Debian Codegraph @

crypto_hashblocks/sha512/avx2/inner.S

raw
1 #include "crypto_asm_hidden.h"
2 // linker define inner
3
4 # qhasm: int64 input_0
5
6 # qhasm: int64 input_1
7
8 # qhasm: int64 input_2
9
10 # qhasm: int64 input_3
11
12 # qhasm: int64 input_4
13
14 # qhasm: int64 input_5
15
16 # qhasm: stack64 input_6
17
18 # qhasm: stack64 input_7
19
20 # qhasm: int64 caller_r11
21
22 # qhasm: int64 caller_r12
23
24 # qhasm: int64 caller_r13
25
26 # qhasm: int64 caller_r14
27
28 # qhasm: int64 caller_r15
29
30 # qhasm: int64 caller_rbx
31
32 # qhasm: int64 caller_rbp
33
34 # qhasm: stack64 r11_stack
35
36 # qhasm: stack64 r12_stack
37
38 # qhasm: stack64 r13_stack
39
40 # qhasm: stack64 r14_stack
41
42 # qhasm: stack64 r15_stack
43
44 # qhasm: stack64 rbx_stack
45
46 # qhasm: stack64 rbp_stack
47
48 # qhasm: int64 statebytes
49
50 # qhasm: stack64 statebytes_stack
51
52 # qhasm: int64 in
53
54 # qhasm: stack64 in_stack
55
56 # qhasm: int64 inlen
57
58 # qhasm: stack64 inlen_stack
59
60 # qhasm: int64 constants
61
62 # qhasm: stack64 constants_stack
63
64 # qhasm: int64 r0
65
66 # qhasm: int64 r1
67
68 # qhasm: int64 r2
69
70 # qhasm: int64 r3
71
72 # qhasm: int64 r4
73
74 # qhasm: int64 r5
75
76 # qhasm: int64 r6
77
78 # qhasm: int64 r7
79
80 # qhasm: int64 i
81
82 # qhasm: stack256 state0123
83
84 # qhasm: stack256 state4567
85
86 # qhasm: reg256 X0
87
88 # qhasm: reg256 X4
89
90 # qhasm: reg256 X8
91
92 # qhasm: reg256 X12
93
94 # qhasm: reg256 X1
95
96 # qhasm: reg256 X5
97
98 # qhasm: reg256 X9
99
100 # qhasm: reg256 X13
101
102 # qhasm: reg256 bigendian64
103
104 # qhasm: reg256 D0
105
106 # qhasm: reg256 D4
107
108 # qhasm: reg256 D8
109
110 # qhasm: reg256 D12
111
112 # qhasm: reg256 W0
113
114 # qhasm: reg256 W2
115
116 # qhasm: reg256 W4
117
118 # qhasm: reg256 W6
119
120 # qhasm: reg256 W8
121
122 # qhasm: reg256 W10
123
124 # qhasm: reg256 W12
125
126 # qhasm: reg256 W14
127
128 # qhasm: stack1280 w
129
130 # qhasm: stack256 wc0123
131
132 # qhasm: stack256 wc4567
133
134 # qhasm: stack256 wc891011
135
136 # qhasm: stack256 wc12131415
137
138 # qhasm: int64 r0andr1
139
140 # qhasm: int64 r2andr3
141
142 # qhasm: int64 r4andr5
143
144 # qhasm: int64 r6andr7
145
146 # qhasm: int64 ch0
147
148 # qhasm: int64 ch1
149
150 # qhasm: int64 ch2
151
152 # qhasm: int64 ch3
153
154 # qhasm: int64 ch4
155
156 # qhasm: int64 ch5
157
158 # qhasm: int64 ch6
159
160 # qhasm: int64 ch7
161
162 # qhasm: int64 maj0
163
164 # qhasm: int64 maj1
165
166 # qhasm: int64 maj2
167
168 # qhasm: int64 maj3
169
170 # qhasm: int64 maj4
171
172 # qhasm: int64 maj5
173
174 # qhasm: int64 maj6
175
176 # qhasm: int64 maj7
177
178 # qhasm: int64 r0Sigma0
179
180 # qhasm: int64 r1Sigma0
181
182 # qhasm: int64 r2Sigma0
183
184 # qhasm: int64 r3Sigma0
185
186 # qhasm: int64 r4Sigma0
187
188 # qhasm: int64 r5Sigma0
189
190 # qhasm: int64 r6Sigma0
191
192 # qhasm: int64 r7Sigma0
193
194 # qhasm: int64 r0Sigma1
195
196 # qhasm: int64 r1Sigma1
197
198 # qhasm: int64 r2Sigma1
199
200 # qhasm: int64 r3Sigma1
201
202 # qhasm: int64 r4Sigma1
203
204 # qhasm: int64 r5Sigma1
205
206 # qhasm: int64 r6Sigma1
207
208 # qhasm: int64 r7Sigma1
209
210 # qhasm: int64 r018
211
212 # qhasm: int64 r118
213
214 # qhasm: int64 r218
215
216 # qhasm: int64 r318
217
218 # qhasm: int64 r418
219
220 # qhasm: int64 r518
221
222 # qhasm: int64 r618
223
224 # qhasm: int64 r718
225
226 # qhasm: int64 r041
227
228 # qhasm: int64 r141
229
230 # qhasm: int64 r241
231
232 # qhasm: int64 r341
233
234 # qhasm: int64 r441
235
236 # qhasm: int64 r541
237
238 # qhasm: int64 r641
239
240 # qhasm: int64 r741
241
242 # qhasm: int64 r034
243
244 # qhasm: int64 r134
245
246 # qhasm: int64 r234
247
248 # qhasm: int64 r334
249
250 # qhasm: int64 r434
251
252 # qhasm: int64 r534
253
254 # qhasm: int64 r634
255
256 # qhasm: int64 r734
257
258 # qhasm: int64 r039
259
260 # qhasm: int64 r139
261
262 # qhasm: int64 r239
263
264 # qhasm: int64 r339
265
266 # qhasm: int64 r439
267
268 # qhasm: int64 r539
269
270 # qhasm: int64 r639
271
272 # qhasm: int64 r739
273
274 # qhasm: reg256 X1right1
275
276 # qhasm: reg256 X1left63
277
278 # qhasm: reg256 X1right8
279
280 # qhasm: reg256 X1left56
281
282 # qhasm: reg256 X1right7
283
284 # qhasm: reg256 X1sigma0
285
286 # qhasm: reg256 X5right1
287
288 # qhasm: reg256 X5left63
289
290 # qhasm: reg256 X5right8
291
292 # qhasm: reg256 X5left56
293
294 # qhasm: reg256 X5right7
295
296 # qhasm: reg256 X5sigma0
297
298 # qhasm: reg256 X9right1
299
300 # qhasm: reg256 X9left63
301
302 # qhasm: reg256 X9right8
303
304 # qhasm: reg256 X9left56
305
306 # qhasm: reg256 X9right7
307
308 # qhasm: reg256 X9sigma0
309
310 # qhasm: reg256 X13right1
311
312 # qhasm: reg256 X13left63
313
314 # qhasm: reg256 X13right8
315
316 # qhasm: reg256 X13left56
317
318 # qhasm: reg256 X13right7
319
320 # qhasm: reg256 X13sigma0
321
322 # qhasm: reg256 W0right19
323
324 # qhasm: reg256 W0right61
325
326 # qhasm: reg256 W0right6
327
328 # qhasm: reg256 W0left45
329
330 # qhasm: reg256 W0left3
331
332 # qhasm: reg256 W0sigma1
333
334 # qhasm: reg256 W2right19
335
336 # qhasm: reg256 W2right61
337
338 # qhasm: reg256 W2right6
339
340 # qhasm: reg256 W2left45
341
342 # qhasm: reg256 W2left3
343
344 # qhasm: reg256 W2sigma1
345
346 # qhasm: reg256 W4right19
347
348 # qhasm: reg256 W4right61
349
350 # qhasm: reg256 W4right6
351
352 # qhasm: reg256 W4left45
353
354 # qhasm: reg256 W4left3
355
356 # qhasm: reg256 W4sigma1
357
358 # qhasm: reg256 W6right19
359
360 # qhasm: reg256 W6right61
361
362 # qhasm: reg256 W6right6
363
364 # qhasm: reg256 W6left45
365
366 # qhasm: reg256 W6left3
367
368 # qhasm: reg256 W6sigma1
369
370 # qhasm: reg256 W8right19
371
372 # qhasm: reg256 W8right61
373
374 # qhasm: reg256 W8right6
375
376 # qhasm: reg256 W8left45
377
378 # qhasm: reg256 W8left3
379
380 # qhasm: reg256 W8sigma1
381
382 # qhasm: reg256 W10right19
383
384 # qhasm: reg256 W10right61
385
386 # qhasm: reg256 W10right6
387
388 # qhasm: reg256 W10left45
389
390 # qhasm: reg256 W10left3
391
392 # qhasm: reg256 W10sigma1
393
394 # qhasm: reg256 W12right19
395
396 # qhasm: reg256 W12right61
397
398 # qhasm: reg256 W12right6
399
400 # qhasm: reg256 W12left45
401
402 # qhasm: reg256 W12left3
403
404 # qhasm: reg256 W12sigma1
405
406 # qhasm: reg256 W14right19
407
408 # qhasm: reg256 W14right61
409
410 # qhasm: reg256 W14right6
411
412 # qhasm: reg256 W14left45
413
414 # qhasm: reg256 W14left3
415
416 # qhasm: reg256 W14sigma1
417
418 # qhasm: enter CRYPTO_SHARED_NAMESPACE(inner)
419 .p2align 7
420 ASM_HIDDEN _CRYPTO_SHARED_NAMESPACE(inner)
421 .global _CRYPTO_SHARED_NAMESPACE(inner)
422 ASM_HIDDEN CRYPTO_SHARED_NAMESPACE(inner)
423 .global CRYPTO_SHARED_NAMESPACE(inner)
424 _CRYPTO_SHARED_NAMESPACE(inner):
425 CRYPTO_SHARED_NAMESPACE(inner):
426 mov %rsp,%r11
427 and $511,%r11
428 add $416,%r11
429 sub %r11,%rsp
430
431 # qhasm: constants = input_3
432 # asm 1: mov <input_3=int64#4,>constants=int64#5
433 # asm 2: mov <input_3=%rcx,>constants=%r8
434 mov %rcx,%r8
435
436 # qhasm: bigendian64 = mem256[input_3+640]
437 # asm 1: vmovupd 640(<input_3=int64#4),>bigendian64=reg256#1
438 # asm 2: vmovupd 640(<input_3=%rcx),>bigendian64=%ymm0
439 vmovupd 640(%rcx),%ymm0
440
441 # qhasm: X0 = mem256[input_0+0]
442 # asm 1: vmovupd 0(<input_0=int64#1),>X0=reg256#2
443 # asm 2: vmovupd 0(<input_0=%rdi),>X0=%ymm1
444 vmovupd 0(%rdi),%ymm1
445
446 # qhasm: statebytes = input_0
447 # asm 1: mov <input_0=int64#1,>statebytes=int64#4
448 # asm 2: mov <input_0=%rdi,>statebytes=%rcx
449 mov %rdi,%rcx
450
451 # qhasm: X4 = mem256[input_0+32]
452 # asm 1: vmovupd 32(<input_0=int64#1),>X4=reg256#3
453 # asm 2: vmovupd 32(<input_0=%rdi),>X4=%ymm2
454 vmovupd 32(%rdi),%ymm2
455
456 # qhasm: 2x 16x X0 = X0[bigendian64]
457 # asm 1: vpshufb <bigendian64=reg256#1,<X0=reg256#2,>X0=reg256#2
458 # asm 2: vpshufb <bigendian64=%ymm0,<X0=%ymm1,>X0=%ymm1
459 vpshufb %ymm0,%ymm1,%ymm1
460
461 # qhasm: 2x 16x X4 = X4[bigendian64]
462 # asm 1: vpshufb <bigendian64=reg256#1,<X4=reg256#3,>X4=reg256#3
463 # asm 2: vpshufb <bigendian64=%ymm0,<X4=%ymm2,>X4=%ymm2
464 vpshufb %ymm0,%ymm2,%ymm2
465
466 # qhasm: state0123 = X0
467 # asm 1: vmovapd <X0=reg256#2,>state0123=stack256#1
468 # asm 2: vmovapd <X0=%ymm1,>state0123=0(%rsp)
469 vmovapd %ymm1,0(%rsp)
470
471 # qhasm: r11_stack = caller_r11
472 # asm 1: movq <caller_r11=int64#9,>r11_stack=stack64#1
473 # asm 2: movq <caller_r11=%r11,>r11_stack=320(%rsp)
474 movq %r11,320(%rsp)
475
476 # qhasm: state4567 = X4
477 # asm 1: vmovapd <X4=reg256#3,>state4567=stack256#2
478 # asm 2: vmovapd <X4=%ymm2,>state4567=32(%rsp)
479 vmovapd %ymm2,32(%rsp)
480
481 # qhasm: r13_stack = caller_r13
482 # asm 1: movq <caller_r13=int64#11,>r13_stack=stack64#2
483 # asm 2: movq <caller_r13=%r13,>r13_stack=328(%rsp)
484 movq %r13,328(%rsp)
485
486 # qhasm: r12_stack = caller_r12
