| .quad 0x0000000000008082
| .quad 0x800000000000808a
| .quad 0x8000000080008000
| .quad 0x000000000000808b
| .quad 0x0000000080000001
| .quad 0x8000000080008081
| .quad 0x8000000000008009
| .quad 0x000000000000008a
| .quad 0x0000000000000088
| .quad 0x0000000080008009
| .quad 0x000000008000000a
| .quad 0x000000008000808b
| .quad 0x800000000000008b
| .quad 0x8000000000008089
| .quad 0x8000000000008003
| .quad 0x8000000000008002
| .quad 0x8000000000000080
| .quad 0x000000000000800a
| .quad 0x800000008000000a
| .quad 0x8000000080008081
| .quad 0x8000000000008080
| .quad 0x0000000080000001
| .quad 0x8000000080008008
| .size iotas,.-iotas
| .asciz "Keccak-1600 absorb and squeeze for PPC64, CRYPTOGAMS by <appro\@openssl.org>"
| ___
|
| $code =~ s/\`([^\`]*)\`/eval $1/gem;
| print $code;
| close STDOUT or die "error closing STDOUT: $!";
|
|