487 # asm 1: movq <caller_r12=int64#10,>r12_stack=stack64#3
488 # asm 2: movq <caller_r12=%r12,>r12_stack=336(%rsp)
489 movq %r12,336(%rsp)
490
491 # qhasm: r14_stack = caller_r14
492 # asm 1: movq <caller_r14=int64#12,>r14_stack=stack64#4
493 # asm 2: movq <caller_r14=%r14,>r14_stack=344(%rsp)
494 movq %r14,344(%rsp)
495
496 # qhasm: rbx_stack = caller_rbx
497 # asm 1: movq <caller_rbx=int64#14,>rbx_stack=stack64#5
498 # asm 2: movq <caller_rbx=%rbx,>rbx_stack=352(%rsp)
499 movq %rbx,352(%rsp)
500
501 # qhasm: r15_stack = caller_r15
502 # asm 1: movq <caller_r15=int64#13,>r15_stack=stack64#6
503 # asm 2: movq <caller_r15=%r15,>r15_stack=360(%rsp)
504 movq %r15,360(%rsp)
505
506 # qhasm: rbp_stack = caller_rbp
507 # asm 1: movq <caller_rbp=int64#15,>rbp_stack=stack64#7
508 # asm 2: movq <caller_rbp=%rbp,>rbp_stack=368(%rsp)
509 movq %rbp,368(%rsp)
510
511 # qhasm: inlen_stack = input_2
512 # asm 1: movq <input_2=int64#3,>inlen_stack=stack64#8
513 # asm 2: movq <input_2=%rdx,>inlen_stack=376(%rsp)
514 movq %rdx,376(%rsp)
515
516 # qhasm: in = input_1
517 # asm 1: mov <input_1=int64#2,>in=int64#1
518 # asm 2: mov <input_1=%rsi,>in=%rdi
519 mov %rsi,%rdi
520
521 # qhasm: statebytes_stack = statebytes
522 # asm 1: movq <statebytes=int64#4,>statebytes_stack=stack64#9
523 # asm 2: movq <statebytes=%rcx,>statebytes_stack=384(%rsp)
524 movq %rcx,384(%rsp)
525
526 # qhasm: r0 = state0123[0]
527 # asm 1: movq <state0123=stack256#1,>r0=int64#2
528 # asm 2: movq <state0123=0(%rsp),>r0=%rsi
529 movq 0(%rsp),%rsi
530
531 # qhasm: r2 = state0123[2]
532 # asm 1: movq <state0123=stack256#1,>r2=int64#3
533 # asm 2: movq <state0123=16(%rsp),>r2=%rdx
534 movq 16(%rsp),%rdx
535
536 # qhasm: constants_stack = constants
537 # asm 1: movq <constants=int64#5,>constants_stack=stack64#10
538 # asm 2: movq <constants=%r8,>constants_stack=392(%rsp)
539 movq %r8,392(%rsp)
540
541 # qhasm: r1 = state0123[1]
542 # asm 1: movq <state0123=stack256#1,>r1=int64#4
543 # asm 2: movq <state0123=8(%rsp),>r1=%rcx
544 movq 8(%rsp),%rcx
545
546 # qhasm: r3 = state0123[3]
547 # asm 1: movq <state0123=stack256#1,>r3=int64#6
548 # asm 2: movq <state0123=24(%rsp),>r3=%r9
549 movq 24(%rsp),%r9
550
551 # qhasm: r5 = state4567[1]
552 # asm 1: movq <state4567=stack256#2,>r5=int64#8
553 # asm 2: movq <state4567=40(%rsp),>r5=%r10
554 movq 40(%rsp),%r10
555
556 # qhasm: r4 = state4567[0]
557 # asm 1: movq <state4567=stack256#2,>r4=int64#9
558 # asm 2: movq <state4567=32(%rsp),>r4=%r11
559 movq 32(%rsp),%r11
560
561 # qhasm: r6 = state4567[2]
562 # asm 1: movq <state4567=stack256#2,>r6=int64#10
563 # asm 2: movq <state4567=48(%rsp),>r6=%r12
564 movq 48(%rsp),%r12
565
566 # qhasm: r7 = state4567[3]
567 # asm 1: movq <state4567=stack256#2,>r7=int64#11
568 # asm 2: movq <state4567=56(%rsp),>r7=%r13
569 movq 56(%rsp),%r13
570
571 # qhasm: new w
572
573 # qhasm: nop9
574 .byte 0x66,0x0f,0x1f,0x84,0x00,0x00,0x00,0x00,0x00
575
576 # qhasm: nop9
577 .byte 0x66,0x0f,0x1f,0x84,0x00,0x00,0x00,0x00,0x00
578
579 # qhasm: nop2
580 .byte 0x66,0x90
581
582 # qhasm: outerloop:
583 ._outerloop:
584
585 # qhasm: X0 = mem256[in + 0]
586 # asm 1: vmovupd 0(<in=int64#1),>X0=reg256#2
587 # asm 2: vmovupd 0(<in=%rdi),>X0=%ymm1
588 vmovupd 0(%rdi),%ymm1
589
590 # qhasm: 2x 16x X0 = X0[bigendian64]
591 # asm 1: vpshufb <bigendian64=reg256#1,<X0=reg256#2,>X0=reg256#2
592 # asm 2: vpshufb <bigendian64=%ymm0,<X0=%ymm1,>X0=%ymm1
593 vpshufb %ymm0,%ymm1,%ymm1
594
595 # qhasm: ch7 = r6
596 # asm 1: mov <r6=int64#10,>ch7=int64#7
597 # asm 2: mov <r6=%r12,>ch7=%rax
598 mov %r12,%rax
599
600 # qhasm: r4Sigma1 = r4>>>14
601 # asm 1: rorx $14,<r4=int64#9,>r4Sigma1=int64#12
602 # asm 2: rorx $14,<r4=%r11,>r4Sigma1=%r14
603 rorx $14,%r11,%r14
604
605 # qhasm: ch7 ^= r5
606 # asm 1: xor <r5=int64#8,<ch7=int64#7
607 # asm 2: xor <r5=%r10,<ch7=%rax
608 xor %r10,%rax
609
610 # qhasm: 4x D0 = X0 + mem256[constants + 0]
611 # asm 1: vpaddq 0(<constants=int64#5),<X0=reg256#2,>D0=reg256#3
612 # asm 2: vpaddq 0(<constants=%r8),<X0=%ymm1,>D0=%ymm2
613 vpaddq 0(%r8),%ymm1,%ymm2
614
615 # qhasm: r418 = r4>>>18
616 # asm 1: rorx $18,<r4=int64#9,>r418=int64#13
617 # asm 2: rorx $18,<r4=%r11,>r418=%r15
618 rorx $18,%r11,%r15
619
620 # qhasm: r4Sigma1 ^= r418
621 # asm 1: xor <r418=int64#13,<r4Sigma1=int64#12
622 # asm 2: xor <r418=%r15,<r4Sigma1=%r14
623 xor %r15,%r14
624
625 # qhasm: ch7 &= r4
626 # asm 1: and <r4=int64#9,<ch7=int64#7
627 # asm 2: and <r4=%r11,<ch7=%rax
628 and %r11,%rax
629
630 # qhasm: r441 = r4>>>41
631 # asm 1: rorx $41,<r4=int64#9,>r441=int64#13
632 # asm 2: rorx $41,<r4=%r11,>r441=%r15
633 rorx $41,%r11,%r15
634
635 # qhasm: r4Sigma1 ^= r441
636 # asm 1: xor <r441=int64#13,<r4Sigma1=int64#12
637 # asm 2: xor <r441=%r15,<r4Sigma1=%r14
638 xor %r15,%r14
639
640 # qhasm: r0Sigma0 = r0>>>28
641 # asm 1: rorx $28,<r0=int64#2,>r0Sigma0=int64#13
642 # asm 2: rorx $28,<r0=%rsi,>r0Sigma0=%r15
643 rorx $28,%rsi,%r15
644
645 # qhasm: ch7 ^= r6
646 # asm 1: xor <r6=int64#10,<ch7=int64#7
647 # asm 2: xor <r6=%r12,<ch7=%rax
648 xor %r12,%rax
649
650 # qhasm: r034 = r0>>>34
651 # asm 1: rorx $34,<r0=int64#2,>r034=int64#14
652 # asm 2: rorx $34,<r0=%rsi,>r034=%rbx
653 rorx $34,%rsi,%rbx
654
655 # qhasm: r039 = r0>>>39
656 # asm 1: rorx $39,<r0=int64#2,>r039=int64#15
657 # asm 2: rorx $39,<r0=%rsi,>r039=%rbp
658 rorx $39,%rsi,%rbp
659
660 # qhasm: inplace state4567[3] = r7
661 # asm 1: movq <r7=int64#11,<state4567=stack256#2
662 # asm 2: movq <r7=%r13,<state4567=56(%rsp)
663 movq %r13,56(%rsp)
664
665 # qhasm: r0Sigma0 ^= r034
666 # asm 1: xor <r034=int64#14,<r0Sigma0=int64#13
667 # asm 2: xor <r034=%rbx,<r0Sigma0=%r15
668 xor %rbx,%r15
669
670 # qhasm: r7 += ch7
671 # asm 1: add <ch7=int64#7,<r7=int64#11
672 # asm 2: add <ch7=%rax,<r7=%r13
673 add %rax,%r13
674
675 # qhasm: maj6 = r1
676 # asm 1: mov <r1=int64#4,>maj6=int64#7
677 # asm 2: mov <r1=%rcx,>maj6=%rax
678 mov %rcx,%rax
679
680 # qhasm: maj6 ^= r0
681 # asm 1: xor <r0=int64#2,<maj6=int64#7
682 # asm 2: xor <r0=%rsi,<maj6=%rax
683 xor %rsi,%rax
684
685 # qhasm: r0Sigma0 ^= r039
686 # asm 1: xor <r039=int64#15,<r0Sigma0=int64#13
687 # asm 2: xor <r039=%rbp,<r0Sigma0=%r15
688 xor %rbp,%r15
689
690 # qhasm: r0andr1 = r1
691 # asm 1: mov <r1=int64#4,>r0andr1=int64#14
692 # asm 2: mov <r1=%rcx,>r0andr1=%rbx
693 mov %rcx,%rbx
694
695 # qhasm: r0andr1 &= r0
696 # asm 1: and <r0=int64#2,<r0andr1=int64#14
697 # asm 2: and <r0=%rsi,<r0andr1=%rbx
698 and %rsi,%rbx
699
700 # qhasm: r7 += r4Sigma1
701 # asm 1: add <r4Sigma1=int64#12,<r7=int64#11
702 # asm 2: add <r4Sigma1=%r14,<r7=%r13
703 add %r14,%r13
704
705 # qhasm: maj7 = r2
706 # asm 1: mov <r2=int64#3,>maj7=int64#12
707 # asm 2: mov <r2=%rdx,>maj7=%r14
708 mov %rdx,%r14
709
710 # qhasm: wc0123 = D0
711 # asm 1: vmovapd <D0=reg256#3,>wc0123=stack256#3
712 # asm 2: vmovapd <D0=%ymm2,>wc0123=64(%rsp)
713 vmovapd %ymm2,64(%rsp)
714
715 # qhasm: r7 += wc0123[0]
716 # asm 1: addq <wc0123=stack256#3,<r7=int64#11
717 # asm 2: addq <wc0123=64(%rsp),<r7=%r13
718 addq 64(%rsp),%r13
719
720 # qhasm: maj7 &= maj6
721 # asm 1: and <maj6=int64#7,<maj7=int64#12
722 # asm 2: and <maj6=%rax,<maj7=%r14
723 and %rax,%r14
724
725 # qhasm: inplace state0123[3] = r3
726 # asm 1: movq <r3=int64#6,<state0123=stack256#1
727 # asm 2: movq <r3=%r9,<state0123=24(%rsp)
728 movq %r9,24(%rsp)
729
730 # qhasm: r3 += r7
731 # asm 1: add <r7=int64#11,<r3=int64#6
732 # asm 2: add <r7=%r13,<r3=%r9
733 add %r13,%r9
734
735 # qhasm: r7 += r0Sigma0
736 # asm 1: add <r0Sigma0=int64#13,<r7=int64#11
737 # asm 2: add <r0Sigma0=%r15,<r7=%r13
738 add %r15,%r13
739
740 # qhasm: maj7 ^= r0andr1
741 # asm 1: xor <r0andr1=int64#14,<maj7=int64#12
742 # asm 2: xor <r0andr1=%rbx,<maj7=%r14
743 xor %rbx,%r14
744
745 # qhasm: ch6 = r5
746 # asm 1: mov <r5=int64#8,>ch6=int64#13
747 # asm 2: mov <r5=%r10,>ch6=%r15
748 mov %r10,%r15
749
750 # qhasm: r3Sigma1 = r3>>>14
751 # asm 1: rorx $14,<r3=int64#6,>r3Sigma1=int64#15
752 # asm 2: rorx $14,<r3=%r9,>r3Sigma1=%rbp
753 rorx $14,%r9,%rbp
754
755 # qhasm: ch6 ^= r4
756 # asm 1: xor <r4=int64#9,<ch6=int64#13
757 # asm 2: xor <r4=%r11,<ch6=%r15
758 xor %r11,%r15
759
760 # qhasm: X4 = mem256[in + 32]
761 # asm 1: vmovupd 32(<in=int64#1),>X4=reg256#3
762 # asm 2: vmovupd 32(<in=%rdi),>X4=%ymm2
763 vmovupd 32(%rdi),%ymm2
764
765 # qhasm: 2x 16x X4 = X4[bigendian64]
766 # asm 1: vpshufb <bigendian64=reg256#1,<X4=reg256#3,>X4=reg256#3
767 # asm 2: vpshufb <bigendian64=%ymm0,<X4=%ymm2,>X4=%ymm2
768 vpshufb %ymm0,%ymm2,%ymm2
769
770 # qhasm: r7 += maj7
771 # asm 1: add <maj7=int64#12,<r7=int64#11
772 # asm 2: add <maj7=%r14,<r7=%r13
773 add %r14,%r13
774
775 # qhasm: r318 = r3>>>18
776 # asm 1: rorx $18,<r3=int64#6,>r318=int64#12
777 # asm 2: rorx $18,<r3=%r9,>r318=%r14
778 rorx $18,%r9,%r14
779
780 # qhasm: 4x D4 = X4 + mem256[constants + 32]
781 # asm 1: vpaddq 32(<constants=int64#5),<X4=reg256#3,>D4=reg256#4
782 # asm 2: vpaddq 32(<constants=%r8),<X4=%ymm2,>D4=%ymm3
783 vpaddq 32(%r8),%ymm2,%ymm3
784
785 # qhasm: ch6 &= r3
786 # asm 1: and <r3=int64#6,<ch6=int64#13
787 # asm 2: and <r3=%r9,<ch6=%r15
788 and %r9,%r15
789
790 # qhasm: r3Sigma1 ^= r318
791 # asm 1: xor <r318=int64#12,<r3Sigma1=int64#15
792 # asm 2: xor <r318=%r14,<r3Sigma1=%rbp
793 xor %r14,%rbp
794
795 # qhasm: maj6 &= r7
796 # asm 1: and <r7=int64#11,<maj6=int64#7
797 # asm 2: and <r7=%r13,<maj6=%rax
798 and %r13,%rax
799
800 # qhasm: inplace state4567[2] = r6
801 # asm 1: movq <r6=int64#10,<state4567=stack256#2
802 # asm 2: movq <r6=%r12,<state4567=48(%rsp)
803 movq %r12,48(%rsp)
804
805 # qhasm: r341 = r3>>>41
806 # asm 1: rorx $41,<r3=int64#6,>r341=int64#12
807 # asm 2: rorx $41,<r3=%r9,>r341=%r14
808 rorx $41,%r9,%r14
809
810 # qhasm: ch6 ^= r5
811 # asm 1: xor <r5=int64#8,<ch6=int64#13
812 # asm 2: xor <r5=%r10,<ch6=%r15
813 xor %r10,%r15
814
815 # qhasm: maj6 ^= r0andr1
816 # asm 1: xor <r0andr1=int64#14,<maj6=int64#7
817 # asm 2: xor <r0andr1=%rbx,<maj6=%rax
818 xor %rbx,%rax
819
820 # qhasm: r7Sigma0 = r7>>>28
821 # asm 1: rorx $28,<r7=int64#11,>r7Sigma0=int64#14
822 # asm 2: rorx $28,<r7=%r13,>r7Sigma0=%rbx
823 rorx $28,%r13,%rbx
824
825 # qhasm: inplace state4567[1] = r5
826 # asm 1: movq <r5=int64#8,<state4567=stack256#2
827 # asm 2: movq <r5=%r10,<state4567=40(%rsp)
828 movq %r10,40(%rsp)
829
830 # qhasm: r6 += ch6
831 # asm 1: add <ch6=int64#13,<r6=int64#10
832 # asm 2: add <ch6=%r15,<r6=%r12
833 add %r15,%r12
834
835 # qhasm: r3Sigma1 ^= r341
836 # asm 1: xor <r341=int64#12,<r3Sigma1=int64#15
837 # asm 2: xor <r341=%r14,<r3Sigma1=%rbp
838 xor %r14,%rbp
839
840 # qhasm: r734 = r7>>>34
841 # asm 1: rorx $34,<r7=int64#11,>r734=int64#12
842 # asm 2: rorx $34,<r7=%r13,>r734=%r14
843 rorx $34,%r13,%r14
844
845 # qhasm: inplace state0123[2] = r2
846 # asm 1: movq <r2=int64#3,<state0123=stack256#1
847 # asm 2: movq <r2=%rdx,<state0123=16(%rsp)
848 movq %rdx,16(%rsp)
849
850 # qhasm: r5 += wc0123[2]
851 # asm 1: addq <wc0123=stack256#3,<r5=int64#8
852 # asm 2: addq <wc0123=80(%rsp),<r5=%r10
853 addq 80(%rsp),%r10
854
855 # qhasm: r739 = r7>>>39
856 # asm 1: rorx $39,<r7=int64#11,>r739=int64#13
857 # asm 2: rorx $39,<r7=%r13,>r739=%r15
858 rorx $39,%r13,%r15
859
860 # qhasm: r7Sigma0 ^= r734
861 # asm 1: xor <r734=int64#12,<r7Sigma0=int64#14
862 # asm 2: xor <r734=%r14,<r7Sigma0=%rbx
863 xor %r14,%rbx
864
865 # qhasm: ch5 = r4
866 # asm 1: mov <r4=int64#9,>ch5=int64#12
867 # asm 2: mov <r4=%r11,>ch5=%r14
868 mov %r11,%r14
869
870 # qhasm: ch5 ^= r3
871 # asm 1: xor <r3=int64#6,<ch5=int64#12
872 # asm 2: xor <r3=%r9,<ch5=%r14
873 xor %r9,%r14
874
875 # qhasm: r6 += wc0123[1]
876 # asm 1: addq <wc0123=stack256#3,<r6=int64#10
877 # asm 2: addq <wc0123=72(%rsp),<r6=%r12
878 addq 72(%rsp),%r12
879
880 # qhasm: r6 += r3Sigma1
881 # asm 1: add <r3Sigma1=int64#15,<r6=int64#10
882 # asm 2: add <r3Sigma1=%rbp,<r6=%r12
883 add %rbp,%r12
884
885 # qhasm: mem256[&w + 0] = X0
886 # asm 1: vmovupd <X0=reg256#2,<w=stack1280#1
887 # asm 2: vmovupd <X0=%ymm1,<w=160(%rsp)
888 vmovupd %ymm1,160(%rsp)
889
890 # qhasm: r7Sigma0 ^= r739
891 # asm 1: xor <r739=int64#13,<r7Sigma0=int64#14
892 # asm 2: xor <r739=%r15,<r7Sigma0=%rbx
893 xor %r15,%rbx
894
895 # qhasm: r2 += r6
896 # asm 1: add <r6=int64#10,<r2=int64#3
897 # asm 2: add <r6=%r12,<r2=%rdx
898 add %r12,%rdx
899
900 # qhasm: r6 += maj6
901 # asm 1: add <maj6=int64#7,<r6=int64#10
902 # asm 2: add <maj6=%rax,<r6=%r12
903 add %rax,%r12
904
905 # qhasm: inplace state4567[0] = r4
906 # asm 1: movq <r4=int64#9,<state4567=stack256#2
907 # asm 2: movq <r4=%r11,<state4567=32(%rsp)
908 movq %r11,32(%rsp)
909
910 # qhasm: r2Sigma1 = r2>>>14
911 # asm 1: rorx $14,<r2=int64#3,>r2Sigma1=int64#7
912 # asm 2: rorx $14,<r2=%rdx,>r2Sigma1=%rax
913 rorx $14,%rdx,%rax
914
915 # qhasm: r6 += r7Sigma0
916 # asm 1: add <r7Sigma0=int64#14,<r6=int64#10
917 # asm 2: add <r7Sigma0=%rbx,<r6=%r12
918 add %rbx,%r12
919
920 # qhasm: ch5 &= r2
921 # asm 1: and <r2=int64#3,<ch5=int64#12
922 # asm 2: and <r2=%rdx,<ch5=%r14
923 and %rdx,%r14
924
925 # qhasm: r218 = r2>>>18
926 # asm 1: rorx $18,<r2=int64#3,>r218=int64#13
927 # asm 2: rorx $18,<r2=%rdx,>r218=%r15
928 rorx $18,%rdx,%r15
929
930 # qhasm: mem256[&w + 32] = X4
931 # asm 1: vmovupd <X4=reg256#3,<w=stack1280#1
932 # asm 2: vmovupd <X4=%ymm2,<w=192(%rsp)
933 vmovupd %ymm2,192(%rsp)
934
935 # qhasm: r2Sigma1 ^= r218
936 # asm 1: xor <r218=int64#13,<r2Sigma1=int64#7
937 # asm 2: xor <r218=%r15,<r2Sigma1=%rax
938 xor %r15,%rax
939
940 # qhasm: ch5 ^= r4
941 # asm 1: xor <r4=int64#9,<ch5=int64#12
942 # asm 2: xor <r4=%r11,<ch5=%r14
943 xor %r11,%r14
944
945 # qhasm: r241 = r2>>>41
946 # asm 1: rorx $41,<r2=int64#3,>r241=int64#13
947 # asm 2: rorx $41,<r2=%rdx,>r241=%r15
948 rorx $41,%rdx,%r15
949
950 # qhasm: maj4 = r7
951 # asm 1: mov <r7=int64#11,>maj4=int64#14
952 # asm 2: mov <r7=%r13,>maj4=%rbx
953 mov %r13,%rbx
954
955 # qhasm: maj4 ^= r6
956 # asm 1: xor <r6=int64#10,<maj4=int64#14
957 # asm 2: xor <r6=%r12,<maj4=%rbx
958 xor %r12,%rbx
959
960 # qhasm: r6Sigma0 = r6>>>28
961 # asm 1: rorx $28,<r6=int64#10,>r6Sigma0=int64#15
962 # asm 2: rorx $28,<r6=%r12,>r6Sigma0=%rbp
963 rorx $28,%r12,%rbp
964
965 # qhasm: wc4567 = D4
966 # asm 1: vmovapd <D4=reg256#4,>wc4567=stack256#4
967 # asm 2: vmovapd <D4=%ymm3,>wc4567=96(%rsp)
968 vmovapd %ymm3,96(%rsp)
969
970 # qhasm: r2Sigma1 ^= r241
971 # asm 1: xor <r241=int64#13,<r2Sigma1=int64#7
972 # asm 2: xor <r241=%r15,<r2Sigma1=%rax
973 xor %r15,%rax
974
975 # qhasm: r5 += ch5
976 # asm 1: add <ch5=int64#12,<r5=int64#8
977 # asm 2: add <ch5=%r14,<r5=%r10
978 add %r14,%r10
979
980 # qhasm: r634 = r6>>>34
981 # asm 1: rorx $34,<r6=int64#10,>r634=int64#12
982 # asm 2: rorx $34,<r6=%r12,>r634=%r14
983 rorx $34,%r12,%r14
984
985 # qhasm: in_stack = in
986 # asm 1: movq <in=int64#1,>in_stack=stack64#11
987 # asm 2: movq <in=%rdi,>in_stack=400(%rsp)
988 movq %rdi,400(%rsp)
989
990 # qhasm: r6Sigma0 ^= r634
991 # asm 1: xor <r634=int64#12,<r6Sigma0=int64#15
992 # asm 2: xor <r634=%r14,<r6Sigma0=%rbp
993 xor %r14,%rbp
994
995 # qhasm: r639 = r6>>>39
996 # asm 1: rorx $39,<r6=int64#10,>r639=int64#12
997 # asm 2: rorx $39,<r6=%r12,>r639=%r14
998 rorx $39,%r12,%r14
999
1000 # qhasm: r6andr7 = r7
1001 # asm 1: mov <r7=int64#11,>r6andr7=int64#13
1002 # asm 2: mov <r7=%r13,>r6andr7=%r15
1003 mov %r13,%r15
1004
1005 # qhasm: r6andr7 &= r6
1006 # asm 1: and <r6=int64#10,<r6andr7=int64#13
1007 # asm 2: and <r6=%r12,<r6andr7=%r15
1008 and %r12,%r15
1009
1010 # qhasm: r6Sigma0 ^= r639
1011 # asm 1: xor <r639=int64#12,<r6Sigma0=int64#15
1012 # asm 2: xor <r639=%r14,<r6Sigma0=%rbp
1013 xor %r14,%rbp
1014
1015 # qhasm: maj5 = r0
1016 # asm 1: mov <r0=int64#2,>maj5=int64#12
1017 # asm 2: mov <r0=%rsi,>maj5=%r14
1018 mov %rsi,%r14
1019
1020 # qhasm: inplace state0123[1] = r1
1021 # asm 1: movq <r1=int64#4,<state0123=stack256#1
1022 # asm 2: movq <r1=%rcx,<state0123=8(%rsp)
1023 movq %rcx,8(%rsp)
1024
1025 # qhasm: r5 += r2Sigma1
1026 # asm 1: add <r2Sigma1=int64#7,<r5=int64#8
1027 # asm 2: add <r2Sigma1=%rax,<r5=%r10
1028 add %rax,%r10
1029
1030 # qhasm: r4 += wc0123[3]
1031 # asm 1: addq <wc0123=stack256#3,<r4=int64#9
1032 # asm 2: addq <wc0123=88(%rsp),<r4=%r11
1033 addq 88(%rsp),%r11
1034
1035 # qhasm: maj5 &= maj4
1036 # asm 1: and <maj4=int64#14,<maj5=int64#12
1037 # asm 2: and <maj4=%rbx,<maj5=%r14
1038 and %rbx,%r14
1039
1040 # qhasm: r1 += r5
1041 # asm 1: add <r5=int64#8,<r1=int64#4
1042 # asm 2: add <r5=%r10,<r1=%rcx
1043 add %r10,%rcx
1044
1045 # qhasm: r5 += r6Sigma0
1046 # asm 1: add <r6Sigma0=int64#15,<r5=int64#8
1047 # asm 2: add <r6Sigma0=%rbp,<r5=%r10
1048 add %rbp,%r10
1049
1050 # qhasm: maj5 ^= r6andr7
1051 # asm 1: xor <r6andr7=int64#13,<maj5=int64#12
1052 # asm 2: xor <r6andr7=%r15,<maj5=%r14
1053 xor %r15,%r14
1054
1055 # qhasm: ch4 = r3
1056 # asm 1: mov <r3=int64#6,>ch4=int64#7
1057 # asm 2: mov <r3=%r9,>ch4=%rax
1058 mov %r9,%rax
1059
1060 # qhasm: r1Sigma1 = r1>>>14
1061 # asm 1: rorx $14,<r1=int64#4,>r1Sigma1=int64#15
1062 # asm 2: rorx $14,<r1=%rcx,>r1Sigma1=%rbp
1063 rorx $14,%rcx,%rbp
1064
1065 # qhasm: ch4 ^= r2
1066 # asm 1: xor <r2=int64#3,<ch4=int64#7
1067 # asm 2: xor <r2=%rdx,<ch4=%rax
1068 xor %rdx,%rax
1069
1070 # qhasm: r5 += maj5
1071 # asm 1: add <maj5=int64#12,<r5=int64#8
1072 # asm 2: add <maj5=%r14,<r5=%r10
1073 add %r14,%r10
1074
1075 # qhasm: ch4 &= r1
1076 # asm 1: and <r1=int64#4,<ch4=int64#7
1077 # asm 2: and <r1=%rcx,<ch4=%rax
1078 and %rcx,%rax
1079
1080 # qhasm: r118 = r1>>>18
1081 # asm 1: rorx $18,<r1=int64#4,>r118=int64#12
1082 # asm 2: rorx $18,<r1=%rcx,>r118=%r14
1083 rorx $18,%rcx,%r14
1084
1085 # qhasm: inplace state0123[0] = r0
1086 # asm 1: movq <r0=int64#2,<state0123=stack256#1
1087 # asm 2: movq <r0=%rsi,<state0123=0(%rsp)
1088 movq %rsi,0(%rsp)
1089
1090 # qhasm: r1Sigma1 ^= r118
1091 # asm 1: xor <r118=int64#12,<r1Sigma1=int64#15
1092 # asm 2: xor <r118=%r14,<r1Sigma1=%rbp
1093 xor %r14,%rbp
1094
1095 # qhasm: maj4 &= r5
1096 # asm 1: and <r5=int64#8,<maj4=int64#14
1097 # asm 2: and <r5=%r10,<maj4=%rbx
1098 and %r10,%rbx
1099
1100 # qhasm: ch4 ^= r3
1101 # asm 1: xor <r3=int64#6,<ch4=int64#7
1102 # asm 2: xor <r3=%r9,<ch4=%rax
1103 xor %r9,%rax
1104
1105 # qhasm: r141 = r1>>>41
1106 # asm 1: rorx $41,<r1=int64#4,>r141=int64#12
1107 # asm 2: rorx $41,<r1=%rcx,>r141=%r14
1108 rorx $41,%rcx,%r14
1109
1110 # qhasm: X8 = mem256[in + 64]
1111 # asm 1: vmovupd 64(<in=int64#1),>X8=reg256#4
1112 # asm 2: vmovupd 64(<in=%rdi),>X8=%ymm3
1113 vmovupd 64(%rdi),%ymm3
1114
1115 # qhasm: r1Sigma1 ^= r141
1116 # asm 1: xor <r141=int64#12,<r1Sigma1=int64#15
1117 # asm 2: xor <r141=%r14,<r1Sigma1=%rbp
1118 xor %r14,%rbp
1119
1120 # qhasm: r5Sigma0 = r5>>>28
1121 # asm 1: rorx $28,<r5=int64#8,>r5Sigma0=int64#12
1122 # asm 2: rorx $28,<r5=%r10,>r5Sigma0=%r14
1123 rorx $28,%r10,%r14
1124
1125 # qhasm: maj4 ^= r6andr7
1126 # asm 1: xor <r6andr7=int64#13,<maj4=int64#14
1127 # asm 2: xor <r6andr7=%r15,<maj4=%rbx
1128 xor %r15,%rbx
1129
1130 # qhasm: r4 += ch4
1131 # asm 1: add <ch4=int64#7,<r4=int64#9
1132 # asm 2: add <ch4=%rax,<r4=%r11
1133 add %rax,%r11
1134
1135 # qhasm: r534 = r5>>>34
1136 # asm 1: rorx $34,<r5=int64#8,>r534=int64#7
1137 # asm 2: rorx $34,<r5=%r10,>r534=%rax
1138 rorx $34,%r10,%rax
1139
1140 # qhasm: r4 += r1Sigma1
1141 # asm 1: add <r1Sigma1=int64#15,<r4=int64#9
1142 # asm 2: add <r1Sigma1=%rbp,<r4=%r11
1143 add %rbp,%r11
1144
1145 # qhasm: r5Sigma0 ^= r534
1146 # asm 1: xor <r534=int64#7,<r5Sigma0=int64#12
1147 # asm 2: xor <r534=%rax,<r5Sigma0=%r14
1148 xor %rax,%r14
1149
1150 # qhasm: r3 += wc4567[0]
1151 # asm 1: addq <wc4567=stack256#4,<r3=int64#6
1152 # asm 2: addq <wc4567=96(%rsp),<r3=%r9
1153 addq 96(%rsp),%r9
1154
1155 # qhasm: X12 = mem256[in + 96]
1156 # asm 1: vmovupd 96(<in=int64#1),>X12=reg256#5
1157 # asm 2: vmovupd 96(<in=%rdi),>X12=%ymm4
1158 vmovupd 96(%rdi),%ymm4
1159
1160 # qhasm: r0 += r4
1161 # asm 1: add <r4=int64#9,<r0=int64#2
1162 # asm 2: add <r4=%r11,<r0=%rsi
1163 add %r11,%rsi
1164
1165 # qhasm: r539 = r5>>>39
1166 # asm 1: rorx $39,<r5=int64#8,>r539=int64#1
1167 # asm 2: rorx $39,<r5=%r10,>r539=%rdi
1168 rorx $39,%r10,%rdi
1169
1170 # qhasm: r4 += maj4
1171 # asm 1: add <maj4=int64#14,<r4=int64#9
1172 # asm 2: add <maj4=%rbx,<r4=%r11
1173 add %rbx,%r11
1174
1175 # qhasm: r5Sigma0 ^= r539
1176 # asm 1: xor <r539=int64#1,<r5Sigma0=int64#12
1177 # asm 2: xor <r539=%rdi,<r5Sigma0=%r14
1178 xor %rdi,%r14
1179
1180 # qhasm: r0Sigma1 = r0>>>14
1181 # asm 1: rorx $14,<r0=int64#2,>r0Sigma1=int64#1
1182 # asm 2: rorx $14,<r0=%rsi,>r0Sigma1=%rdi
1183 rorx $14,%rsi,%rdi
1184
1185 # qhasm: r4 += r5Sigma0
1186 # asm 1: add <r5Sigma0=int64#12,<r4=int64#9
1187 # asm 2: add <r5Sigma0=%r14,<r4=%r11
1188 add %r14,%r11
1189
1190 # qhasm: ch3 = r2
1191 # asm 1: mov <r2=int64#3,>ch3=int64#7
1192 # asm 2: mov <r2=%rdx,>ch3=%rax
1193 mov %rdx,%rax
1194
1195 # qhasm: r018 = r0>>>18
1196 # asm 1: rorx $18,<r0=int64#2,>r018=int64#12
1197 # asm 2: rorx $18,<r0=%rsi,>r018=%r14
1198 rorx $18,%rsi,%r14
1199
1200 # qhasm: ch3 ^= r1
1201 # asm 1: xor <r1=int64#4,<ch3=int64#7
1202 # asm 2: xor <r1=%rcx,<ch3=%rax
1203 xor %rcx,%rax
1204
1205 # qhasm: 2x 16x X8 = X8[bigendian64]
1206 # asm 1: vpshufb <bigendian64=reg256#1,<X8=reg256#4,>X8=reg256#4
1207 # asm 2: vpshufb <bigendian64=%ymm0,<X8=%ymm3,>X8=%ymm3
1208 vpshufb %ymm0,%ymm3,%ymm3
1209
1210 # qhasm: r0Sigma1 ^= r018
1211 # asm 1: xor <r018=int64#12,<r0Sigma1=int64#1
1212 # asm 2: xor <r018=%r14,<r0Sigma1=%rdi
1213 xor %r14,%rdi
1214
1215 # qhasm: ch3 &= r0
1216 # asm 1: and <r0=int64#2,<ch3=int64#7
1217 # asm 2: and <r0=%rsi,<ch3=%rax
1218 and %rsi,%rax
1219
1220 # qhasm: r041 = r0>>>41
1221 # asm 1: rorx $41,<r0=int64#2,>r041=int64#12
1222 # asm 2: rorx $41,<r0=%rsi,>r041=%r14
1223 rorx $41,%rsi,%r14
1224
1225 # qhasm: r4Sigma0 = r4>>>28
1226 # asm 1: rorx $28,<r4=int64#9,>r4Sigma0=int64#13
1227 # asm 2: rorx $28,<r4=%r11,>r4Sigma0=%r15
1228 rorx $28,%r11,%r15
1229
1230 # qhasm: 4x D8 = X8 + mem256[constants + 64]
1231 # asm 1: vpaddq 64(<constants=int64#5),<X8=reg256#4,>D8=reg256#6
1232 # asm 2: vpaddq 64(<constants=%r8),<X8=%ymm3,>D8=%ymm5
1233 vpaddq 64(%r8),%ymm3,%ymm5
1234
1235 # qhasm: r0Sigma1 ^= r041
1236 # asm 1: xor <r041=int64#12,<r0Sigma1=int64#1
1237 # asm 2: xor <r041=%r14,<r0Sigma1=%rdi
1238 xor %r14,%rdi
1239
1240 # qhasm: ch3 ^= r2
1241 # asm 1: xor <r2=int64#3,<ch3=int64#7
1242 # asm 2: xor <r2=%rdx,<ch3=%rax
1243 xor %rdx,%rax
1244
1245 # qhasm: mem256[&w + 64] = X8
1246 # asm 1: vmovupd <X8=reg256#4,<w=stack1280#1
1247 # asm 2: vmovupd <X8=%ymm3,<w=224(%rsp)
1248 vmovupd %ymm3,224(%rsp)
1249
1250 # qhasm: r3 += ch3
1251 # asm 1: add <ch3=int64#7,<r3=int64#6
1252 # asm 2: add <ch3=%rax,<r3=%r9
1253 add %rax,%r9
1254
1255 # qhasm: r434 = r4>>>34
1256 # asm 1: rorx $34,<r4=int64#9,>r434=int64#7
1257 # asm 2: rorx $34,<r4=%r11,>r434=%rax
1258 rorx $34,%r11,%rax
1259
1260 # qhasm: r439 = r4>>>39
1261 # asm 1: rorx $39,<r4=int64#9,>r439=int64#12
1262 # asm 2: rorx $39,<r4=%r11,>r439=%r14
1263 rorx $39,%r11,%r14
1264
1265 # qhasm: maj2 = r5
1266 # asm 1: mov <r5=int64#8,>maj2=int64#14
1267 # asm 2: mov <r5=%r10,>maj2=%rbx
1268 mov %r10,%rbx
1269
1270 # qhasm: maj2 ^= r4
1271 # asm 1: xor <r4=int64#9,<maj2=int64#14
1272 # asm 2: xor <r4=%r11,<maj2=%rbx
1273 xor %r11,%rbx
1274
1275 # qhasm: wc891011 = D8
1276 # asm 1: vmovapd <D8=reg256#6,>wc891011=stack256#3
1277 # asm 2: vmovapd <D8=%ymm5,>wc891011=64(%rsp)
1278 vmovapd %ymm5,64(%rsp)
1279
1280 # qhasm: r4Sigma0 ^= r434
1281 # asm 1: xor <r434=int64#7,<r4Sigma0=int64#13
1282 # asm 2: xor <r434=%rax,<r4Sigma0=%r15
1283 xor %rax,%r15
1284
1285 # qhasm: r3 += r0Sigma1
1286 # asm 1: add <r0Sigma1=int64#1,<r3=int64#6
1287 # asm 2: add <r0Sigma1=%rdi,<r3=%r9
1288 add %rdi,%r9
1289
1290 # qhasm: r4andr5 = r5
1291 # asm 1: mov <r5=int64#8,>r4andr5=int64#1
1292 # asm 2: mov <r5=%r10,>r4andr5=%rdi
1293 mov %r10,%rdi
1294
1295 # qhasm: r4andr5 &= r4
1296 # asm 1: and <r4=int64#9,<r4andr5=int64#1
1297 # asm 2: and <r4=%r11,<r4andr5=%rdi
1298 and %r11,%rdi
1299
1300 # qhasm: r2 += wc4567[1]
1301 # asm 1: addq <wc4567=stack256#4,<r2=int64#3
1302 # asm 2: addq <wc4567=104(%rsp),<r2=%rdx
1303 addq 104(%rsp),%rdx
1304
1305 # qhasm: r4Sigma0 ^= r439
1306 # asm 1: xor <r439=int64#12,<r4Sigma0=int64#13
1307 # asm 2: xor <r439=%r14,<r4Sigma0=%r15
1308 xor %r14,%r15
1309
1310 # qhasm: maj3 = r6
1311 # asm 1: mov <r6=int64#10,>maj3=int64#7
1312 # asm 2: mov <r6=%r12,>maj3=%rax
1313 mov %r12,%rax
1314
1315 # qhasm: maj3 &= maj2
1316 # asm 1: and <maj2=int64#14,<maj3=int64#7
1317 # asm 2: and <maj2=%rbx,<maj3=%rax
1318 and %rbx,%rax
1319
1320 # qhasm: r7 += r3
1321 # asm 1: add <r3=int64#6,<r7=int64#11
1322 # asm 2: add <r3=%r9,<r7=%r13
1323 add %r9,%r13
1324
1325 # qhasm: r3 += r4Sigma0
1326 # asm 1: add <r4Sigma0=int64#13,<r3=int64#6
1327 # asm 2: add <r4Sigma0=%r15,<r3=%r9
1328 add %r15,%r9
1329
1330 # qhasm: 2x 16x X12 = X12[bigendian64]
1331 # asm 1: vpshufb <bigendian64=reg256#1,<X12=reg256#5,>X12=reg256#5
1332 # asm 2: vpshufb <bigendian64=%ymm0,<X12=%ymm4,>X12=%ymm4
1333 vpshufb %ymm0,%ymm4,%ymm4
1334
1335 # qhasm: ch2 = r1
1336 # asm 1: mov <r1=int64#4,>ch2=int64#12
1337 # asm 2: mov <r1=%rcx,>ch2=%r14
1338 mov %rcx,%r14
1339
1340 # qhasm: maj3 ^= r4andr5
1341 # asm 1: xor <r4andr5=int64#1,<maj3=int64#7
1342 # asm 2: xor <r4andr5=%rdi,<maj3=%rax
1343 xor %rdi,%rax
1344
1345 # qhasm: ch2 ^= r0
1346 # asm 1: xor <r0=int64#2,<ch2=int64#12
1347 # asm 2: xor <r0=%rsi,<ch2=%r14
1348 xor %rsi,%r14
1349
1350 # qhasm: r3 += maj3
1351 # asm 1: add <maj3=int64#7,<r3=int64#6
1352 # asm 2: add <maj3=%rax,<r3=%r9
1353 add %rax,%r9
1354
1355 # qhasm: r7Sigma1 = r7>>>14
1356 # asm 1: rorx $14,<r7=int64#11,>r7Sigma1=int64#7
1357 # asm 2: rorx $14,<r7=%r13,>r7Sigma1=%rax
1358 rorx $14,%r13,%rax
1359
1360 # qhasm: 4x D12 = X12 + mem256[constants + 96]
1361 # asm 1: vpaddq 96(<constants=int64#5),<X12=reg256#5,>D12=reg256#6
1362 # asm 2: vpaddq 96(<constants=%r8),<X12=%ymm4,>D12=%ymm5
1363 vpaddq 96(%r8),%ymm4,%ymm5
1364
1365 # qhasm: ch2 &= r7
1366 # asm 1: and <r7=int64#11,<ch2=int64#12
1367 # asm 2: and <r7=%r13,<ch2=%r14
1368 and %r13,%r14
1369
1370 # qhasm: r718 = r7>>>18
1371 # asm 1: rorx $18,<r7=int64#11,>r718=int64#13
1372 # asm 2: rorx $18,<r7=%r13,>r718=%r15
1373 rorx $18,%r13,%r15
1374
1375 # qhasm: r7Sigma1 ^= r718
1376 # asm 1: xor <r718=int64#13,<r7Sigma1=int64#7
1377 # asm 2: xor <r718=%r15,<r7Sigma1=%rax
1378 xor %r15,%rax
1379
1380 # qhasm: maj2 &= r3
1381 # asm 1: and <r3=int64#6,<maj2=int64#14
1382 # asm 2: and <r3=%r9,<maj2=%rbx
1383 and %r9,%rbx
1384
1385 # qhasm: ch2 ^= r1
1386 # asm 1: xor <r1=int64#4,<ch2=int64#12
1387 # asm 2: xor <r1=%rcx,<ch2=%r14
1388 xor %rcx,%r14
1389
1390 # qhasm: r741 = r7>>>41
1391 # asm 1: rorx $41,<r7=int64#11,>r741=int64#13
1392 # asm 2: rorx $41,<r7=%r13,>r741=%r15
1393 rorx $41,%r13,%r15
1394
1395 # qhasm: mem256[&w + 96] = X12
1396 # asm 1: vmovupd <X12=reg256#5,<w=stack1280#1
1397 # asm 2: vmovupd <X12=%ymm4,<w=256(%rsp)
1398 vmovupd %ymm4,256(%rsp)
1399
1400 # qhasm: r7Sigma1 ^= r741
1401 # asm 1: xor <r741=int64#13,<r7Sigma1=int64#7
1402 # asm 2: xor <r741=%r15,<r7Sigma1=%rax
1403 xor %r15,%rax
1404
1405 # qhasm: maj2 ^= r4andr5
1406 # asm 1: xor <r4andr5=int64#1,<maj2=int64#14
1407 # asm 2: xor <r4andr5=%rdi,<maj2=%rbx
1408 xor %rdi,%rbx
1409
1410 # qhasm: r3Sigma0 = r3>>>28
1411 # asm 1: rorx $28,<r3=int64#6,>r3Sigma0=int64#1
1412 # asm 2: rorx $28,<r3=%r9,>r3Sigma0=%rdi
1413 rorx $28,%r9,%rdi
1414
1415 # qhasm: wc12131415 = D12
1416 # asm 1: vmovapd <D12=reg256#6,>wc12131415=stack256#5
1417 # asm 2: vmovapd <D12=%ymm5,>wc12131415=128(%rsp)
1418 vmovapd %ymm5,128(%rsp)
1419
1420 # qhasm: r2 += ch2
1421 # asm 1: add <ch2=int64#12,<r2=int64#3
1422 # asm 2: add <ch2=%r14,<r2=%rdx
1423 add %r14,%rdx
1424
1425 # qhasm: r2 += r7Sigma1
1426 # asm 1: add <r7Sigma1=int64#7,<r2=int64#3
1427 # asm 2: add <r7Sigma1=%rax,<r2=%rdx
1428 add %rax,%rdx
1429
1430 # qhasm: r1 += wc4567[2]
1431 # asm 1: addq <wc4567=stack256#4,<r1=int64#4
1432 # asm 2: addq <wc4567=112(%rsp),<r1=%rcx
1433 addq 112(%rsp),%rcx
1434
1435 # qhasm: ch1 = r0
1436 # asm 1: mov <r0=int64#2,>ch1=int64#7
1437 # asm 2: mov <r0=%rsi,>ch1=%rax
1438 mov %rsi,%rax
1439
1440 # qhasm: r334 = r3>>>34
1441 # asm 1: rorx $34,<r3=int64#6,>r334=int64#12
1442 # asm 2: rorx $34,<r3=%r9,>r334=%r14
1443 rorx $34,%r9,%r14
1444
1445 # qhasm: ch1 ^= r7
1446 # asm 1: xor <r7=int64#11,<ch1=int64#7
1447 # asm 2: xor <r7=%r13,<ch1=%rax
1448 xor %r13,%rax
1449
1450 # qhasm: r3Sigma0 ^= r334
1451 # asm 1: xor <r334=int64#12,<r3Sigma0=int64#1
1452 # asm 2: xor <r334=%r14,<r3Sigma0=%rdi
1453 xor %r14,%rdi
1454
1455 # qhasm: r6 += r2
1456 # asm 1: add <r2=int64#3,<r6=int64#10
1457 # asm 2: add <r2=%rdx,<r6=%r12
1458 add %rdx,%r12
1459
1460 # qhasm: r339 = r3>>>39
1461 # asm 1: rorx $39,<r3=int64#6,>r339=int64#12
1462 # asm 2: rorx $39,<r3=%r9,>r339=%r14
1463 rorx $39,%r9,%r14
1464
1465 # qhasm: r2 += maj2
1466 # asm 1: add <maj2=int64#14,<r2=int64#3
1467 # asm 2: add <maj2=%rbx,<r2=%rdx
1468 add %rbx,%rdx
1469
1470 # qhasm: r3Sigma0 ^= r339
1471 # asm 1: xor <r339=int64#12,<r3Sigma0=int64#1
1472 # asm 2: xor <r339=%r14,<r3Sigma0=%rdi
1473 xor %r14,%rdi
1474
1475 # qhasm: r6Sigma1 = r6>>>14
1476 # asm 1: rorx $14,<r6=int64#10,>r6Sigma1=int64#12
1477 # asm 2: rorx $14,<r6=%r12,>r6Sigma1=%r14
1478 rorx $14,%r12,%r14
1479
1480 # qhasm: r618 = r6>>>18
1481 # asm 1: rorx $18,<r6=int64#10,>r618=int64#13
1482 # asm 2: rorx $18,<r6=%r12,>r618=%r15
1483 rorx $18,%r12,%r15
1484
1485 # qhasm: r641 = r6>>>41
1486 # asm 1: rorx $41,<r6=int64#10,>r641=int64#14
1487 # asm 2: rorx $41,<r6=%r12,>r641=%rbx
1488 rorx $41,%r12,%rbx
1489
1490 # qhasm: ch1 &= r6
1491 # asm 1: and <r6=int64#10,<ch1=int64#7
1492 # asm 2: and <r6=%r12,<ch1=%rax
1493 and %r12,%rax
1494
1495 # qhasm: r2 += r3Sigma0
1496 # asm 1: add <r3Sigma0=int64#1,<r2=int64#3
1497 # asm 2: add <r3Sigma0=%rdi,<r2=%rdx
1498 add %rdi,%rdx
1499
1500 # qhasm: r6Sigma1 ^= r618
1501 # asm 1: xor <r618=int64#13,<r6Sigma1=int64#12
1502 # asm 2: xor <r618=%r15,<r6Sigma1=%r14
1503 xor %r15,%r14
1504
1505 # qhasm: r2Sigma0 = r2>>>28
1506 # asm 1: rorx $28,<r2=int64#3,>r2Sigma0=int64#1
1507 # asm 2: rorx $28,<r2=%rdx,>r2Sigma0=%rdi
1508 rorx $28,%rdx,%rdi
1509
1510 # qhasm: r6Sigma1 ^= r641
1511 # asm 1: xor <r641=int64#14,<r6Sigma1=int64#12
1512 # asm 2: xor <r641=%rbx,<r6Sigma1=%r14
1513 xor %rbx,%r14
1514
1515 # qhasm: ch1 ^= r0
1516 # asm 1: xor <r0=int64#2,<ch1=int64#7
1517 # asm 2: xor <r0=%rsi,<ch1=%rax
1518 xor %rsi,%rax
1519
1520 # qhasm: r234 = r2>>>34
1521 # asm 1: rorx $34,<r2=int64#3,>r234=int64#13
1522 # asm 2: rorx $34,<r2=%rdx,>r234=%r15
1523 rorx $34,%rdx,%r15
1524
1525 # qhasm: maj0 = r3
1526 # asm 1: mov <r3=int64#6,>maj0=int64#14
1527 # asm 2: mov <r3=%r9,>maj0=%rbx
1528 mov %r9,%rbx
1529
1530 # qhasm: maj0 ^= r2
1531 # asm 1: xor <r2=int64#3,<maj0=int64#14
1532 # asm 2: xor <r2=%rdx,<maj0=%rbx
1533 xor %rdx,%rbx
1534
1535 # qhasm: r2Sigma0 ^= r234
1536 # asm 1: xor <r234=int64#13,<r2Sigma0=int64#1
1537 # asm 2: xor <r234=%r15,<r2Sigma0=%rdi
1538 xor %r15,%rdi
1539
1540 # qhasm: r1 += ch1
1541 # asm 1: add <ch1=int64#7,<r1=int64#4
1542 # asm 2: add <ch1=%rax,<r1=%rcx
1543 add %rax,%rcx
1544
1545 # qhasm: r239 = r2>>>39
1546 # asm 1: rorx $39,<r2=int64#3,>r239=int64#7
1547 # asm 2: rorx $39,<r2=%rdx,>r239=%rax
1548 rorx $39,%rdx,%rax
1549
1550 # qhasm: r2andr3 = r3
1551 # asm 1: mov <r3=int64#6,>r2andr3=int64#13
1552 # asm 2: mov <r3=%r9,>r2andr3=%r15
1553 mov %r9,%r15
1554
1555 # qhasm: r2andr3 &= r2
1556 # asm 1: and <r2=int64#3,<r2andr3=int64#13
1557 # asm 2: and <r2=%rdx,<r2andr3=%r15
1558 and %rdx,%r15
1559
1560 # qhasm: r2Sigma0 ^= r239
1561 # asm 1: xor <r239=int64#7,<r2Sigma0=int64#1
1562 # asm 2: xor <r239=%rax,<r2Sigma0=%rdi
1563 xor %rax,%rdi
1564
1565 # qhasm: r1 += r6Sigma1
1566 # asm 1: add <r6Sigma1=int64#12,<r1=int64#4
1567 # asm 2: add <r6Sigma1=%r14,<r1=%rcx
1568 add %r14,%rcx
1569
1570 # qhasm: maj1 = r4
1571 # asm 1: mov <r4=int64#9,>maj1=int64#7
1572 # asm 2: mov <r4=%r11,>maj1=%rax
1573 mov %r11,%rax
1574
1575 # qhasm: maj1 &= maj0
1576 # asm 1: and <maj0=int64#14,<maj1=int64#7
1577 # asm 2: and <maj0=%rbx,<maj1=%rax
1578 and %rbx,%rax
1579
1580 # qhasm: r0 += wc4567[3]
1581 # asm 1: addq <wc4567=stack256#4,<r0=int64#2
1582 # asm 2: addq <wc4567=120(%rsp),<r0=%rsi
1583 addq 120(%rsp),%rsi
1584
1585 # qhasm: r5 += r1
1586 # asm 1: add <r1=int64#4,<r5=int64#8
1587 # asm 2: add <r1=%rcx,<r5=%r10
1588 add %rcx,%r10
1589
1590 # qhasm: r1 += r2Sigma0
1591 # asm 1: add <r2Sigma0=int64#1,<r1=int64#4
1592 # asm 2: add <r2Sigma0=%rdi,<r1=%rcx
1593 add %rdi,%rcx
1594
1595 # qhasm: ch0 = r7
1596 # asm 1: mov <r7=int64#11,>ch0=int64#1
1597 # asm 2: mov <r7=%r13,>ch0=%rdi
1598 mov %r13,%rdi
1599
1600 # qhasm: maj1 ^= r2andr3
1601 # asm 1: xor <r2andr3=int64#13,<maj1=int64#7
1602 # asm 2: xor <r2andr3=%r15,<maj1=%rax
1603 xor %r15,%rax
1604
1605 # qhasm: ch0 ^= r6
1606 # asm 1: xor <r6=int64#10,<ch0=int64#1
1607 # asm 2: xor <r6=%r12,<ch0=%rdi
1608 xor %r12,%rdi
1609
1610 # qhasm: r5Sigma1 = r5>>>14
1611 # asm 1: rorx $14,<r5=int64#8,>r5Sigma1=int64#12
1612 # asm 2: rorx $14,<r5=%r10,>r5Sigma1=%r14
1613 rorx $14,%r10,%r14
1614
1615 # qhasm: r1 += maj1
1616 # asm 1: add <maj1=int64#7,<r1=int64#4
1617 # asm 2: add <maj1=%rax,<r1=%rcx
1618 add %rax,%rcx
1619
1620 # qhasm: ch0 &= r5
1621 # asm 1: and <r5=int64#8,<ch0=int64#1
1622 # asm 2: and <r5=%r10,<ch0=%rdi
1623 and %r10,%rdi
1624
1625 # qhasm: r518 = r5>>>18
1626 # asm 1: rorx $18,<r5=int64#8,>r518=int64#7
1627 # asm 2: rorx $18,<r5=%r10,>r518=%rax
1628 rorx $18,%r10,%rax
1629
1630 # qhasm: r5Sigma1 ^= r518
1631 # asm 1: xor <r518=int64#7,<r5Sigma1=int64#12
1632 # asm 2: xor <r518=%rax,<r5Sigma1=%r14
1633 xor %rax,%r14
1634
1635 # qhasm: maj0 &= r1
1636 # asm 1: and <r1=int64#4,<maj0=int64#14
1637 # asm 2: and <r1=%rcx,<maj0=%rbx
1638 and %rcx,%rbx
1639
1640 # qhasm: ch0 ^= r7
1641 # asm 1: xor <r7=int64#11,<ch0=int64#1
1642 # asm 2: xor <r7=%r13,<ch0=%rdi
1643 xor %r13,%rdi
1644
1645 # qhasm: r541 = r5>>>41
1646 # asm 1: rorx $41,<r5=int64#8,>r541=int64#7
1647 # asm 2: rorx $41,<r5=%r10,>r541=%rax
1648 rorx $41,%r10,%rax
1649
1650 # qhasm: r5Sigma1 ^= r541
1651 # asm 1: xor <r541=int64#7,<r5Sigma1=int64#12
1652 # asm 2: xor <r541=%rax,<r5Sigma1=%r14
1653 xor %rax,%r14
1654
1655 # qhasm: maj0 ^= r2andr3
1656 # asm 1: xor <r2andr3=int64#13,<maj0=int64#14
1657 # asm 2: xor <r2andr3=%r15,<maj0=%rbx
1658 xor %r15,%rbx
1659
1660 # qhasm: r1Sigma0 = r1>>>28
1661 # asm 1: rorx $28,<r1=int64#4,>r1Sigma0=int64#7
1662 # asm 2: rorx $28,<r1=%rcx,>r1Sigma0=%rax
1663 rorx $28,%rcx,%rax
1664
1665 # qhasm: r0 += ch0
1666 # asm 1: add <ch0=int64#1,<r0=int64#2
1667 # asm 2: add <ch0=%rdi,<r0=%rsi
1668 add %rdi,%rsi
1669
1670 # qhasm: r0 += r5Sigma1
1671 # asm 1: add <r5Sigma1=int64#12,<r0=int64#2
1672 # asm 2: add <r5Sigma1=%r14,<r0=%rsi
1673 add %r14,%rsi
1674
1675 # qhasm: r134 = r1>>>34
1676 # asm 1: rorx $34,<r1=int64#4,>r134=int64#1
1677 # asm 2: rorx $34,<r1=%rcx,>r134=%rdi
1678 rorx $34,%rcx,%rdi
1679
1680 # qhasm: r1Sigma0 ^= r134
1681 # asm 1: xor <r134=int64#1,<r1Sigma0=int64#7
1682 # asm 2: xor <r134=%rdi,<r1Sigma0=%rax
1683 xor %rdi,%rax
1684
1685 # qhasm: r4 += r0
1686 # asm 1: add <r0=int64#2,<r4=int64#9
1687 # asm 2: add <r0=%rsi,<r4=%r11
1688 add %rsi,%r11
1689
1690 # qhasm: r0 += maj0
1691 # asm 1: add <maj0=int64#14,<r0=int64#2
1692 # asm 2: add <maj0=%rbx,<r0=%rsi
1693 add %rbx,%rsi
1694
1695 # qhasm: r139 = r1>>>39
1696 # asm 1: rorx $39,<r1=int64#4,>r139=int64#1
1697 # asm 2: rorx $39,<r1=%rcx,>r139=%rdi
1698 rorx $39,%rcx,%rdi
1699
1700 # qhasm: r1Sigma0 ^= r139
1701 # asm 1: xor <r139=int64#1,<r1Sigma0=int64#7
1702 # asm 2: xor <r139=%rdi,<r1Sigma0=%rax
1703 xor %rdi,%rax
1704
1705 # qhasm: r0 += r1Sigma0
1706 # asm 1: add <r1Sigma0=int64#7,<r0=int64#2
1707 # asm 2: add <r1Sigma0=%rax,<r0=%rsi
1708 add %rax,%rsi
1709
1710 # qhasm: i = 4
1711 # asm 1: mov $4,>i=int64#1
1712 # asm 2: mov $4,>i=%rdi
1713 mov $4,%rdi
1714
1715 # qhasm: innerloop:
1716 ._innerloop:
1717
1718 # qhasm: X1 = mem256[&w + 8]
1719 # asm 1: vmovupd <w=stack1280#1,>X1=reg256#6
1720 # asm 2: vmovupd <w=168(%rsp),>X1=%ymm5
1721 vmovupd 168(%rsp),%ymm5
1722
1723 # qhasm: 4x X1right1 = X1 unsigned>> 1
1724 # asm 1: vpsrlq $1,<X1=reg256#6,>X1right1=reg256#7
1725 # asm 2: vpsrlq $1,<X1=%ymm5,>X1right1=%ymm6
1726 vpsrlq $1,%ymm5,%ymm6
1727
1728 # qhasm: r4Sigma1 = r4>>>14
1729 # asm 1: rorx $14,<r4=int64#9,>r4Sigma1=int64#7
1730 # asm 2: rorx $14,<r4=%r11,>r4Sigma1=%rax
1731 rorx $14,%r11,%rax
1732
1733 # qhasm: r7 += wc891011[0]
1734 # asm 1: addq <wc891011=stack256#3,<r7=int64#11
1735 # asm 2: addq <wc891011=64(%rsp),<r7=%r13
1736 addq 64(%rsp),%r13
1737
1738 # qhasm: 4x X1left63 = X1 << 63
1739 # asm 1: vpsllq $63,<X1=reg256#6,>X1left63=reg256#8
1740 # asm 2: vpsllq $63,<X1=%ymm5,>X1left63=%ymm7
1741 vpsllq $63,%ymm5,%ymm7
1742
1743 # qhasm: ch7 = r6
1744 # asm 1: mov <r6=int64#10,>ch7=int64#12
1745 # asm 2: mov <r6=%r12,>ch7=%r14
1746 mov %r12,%r14
1747
1748 # qhasm: ch7 ^= r5
1749 # asm 1: xor <r5=int64#8,<ch7=int64#12
1750 # asm 2: xor <r5=%r10,<ch7=%r14
1751 xor %r10,%r14
1752
1753 # qhasm: r418 = r4>>>18
1754 # asm 1: rorx $18,<r4=int64#9,>r418=int64#13
1755 # asm 2: rorx $18,<r4=%r11,>r418=%r15
1756 rorx $18,%r11,%r15
1757
1758 # qhasm: ch7 &= r4
1759 # asm 1: and <r4=int64#9,<ch7=int64#12
1760 # asm 2: and <r4=%r11,<ch7=%r14
1761 and %r11,%r14
1762
1763 # qhasm: maj6 = r1
1764 # asm 1: mov <r1=int64#4,>maj6=int64#14
1765 # asm 2: mov <r1=%rcx,>maj6=%rbx
1766 mov %rcx,%rbx
1767
1768 # qhasm: maj6 ^= r0
1769 # asm 1: xor <r0=int64#2,<maj6=int64#14
1770 # asm 2: xor <r0=%rsi,<maj6=%rbx
1771 xor %rsi,%rbx
1772
1773 # qhasm: W14 = mem128[&w + 112],0
1774 # asm 1: vmovupd <w=stack1280#1,>W14=reg256#9%128
1775 # asm 2: vmovupd <w=272(%rsp),>W14=%xmm8
1776 vmovupd 272(%rsp),%xmm8
1777
1778 # qhasm: r441 = r4>>>41
1779 # asm 1: rorx $41,<r4=int64#9,>r441=int64#15
1780 # asm 2: rorx $41,<r4=%r11,>r441=%rbp
1781 rorx $41,%r11,%rbp
1782
1783 # qhasm: r4Sigma1 ^= r418
1784 # asm 1: xor <r418=int64#13,<r4Sigma1=int64#7
1785 # asm 2: xor <r418=%r15,<r4Sigma1=%rax
1786 xor %r15,%rax
1787
1788 # qhasm: ch7 ^= r6
1789 # asm 1: xor <r6=int64#10,<ch7=int64#12
1790 # asm 2: xor <r6=%r12,<ch7=%r14
1791 xor %r12,%r14
1792
1793 # qhasm: X1sigma0 = X1right1 ^ X1left63
1794 # asm 1: vpxor <X1right1=reg256#7,<X1left63=reg256#8,>X1sigma0=reg256#7
1795 # asm 2: vpxor <X1right1=%ymm6,<X1left63=%ymm7,>X1sigma0=%ymm6
1796 vpxor %ymm6,%ymm7,%ymm6
1797
1798 # qhasm: 4x X1right8 = X1 unsigned>> 8
1799 # asm 1: vpsrlq $8,<X1=reg256#6,>X1right8=reg256#8
1800 # asm 2: vpsrlq $8,<X1=%ymm5,>X1right8=%ymm7
1801 vpsrlq $8,%ymm5,%ymm7
1802
1803 # qhasm: r4Sigma1 ^= r441
1804 # asm 1: xor <r441=int64#15,<r4Sigma1=int64#7
1805 # asm 2: xor <r441=%rbp,<r4Sigma1=%rax
1806 xor %rbp,%rax
1807
1808 # qhasm: r7 += ch7
1809 # asm 1: add <ch7=int64#12,<r7=int64#11
1810 # asm 2: add <ch7=%r14,<r7=%r13
1811 add %r14,%r13
1812
1813 # qhasm: r0Sigma0 = r0>>>28
1814 # asm 1: rorx $28,<r0=int64#2,>r0Sigma0=int64#12
1815 # asm 2: rorx $28,<r0=%rsi,>r0Sigma0=%r14
1816 rorx $28,%rsi,%r14
1817
1818 # qhasm: r034 = r0>>>34
1819 # asm 1: rorx $34,<r0=int64#2,>r034=int64#13
1820 # asm 2: rorx $34,<r0=%rsi,>r034=%r15
1821 rorx $34,%rsi,%r15
1822
1823 # qhasm: X1sigma0 = X1sigma0 ^ X1right8
1824 # asm 1: vpxor <X1sigma0=reg256#7,<X1right8=reg256#8,>X1sigma0=reg256#7
1825 # asm 2: vpxor <X1sigma0=%ymm6,<X1right8=%ymm7,>X1sigma0=%ymm6
1826 vpxor %ymm6,%ymm7,%ymm6
1827
1828 # qhasm: r7 += r4Sigma1
1829 # asm 1: add <r4Sigma1=int64#7,<r7=int64#11
1830 # asm 2: add <r4Sigma1=%rax,<r7=%r13
1831 add %rax,%r13
1832
1833 # qhasm: maj7 = r2
1834 # asm 1: mov <r2=int64#3,>maj7=int64#7
1835 # asm 2: mov <r2=%rdx,>maj7=%rax
1836 mov %rdx,%rax
1837
1838 # qhasm: maj7 &= maj6
1839 # asm 1: and <maj6=int64#14,<maj7=int64#7
1840 # asm 2: and <maj6=%rbx,<maj7=%rax
1841 and %rbx,%rax
1842
1843 # qhasm: r0Sigma0 ^= r034
1844 # asm 1: xor <r034=int64#13,<r0Sigma0=int64#12
1845 # asm 2: xor <r034=%r15,<r0Sigma0=%r14
1846 xor %r15,%r14
1847
1848 # qhasm: 2x,0 W14right19 = W14 unsigned>> 19
1849 # asm 1: vpsrlq $19,<W14=reg256#9%128,>W14right19=reg256#8%128
1850 # asm 2: vpsrlq $19,<W14=%xmm8,>W14right19=%xmm7
1851 vpsrlq $19,%xmm8,%xmm7
1852
1853 # qhasm: r039 = r0>>>39
1854 # asm 1: rorx $39,<r0=int64#2,>r039=int64#13
1855 # asm 2: rorx $39,<r0=%rsi,>r039=%r15
1856 rorx $39,%rsi,%r15
1857
1858 # qhasm: 4x X1left56 = X1 << 56
1859 # asm 1: vpsllq $56,<X1=reg256#6,>X1left56=reg256#10
1860 # asm 2: vpsllq $56,<X1=%ymm5,>X1left56=%ymm9
1861 vpsllq $56,%ymm5,%ymm9
1862
1863 # qhasm: r3 += r7
1864 # asm 1: add <r7=int64#11,<r3=int64#6
1865 # asm 2: add <r7=%r13,<r3=%r9
1866 add %r13,%r9
1867
1868 # qhasm: r0Sigma0 ^= r039
1869 # asm 1: xor <r039=int64#13,<r0Sigma0=int64#12
1870 # asm 2: xor <r039=%r15,<r0Sigma0=%r14
1871 xor %r15,%r14
1872
1873 # qhasm: r6 += wc891011[1]
1874 # asm 1: addq <wc891011=stack256#3,<r6=int64#10
1875 # asm 2: addq <wc891011=72(%rsp),<r6=%r12
1876 addq 72(%rsp),%r12
1877
1878 # qhasm: r0andr1 = r1
1879 # asm 1: mov <r1=int64#4,>r0andr1=int64#13
1880 # asm 2: mov <r1=%rcx,>r0andr1=%r15
1881 mov %rcx,%r15
1882
1883 # qhasm: r0andr1 &= r0
1884 # asm 1: and <r0=int64#2,<r0andr1=int64#13
1885 # asm 2: and <r0=%rsi,<r0andr1=%r15
1886 and %rsi,%r15
1887
1888 # qhasm: 2x,0 W14left45 = W14 << 45
1889 # asm 1: vpsllq $45,<W14=reg256#9%128,>W14left45=reg256#11%128
1890 # asm 2: vpsllq $45,<W14=%xmm8,>W14left45=%xmm10
1891 vpsllq $45,%xmm8,%xmm10
1892
1893 # qhasm: r7 += r0Sigma0
1894 # asm 1: add <r0Sigma0=int64#12,<r7=int64#11
1895 # asm 2: add <r0Sigma0=%r14,<r7=%r13
1896 add %r14,%r13
1897
1898 # qhasm: maj7 ^= r0andr1
1899 # asm 1: xor <r0andr1=int64#13,<maj7=int64#7
1900 # asm 2: xor <r0andr1=%r15,<maj7=%rax
1901 xor %r15,%rax
1902
1903 # qhasm: ch6 = r5
1904 # asm 1: mov <r5=int64#8,>ch6=int64#12
1905 # asm 2: mov <r5=%r10,>ch6=%r14
1906 mov %r10,%r14
1907
1908 # qhasm: ch6 ^= r4
1909 # asm 1: xor <r4=int64#9,<ch6=int64#12
1910 # asm 2: xor <r4=%r11,<ch6=%r14
1911 xor %r11,%r14
1912
1913 # qhasm: 2x,0 W14right61 = W14 unsigned>> 61
1914 # asm 1: vpsrlq $61,<W14=reg256#9%128,>W14right61=reg256#12%128
1915 # asm 2: vpsrlq $61,<W14=%xmm8,>W14right61=%xmm11
1916 vpsrlq $61,%xmm8,%xmm11
1917
1918 # qhasm: r3Sigma1 = r3>>>14
1919 # asm 1: rorx $14,<r3=int64#6,>r3Sigma1=int64#15
1920 # asm 2: rorx $14,<r3=%r9,>r3Sigma1=%rbp
1921 rorx $14,%r9,%rbp
1922
1923 # qhasm: X1sigma0 = X1sigma0 ^ X1left56
1924 # asm 1: vpxor <X1sigma0=reg256#7,<X1left56=reg256#10,>X1sigma0=reg256#7
1925 # asm 2: vpxor <X1sigma0=%ymm6,<X1left56=%ymm9,>X1sigma0=%ymm6
1926 vpxor %ymm6,%ymm9,%ymm6
1927
1928 # qhasm: r7 += maj7
1929 # asm 1: add <maj7=int64#7,<r7=int64#11
1930 # asm 2: add <maj7=%rax,<r7=%r13
1931 add %rax,%r13
1932
1933 # qhasm: r318 = r3>>>18
1934 # asm 1: rorx $18,<r3=int64#6,>r318=int64#7
1935 # asm 2: rorx $18,<r3=%r9,>r318=%rax
1936 rorx $18,%r9,%rax
1937
1938 # qhasm: 4x X1right7 = X1 unsigned>> 7
1939 # asm 1: vpsrlq $7,<X1=reg256#6,>X1right7=reg256#6
1940 # asm 2: vpsrlq $7,<X1=%ymm5,>X1right7=%ymm5
1941 vpsrlq $7,%ymm5,%ymm5
1942
1943 # qhasm: 1x,0 W14sigma1 = W14right19 ^ W14left45
1944 # asm 1: vpxor <W14right19=reg256#8%128,<W14left45=reg256#11%128,>W14sigma1=reg256#8%128
1945 # asm 2: vpxor <W14right19=%xmm7,<W14left45=%xmm10,>W14sigma1=%xmm7
1946 vpxor %xmm7,%xmm10,%xmm7
1947
1948 # qhasm: ch6 &= r3
1949 # asm 1: and <r3=int64#6,<ch6=int64#12
1950 # asm 2: and <r3=%r9,<ch6=%r14
1951 and %r9,%r14
1952
1953 # qhasm: r3Sigma1 ^= r318
1954 # asm 1: xor <r318=int64#7,<r3Sigma1=int64#15
1955 # asm 2: xor <r318=%rax,<r3Sigma1=%rbp
1956 xor %rax,%rbp
1957
1958 # qhasm: r341 = r3>>>41
1959 # asm 1: rorx $41,<r3=int64#6,>r341=int64#7
1960 # asm 2: rorx $41,<r3=%r9,>r341=%rax
1961 rorx $41,%r9,%rax
1962
1963 # qhasm: maj6 &= r7
1964 # asm 1: and <r7=int64#11,<maj6=int64#14
1965 # asm 2: and <r7=%r13,<maj6=%rbx
1966 and %r13,%rbx
1967
1968 # qhasm: X1sigma0 = X1sigma0 ^ X1right7
1969 # asm 1: vpxor <X1sigma0=reg256#7,<X1right7=reg256#6,>X1sigma0=reg256#6
1970 # asm 2: vpxor <X1sigma0=%ymm6,<X1right7=%ymm5,>X1sigma0=%ymm5
1971 vpxor %ymm6,%ymm5,%ymm5
1972
1973 # qhasm: 1x,0 W14sigma1 ^= W14right61
1974 # asm 1: vpxor <W14right61=reg256#12%128,<W14sigma1=reg256#8%128,<W14sigma1=reg256#8%128
1975 # asm 2: vpxor <W14right61=%xmm11,<W14sigma1=%xmm7,<W14sigma1=%xmm7
1976 vpxor %xmm11,%xmm7,%xmm7
1977
1978 # qhasm: 4x X0 = X0 + X1sigma0
1979 # asm 1: vpaddq <X0=reg256#2,<X1sigma0=reg256#6,>X0=reg256#2
1980 # asm 2: vpaddq <X0=%ymm1,<X1sigma0=%ymm5,>X0=%ymm1
1981 vpaddq %ymm1,%ymm5,%ymm1
1982
1983 # qhasm: r3Sigma1 ^= r341
1984 # asm 1: xor <r341=int64#7,<r3Sigma1=int64#15
1985 # asm 2: xor <r341=%rax,<r3Sigma1=%rbp
1986 xor %rax,%rbp
1987
1988 # qhasm: maj6 ^= r0andr1
1989 # asm 1: xor <r0andr1=int64#13,<maj6=int64#14
1990 # asm 2: xor <r0andr1=%r15,<maj6=%rbx
1991 xor %r15,%rbx
1992
1993 # qhasm: 2x,0 W14left3 = W14 << 3
1994 # asm 1: vpsllq $3,<W14=reg256#9%128,>W14left3=reg256#6%128
1995 # asm 2: vpsllq $3,<W14=%xmm8,>W14left3=%xmm5
1996 vpsllq $3,%xmm8,%xmm5
1997
1998 # qhasm: r7Sigma0 = r7>>>28
1999 # asm 1: rorx $28,<r7=int64#11,>r7Sigma0=int64#7
2000 # asm 2: rorx $28,<r7=%r13,>r7Sigma0=%rax
2001 rorx $28,%r13,%rax
2002
2003 # qhasm: ch6 ^= r5
2004 # asm 1: xor <r5=int64#8,<ch6=int64#12
2005 # asm 2: xor <r5=%r10,<ch6=%r14
2006 xor %r10,%r14
2007
2008 # qhasm: r6 += r3Sigma1
2009 # asm 1: add <r3Sigma1=int64#15,<r6=int64#10
2010 # asm 2: add <r3Sigma1=%rbp,<r6=%r12
2011 add %rbp,%r12
2012
2013 # qhasm: 4x X0 = X0 + mem256[&w + 72]
2014 # asm 1: vpaddq <w=stack1280#1,<X0=reg256#2,>X0=reg256#2
2015 # asm 2: vpaddq <w=232(%rsp),<X0=%ymm1,>X0=%ymm1
2016 vpaddq 232(%rsp),%ymm1,%ymm1
2017
2018 # qhasm: r734 = r7>>>34
2019 # asm 1: rorx $34,<r7=int64#11,>r734=int64#13
2020 # asm 2: rorx $34,<r7=%r13,>r734=%r15
2021 rorx $34,%r13,%r15
2022
2023 # qhasm: r5 += wc891011[2]
2024 # asm 1: addq <wc891011=stack256#3,<r5=int64#8
2025 # asm 2: addq <wc891011=80(%rsp),<r5=%r10
2026 addq 80(%rsp),%r10
2027
2028 # qhasm: r6 += ch6
2029 # asm 1: add <ch6=int64#12,<r6=int64#10
2030 # asm 2: add <ch6=%r14,<r6=%r12
2031 add %r14,%r12
2032
2033 # qhasm: r7Sigma0 ^= r734
2034 # asm 1: xor <r734=int64#13,<r7Sigma0=int64#7
2035 # asm 2: xor <r734=%r15,<r7Sigma0=%rax
2036 xor %r15,%rax
2037
2038 # qhasm: r739 = r7>>>39
2039 # asm 1: rorx $39,<r7=int64#11,>r739=int64#12
2040 # asm 2: rorx $39,<r7=%r13,>r739=%r14
2041 rorx $39,%r13,%r14
2042
2043 # qhasm: r2 += r6
2044 # asm 1: add <r6=int64#10,<r2=int64#3
2045 # asm 2: add <r6=%r12,<r2=%rdx
2046 add %r12,%rdx
2047
2048 # qhasm: 1x,0 W14sigma1 ^= W14left3
2049 # asm 1: vpxor <W14left3=reg256#6%128,<W14sigma1=reg256#8%128,<W14sigma1=reg256#8%128
2050 # asm 2: vpxor <W14left3=%xmm5,<W14sigma1=%xmm7,<W14sigma1=%xmm7
2051 vpxor %xmm5,%xmm7,%xmm7
2052
2053 # qhasm: 2x,0 W14right6 = W14 unsigned>> 6
2054 # asm 1: vpsrlq $6,<W14=reg256#9%128,>W14right6=reg256#6%128
2055 # asm 2: vpsrlq $6,<W14=%xmm8,>W14right6=%xmm5
2056 vpsrlq $6,%xmm8,%xmm5
2057
2058 # qhasm: r7Sigma0 ^= r739
2059 # asm 1: xor <r739=int64#12,<r7Sigma0=int64#7
2060 # asm 2: xor <r739=%r14,<r7Sigma0=%rax
2061 xor %r14,%rax
2062
2063 # qhasm: r6 += maj6
2064 # asm 1: add <maj6=int64#14,<r6=int64#10
2065 # asm 2: add <maj6=%rbx,<r6=%r12
2066 add %rbx,%r12
2067
2068 # qhasm: ch5 = r4
2069 # asm 1: mov <r4=int64#9,>ch5=int64#12
2070 # asm 2: mov <r4=%r11,>ch5=%r14
2071 mov %r11,%r14
2072
2073 # qhasm: ch5 ^= r3
2074 # asm 1: xor <r3=int64#6,<ch5=int64#12
2075 # asm 2: xor <r3=%r9,<ch5=%r14
2076 xor %r9,%r14
2077
2078 # qhasm: r6 += r7Sigma0
2079 # asm 1: add <r7Sigma0=int64#7,<r6=int64#10
2080 # asm 2: add <r7Sigma0=%rax,<r6=%r12
2081 add %rax,%r12
2082
2083 # qhasm: r2Sigma1 = r2>>>14
2084 # asm 1: rorx $14,<r2=int64#3,>r2Sigma1=int64#7
2085 # asm 2: rorx $14,<r2=%rdx,>r2Sigma1=%rax
2086 rorx $14,%rdx,%rax
2087
2088 # qhasm: 1x,0 W14sigma1 ^= W14right6
2089 # asm 1: vpxor <W14right6=reg256#6%128,<W14sigma1=reg256#8%128,<W14sigma1=reg256#8%128
2090 # asm 2: vpxor <W14right6=%xmm5,<W14sigma1=%xmm7,<W14sigma1=%xmm7
2091 vpxor %xmm5,%xmm7,%xmm7
2092
2093 # qhasm: ch5 &= r2
2094 # asm 1: and <r2=int64#3,<ch5=int64#12
2095 # asm 2: and <r2=%rdx,<ch5=%r14
2096 and %rdx,%r14
2097
2098 # qhasm: r218 = r2>>>18
2099 # asm 1: rorx $18,<r2=int64#3,>r218=int64#13
2100 # asm 2: rorx $18,<r2=%rdx,>r218=%r15
2101 rorx $18,%rdx,%r15
2102
2103 # qhasm: r241 = r2>>>41
2104 # asm 1: rorx $41,<r2=int64#3,>r241=int64#14
2105 # asm 2: rorx $41,<r2=%rdx,>r241=%rbx
2106 rorx $41,%rdx,%rbx
2107
2108 # qhasm: 4x X0 = W14sigma1 + X0
2109 # asm 1: vpaddq <W14sigma1=reg256#8,<X0=reg256#2,>X0=reg256#2
2110 # asm 2: vpaddq <W14sigma1=%ymm7,<X0=%ymm1,>X0=%ymm1
2111 vpaddq %ymm7,%ymm1,%ymm1
2112
2113 # qhasm: ch5 ^= r4
2114 # asm 1: xor <r4=int64#9,<ch5=int64#12
2115 # asm 2: xor <r4=%r11,<ch5=%r14
2116 xor %r11,%r14
2117
2118 # qhasm: r2Sigma1 ^= r218
2119 # asm 1: xor <r218=int64#13,<r2Sigma1=int64#7
2120 # asm 2: xor <r218=%r15,<r2Sigma1=%rax
2121 xor %r15,%rax
2122
2123 # qhasm: 2x,0 W0right19 = X0 unsigned>> 19
2124 # asm 1: vpsrlq $19,<X0=reg256#2%128,>W0right19=reg256#6%128
2125 # asm 2: vpsrlq $19,<X0=%xmm1,>W0right19=%xmm5
2126 vpsrlq $19,%xmm1,%xmm5
2127
2128 # qhasm: r6Sigma0 = r6>>>28
2129 # asm 1: rorx $28,<r6=int64#10,>r6Sigma0=int64#13
2130 # asm 2: rorx $28,<r6=%r12,>r6Sigma0=%r15
2131 rorx $28,%r12,%r15
2132
2133 # qhasm: r5 += ch5
2134 # asm 1: add <ch5=int64#12,<r5=int64#8
2135 # asm 2: add <ch5=%r14,<r5=%r10
2136 add %r14,%r10
2137
2138 # qhasm: 2x,0 W0left45 = X0 << 45
2139 # asm 1: vpsllq $45,<X0=reg256#2%128,>W0left45=reg256#7%128
2140 # asm 2: vpsllq $45,<X0=%xmm1,>W0left45=%xmm6
2141 vpsllq $45,%xmm1,%xmm6
2142
2143 # qhasm: r634 = r6>>>34
2144 # asm 1: rorx $34,<r6=int64#10,>r634=int64#12
2145 # asm 2: rorx $34,<r6=%r12,>r634=%r14
2146 rorx $34,%r12,%r14
2147
2148 # qhasm: r2Sigma1 ^= r241
2149 # asm 1: xor <r241=int64#14,<r2Sigma1=int64#7
2150 # asm 2: xor <r241=%rbx,<r2Sigma1=%rax
2151 xor %rbx,%rax
2152
2153 # qhasm: maj4 = r7
2154 # asm 1: mov <r7=int64#11,>maj4=int64#14
2155 # asm 2: mov <r7=%r13,>maj4=%rbx
2156 mov %r13,%rbx
2157
2158 # qhasm: X5 = mem256[&w + 40]
2159 # asm 1: vmovupd <w=stack1280#1,>X5=reg256#8
2160 # asm 2: vmovupd <w=200(%rsp),>X5=%ymm7
2161 vmovupd 200(%rsp),%ymm7
2162
2163 # qhasm: maj4 ^= r6
2164 # asm 1: xor <r6=int64#10,<maj4=int64#14
2165 # asm 2: xor <r6=%r12,<maj4=%rbx
2166 xor %r12,%rbx
2167
2168 # qhasm: 2x,0 W0right61 = X0 unsigned>> 61
2169 # asm 1: vpsrlq $61,<X0=reg256#2%128,>W0right61=reg256#9%128
2170 # asm 2: vpsrlq $61,<X0=%xmm1,>W0right61=%xmm8
2171 vpsrlq $61,%xmm1,%xmm8
2172
2173 # qhasm: 1x,0 W0sigma1 = W0right19 ^ W0left45
2174 # asm 1: vpxor <W0right19=reg256#6%128,<W0left45=reg256#7%128,>W0sigma1=reg256#6%128
2175 # asm 2: vpxor <W0right19=%xmm5,<W0left45=%xmm6,>W0sigma1=%xmm5
2176 vpxor %xmm5,%xmm6,%xmm5
2177
2178 # qhasm: r6Sigma0 ^= r634
2179 # asm 1: xor <r634=int64#12,<r6Sigma0=int64#13
2180 # asm 2: xor <r634=%r14,<r6Sigma0=%r15
2181 xor %r14,%r15
2182
2183 # qhasm: 1x,0 W0sigma1 ^= W0right61
2184 # asm 1: vpxor <W0right61=reg256#9%128,<W0sigma1=reg256#6%128,<W0sigma1=reg256#6%128
2185 # asm 2: vpxor <W0right61=%xmm8,<W0sigma1=%xmm5,<W0sigma1=%xmm5
2186 vpxor %xmm8,%xmm5,%xmm5
2187
2188 # qhasm: r639 = r6>>>39
2189 # asm 1: rorx $39,<r6=int64#10,>r639=int64#12
2190 # asm 2: rorx $39,<r6=%r12,>r639=%r14
2191 rorx $39,%r12,%r14
2192
2193 # qhasm: 2x,0 W0left3 = X0 << 3
2194 # asm 1: vpsllq $3,<X0=reg256#2%128,>W0left3=reg256#7%128
2195 # asm 2: vpsllq $3,<X0=%xmm1,>W0left3=%xmm6
2196 vpsllq $3,%xmm1,%xmm6
2197
2198 # qhasm: r6Sigma0 ^= r639
2199 # asm 1: xor <r639=int64#12,<r6Sigma0=int64#13
2200 # asm 2: xor <r639=%r14,<r6Sigma0=%r15
2201 xor %r14,%r15
2202
2203 # qhasm: r5 += r2Sigma1
2204 # asm 1: add <r2Sigma1=int64#7,<r5=int64#8
2205 # asm 2: add <r2Sigma1=%rax,<r5=%r10
2206 add %rax,%r10
2207
2208 # qhasm: 2x,0 W0right6 = X0 unsigned>> 6
2209 # asm 1: vpsrlq $6,<X0=reg256#2%128,>W0right6=reg256#9%128
2210 # asm 2: vpsrlq $6,<X0=%xmm1,>W0right6=%xmm8
2211 vpsrlq $6,%xmm1,%xmm8
2212
2213 # qhasm: 1x,0 W0sigma1 ^= W0left3
2214 # asm 1: vpxor <W0left3=reg256#7%128,<W0sigma1=reg256#6%128,<W0sigma1=reg256#6%128
2215 # asm 2: vpxor <W0left3=%xmm6,<W0sigma1=%xmm5,<W0sigma1=%xmm5
2216 vpxor %xmm6,%xmm5,%xmm5
2217
2218 # qhasm: r1 += r5
2219 # asm 1: add <r5=int64#8,<r1=int64#4
2220 # asm 2: add <r5=%r10,<r1=%rcx
2221 add %r10,%rcx
2222
2223 # qhasm: r6andr7 = r7
2224 # asm 1: mov <r7=int64#11,>r6andr7=int64#7
2225 # asm 2: mov <r7=%r13,>r6andr7=%rax
2226 mov %r13,%rax
2227
2228 # qhasm: r6andr7 &= r6
2229 # asm 1: and <r6=int64#10,<r6andr7=int64#7
2230 # asm 2: and <r6=%r12,<r6andr7=%rax
2231 and %r12,%rax
2232
2233 # qhasm: r1Sigma1 = r1>>>14
2234 # asm 1: rorx $14,<r1=int64#4,>r1Sigma1=int64#12
2235 # asm 2: rorx $14,<r1=%rcx,>r1Sigma1=%r14
2236 rorx $14,%rcx,%r14
2237
2238 # qhasm: r5 += r6Sigma0
2239 # asm 1: add <r6Sigma0=int64#13,<r5=int64#8
2240 # asm 2: add <r6Sigma0=%r15,<r5=%r10
2241 add %r15,%r10
2242
2243 # qhasm: 1x,0 W0sigma1 ^= W0right6
2244 # asm 1: vpxor <W0right6=reg256#9%128,<W0sigma1=reg256#6%128,<W0sigma1=reg256#6%128
2245 # asm 2: vpxor <W0right6=%xmm8,<W0sigma1=%xmm5,<W0sigma1=%xmm5
2246 vpxor %xmm8,%xmm5,%xmm5
2247
2248 # qhasm: maj5 = r0
2249 # asm 1: mov <r0=int64#2,>maj5=int64#13
2250 # asm 2: mov <r0=%rsi,>maj5=%r15
2251 mov %rsi,%r15
2252
2253 # qhasm: maj5 &= maj4
2254 # asm 1: and <maj4=int64#14,<maj5=int64#13
2255 # asm 2: and <maj4=%rbx,<maj5=%r15
2256 and %rbx,%r15
2257
2258 # qhasm: W0sigma1 = W0sigma1[2,3,0,1]
2259 # asm 1: vpermq $0x4e,<W0sigma1=reg256#6,>W0sigma1=reg256#6
2260 # asm 2: vpermq $0x4e,<W0sigma1=%ymm5,>W0sigma1=%ymm5
2261 vpermq $0x4e,%ymm5,%ymm5
2262
2263 # qhasm: maj5 ^= r6andr7
2264 # asm 1: xor <r6andr7=int64#7,<maj5=int64#13
2265 # asm 2: xor <r6andr7=%rax,<maj5=%r15
2266 xor %rax,%r15
2267
2268 # qhasm: ch4 = r3
2269 # asm 1: mov <r3=int64#6,>ch4=int64#15
2270 # asm 2: mov <r3=%r9,>ch4=%rbp
2271 mov %r9,%rbp
2272
2273 # qhasm: 4x X5right1 = X5 unsigned>> 1
2274 # asm 1: vpsrlq $1,<X5=reg256#8,>X5right1=reg256#7
2275 # asm 2: vpsrlq $1,<X5=%ymm7,>X5right1=%ymm6
2276 vpsrlq $1,%ymm7,%ymm6
2277
2278 # qhasm: ch4 ^= r2
2279 # asm 1: xor <r2=int64#3,<ch4=int64#15
2280 # asm 2: xor <r2=%rdx,<ch4=%rbp
2281 xor %rdx,%rbp
2282
2283 # qhasm: r5 += maj5
2284 # asm 1: add <maj5=int64#13,<r5=int64#8
2285 # asm 2: add <maj5=%r15,<r5=%r10
2286 add %r15,%r10
2287
2288 # qhasm: r118 = r1>>>18
2289 # asm 1: rorx $18,<r1=int64#4,>r118=int64#13
2290 # asm 2: rorx $18,<r1=%rcx,>r118=%r15
2291 rorx $18,%rcx,%r15
2292
2293 # qhasm: ch4 &= r1
2294 # asm 1: and <r1=int64#4,<ch4=int64#15
2295 # asm 2: and <r1=%rcx,<ch4=%rbp
2296 and %rcx,%rbp
2297
2298 # qhasm: maj4 &= r5
2299 # asm 1: and <r5=int64#8,<maj4=int64#14
2300 # asm 2: and <r5=%r10,<maj4=%rbx
2301 and %r10,%rbx
2302
2303 # qhasm: ch4 ^= r3
2304 # asm 1: xor <r3=int64#6,<ch4=int64#15
2305 # asm 2: xor <r3=%r9,<ch4=%rbp
2306 xor %r9,%rbp
2307
2308 # qhasm: r4 += wc891011[3]
2309 # asm 1: addq <wc891011=stack256#3,<r4=int64#9
2310 # asm 2: addq <wc891011=88(%rsp),<r4=%r11
2311 addq 88(%rsp),%r11
2312
2313 # qhasm: r1Sigma1 ^= r118
2314 # asm 1: xor <r118=int64#13,<r1Sigma1=int64#12
2315 # asm 2: xor <r118=%r15,<r1Sigma1=%r14
2316 xor %r15,%r14
2317
2318 # qhasm: r141 = r1>>>41
2319 # asm 1: rorx $41,<r1=int64#4,>r141=int64#13
2320 # asm 2: rorx $41,<r1=%rcx,>r141=%r15
2321 rorx $41,%rcx,%r15
2322
2323 # qhasm: 4x X0 = X0 + W0sigma1
2324 # asm 1: vpaddq <X0=reg256#2,<W0sigma1=reg256#6,>X0=reg256#2
2325 # asm 2: vpaddq <X0=%ymm1,<W0sigma1=%ymm5,>X0=%ymm1
2326 vpaddq %ymm1,%ymm5,%ymm1
2327
2328 # qhasm: r4 += ch4
2329 # asm 1: add <ch4=int64#15,<r4=int64#9
2330 # asm 2: add <ch4=%rbp,<r4=%r11
2331 add %rbp,%r11
2332
2333 # qhasm: maj4 ^= r6andr7
2334 # asm 1: xor <r6andr7=int64#7,<maj4=int64#14
2335 # asm 2: xor <r6andr7=%rax,<maj4=%rbx
2336 xor %rax,%rbx
2337
2338 # qhasm: r5Sigma0 = r5>>>28
2339 # asm 1: rorx $28,<r5=int64#8,>r5Sigma0=int64#7
2340 # asm 2: rorx $28,<r5=%r10,>r5Sigma0=%rax
2341 rorx $28,%r10,%rax
2342
2343 # qhasm: 4x D0 = X0 + mem256[constants + 128]
2344 # asm 1: vpaddq 128(<constants=int64#5),<X0=reg256#2,>D0=reg256#6
2345 # asm 2: vpaddq 128(<constants=%r8),<X0=%ymm1,>D0=%ymm5
2346 vpaddq 128(%r8),%ymm1,%ymm5
2347
2348 # qhasm: r1Sigma1 ^= r141
2349 # asm 1: xor <r141=int64#13,<r1Sigma1=int64#12
2350 # asm 2: xor <r141=%r15,<r1Sigma1=%r14
2351 xor %r15,%r14
2352
2353 # qhasm: r534 = r5>>>34
2354 # asm 1: rorx $34,<r5=int64#8,>r534=int64#13
2355 # asm 2: rorx $34,<r5=%r10,>r534=%r15
2356 rorx $34,%r10,%r15
2357
2358 # qhasm: mem256[&w + 128] = X0
2359 # asm 1: vmovupd <X0=reg256#2,<w=stack1280#1
2360 # asm 2: vmovupd <X0=%ymm1,<w=288(%rsp)
2361 vmovupd %ymm1,288(%rsp)
2362
2363 # qhasm: r4 += r1Sigma1
2364 # asm 1: add <r1Sigma1=int64#12,<r4=int64#9
2365 # asm 2: add <r1Sigma1=%r14,<r4=%r11
2366 add %r14,%r11
2367
2368 # qhasm: r5Sigma0 ^= r534
2369 # asm 1: xor <r534=int64#13,<r5Sigma0=int64#7
2370 # asm 2: xor <r534=%r15,<r5Sigma0=%rax
2371 xor %r15,%rax
2372
2373 # qhasm: r539 = r5>>>39
2374 # asm 1: rorx $39,<r5=int64#8,>r539=int64#12
2375 # asm 2: rorx $39,<r5=%r10,>r539=%r14
2376 rorx $39,%r10,%r14
2377
2378 # qhasm: r3 += wc12131415[0]
2379 # asm 1: addq <wc12131415=stack256#5,<r3=int64#6
2380 # asm 2: addq <wc12131415=128(%rsp),<r3=%r9
2381 addq 128(%rsp),%r9
2382
2383 # qhasm: mem256[&w + 0] = X0
2384 # asm 1: vmovupd <X0=reg256#2,<w=stack1280#1
2385 # asm 2: vmovupd <X0=%ymm1,<w=160(%rsp)
2386 vmovupd %ymm1,160(%rsp)
2387
2388 # qhasm: r0 += r4
2389 # asm 1: add <r4=int64#9,<r0=int64#2
2390 # asm 2: add <r4=%r11,<r0=%rsi
2391 add %r11,%rsi
2392
2393 # qhasm: r4 += maj4
2394 # asm 1: add <maj4=int64#14,<r4=int64#9
2395 # asm 2: add <maj4=%rbx,<r4=%r11
2396 add %rbx,%r11
2397
2398 # qhasm: r5Sigma0 ^= r539
2399 # asm 1: xor <r539=int64#12,<r5Sigma0=int64#7
2400 # asm 2: xor <r539=%r14,<r5Sigma0=%rax
2401 xor %r14,%rax
2402
2403 # qhasm: constants += 128
2404 # asm 1: add $128,<constants=int64#5
2405 # asm 2: add $128,<constants=%r8
2406 add $128,%r8
2407
2408 # qhasm: wc0123 = D0
2409 # asm 1: vmovapd <D0=reg256#6,>wc0123=stack256#3
2410 # asm 2: vmovapd <D0=%ymm5,>wc0123=64(%rsp)
2411 vmovapd %ymm5,64(%rsp)
2412
2413 # qhasm: W2 = mem128[&w + 16],0
2414 # asm 1: vmovupd <w=stack1280#1,>W2=reg256#6%128
2415 # asm 2: vmovupd <w=176(%rsp),>W2=%xmm5
2416 vmovupd 176(%rsp),%xmm5
2417
2418 # qhasm: r4 += r5Sigma0
2419 # asm 1: add <r5Sigma0=int64#7,<r4=int64#9
2420 # asm 2: add <r5Sigma0=%rax,<r4=%r11
2421 add %rax,%r11
2422
2423 # qhasm: r0Sigma1 = r0>>>14
2424 # asm 1: rorx $14,<r0=int64#2,>r0Sigma1=int64#7
2425 # asm 2: rorx $14,<r0=%rsi,>r0Sigma1=%rax
2426 rorx $14,%rsi,%rax
2427
2428 # qhasm: ch3 = r2
2429 # asm 1: mov <r2=int64#3,>ch3=int64#12
2430 # asm 2: mov <r2=%rdx,>ch3=%r14
2431 mov %rdx,%r14
2432
2433 # qhasm: ch3 ^= r1
2434 # asm 1: xor <r1=int64#4,<ch3=int64#12
2435 # asm 2: xor <r1=%rcx,<ch3=%r14
2436 xor %rcx,%r14
2437
2438 # qhasm: 4x X5left63 = X5 << 63
2439 # asm 1: vpsllq $63,<X5=reg256#8,>X5left63=reg256#9
2440 # asm 2: vpsllq $63,<X5=%ymm7,>X5left63=%ymm8
2441 vpsllq $63,%ymm7,%ymm8
2442
2443 # qhasm: r018 = r0>>>18
2444 # asm 1: rorx $18,<r0=int64#2,>r018=int64#13
2445 # asm 2: rorx $18,<r0=%rsi,>r018=%r15
2446 rorx $18,%rsi,%r15
2447
2448 # qhasm: maj2 = r5
2449 # asm 1: mov <r5=int64#8,>maj2=int64#14
2450 # asm 2: mov <r5=%r10,>maj2=%rbx
2451 mov %r10,%rbx
2452
2453 # qhasm: ch3 &= r0
2454 # asm 1: and <r0=int64#2,<ch3=int64#12
2455 # asm 2: and <r0=%rsi,<ch3=%r14
2456 and %rsi,%r14
2457
2458 # qhasm: maj2 ^= r4
2459 # asm 1: xor <r4=int64#9,<maj2=int64#14
2460 # asm 2: xor <r4=%r11,<maj2=%rbx
2461 xor %r11,%rbx
2462
2463 # qhasm: r041 = r0>>>41
2464 # asm 1: rorx $41,<r0=int64#2,>r041=int64#15
2465 # asm 2: rorx $41,<r0=%rsi,>r041=%rbp
2466 rorx $41,%rsi,%rbp
2467
2468 # qhasm: X5sigma0 = X5right1 ^ X5left63
2469 # asm 1: vpxor <X5right1=reg256#7,<X5left63=reg256#9,>X5sigma0=reg256#7
2470 # asm 2: vpxor <X5right1=%ymm6,<X5left63=%ymm8,>X5sigma0=%ymm6
2471 vpxor %ymm6,%ymm8,%ymm6
2472
2473 # qhasm: r0Sigma1 ^= r018
2474 # asm 1: xor <r018=int64#13,<r0Sigma1=int64#7
2475 # asm 2: xor <r018=%r15,<r0Sigma1=%rax
2476 xor %r15,%rax
2477
2478 # qhasm: ch3 ^= r2
2479 # asm 1: xor <r2=int64#3,<ch3=int64#12
2480 # asm 2: xor <r2=%rdx,<ch3=%r14
2481 xor %rdx,%r14
2482
2483 # qhasm: 4x X5right8 = X5 unsigned>> 8
2484 # asm 1: vpsrlq $8,<X5=reg256#8,>X5right8=reg256#9
2485 # asm 2: vpsrlq $8,<X5=%ymm7,>X5right8=%ymm8
2486 vpsrlq $8,%ymm7,%ymm8
2487
2488 # qhasm: r4Sigma0 = r4>>>28
2489 # asm 1: rorx $28,<r4=int64#9,>r4Sigma0=int64#13
2490 # asm 2: rorx $28,<r4=%r11,>r4Sigma0=%r15
2491 rorx $28,%r11,%r15
2492
2493 # qhasm: r0Sigma1 ^= r041
2494 # asm 1: xor <r041=int64#15,<r0Sigma1=int64#7
2495 # asm 2: xor <r041=%rbp,<r0Sigma1=%rax
2496 xor %rbp,%rax
2497
2498 # qhasm: r3 += ch3
2499 # asm 1: add <ch3=int64#12,<r3=int64#6
2500 # asm 2: add <ch3=%r14,<r3=%r9
2501 add %r14,%r9
2502
2503 # qhasm: r434 = r4>>>34
2504 # asm 1: rorx $34,<r4=int64#9,>r434=int64#12
2505 # asm 2: rorx $34,<r4=%r11,>r434=%r14
2506 rorx $34,%r11,%r14
2507
2508 # qhasm: X5sigma0 = X5sigma0 ^ X5right8
2509 # asm 1: vpxor <X5sigma0=reg256#7,<X5right8=reg256#9,>X5sigma0=reg256#7
2510 # asm 2: vpxor <X5sigma0=%ymm6,<X5right8=%ymm8,>X5sigma0=%ymm6
2511 vpxor %ymm6,%ymm8,%ymm6
2512
2513 # qhasm: r3 += r0Sigma1
2514 # asm 1: add <r0Sigma1=int64#7,<r3=int64#6
2515 # asm 2: add <r0Sigma1=%rax,<r3=%r9
2516 add %rax,%r9
2517
2518 # qhasm: maj3 = r6
2519 # asm 1: mov <r6=int64#10,>maj3=int64#7
2520 # asm 2: mov <r6=%r12,>maj3=%rax
2521 mov %r12,%rax
2522
2523 # qhasm: maj3 &= maj2
2524 # asm 1: and <maj2=int64#14,<maj3=int64#7
2525 # asm 2: and <maj2=%rbx,<maj3=%rax
2526 and %rbx,%rax
2527
2528 # qhasm: r4Sigma0 ^= r434
2529 # asm 1: xor <r434=int64#12,<r4Sigma0=int64#13
2530 # asm 2: xor <r434=%r14,<r4Sigma0=%r15
2531 xor %r14,%r15
2532
2533 # qhasm: r439 = r4>>>39
2534 # asm 1: rorx $39,<r4=int64#9,>r439=int64#12
2535 # asm 2: rorx $39,<r4=%r11,>r439=%r14
2536 rorx $39,%r11,%r14
2537
2538 # qhasm: 2x,0 W2right19 = W2 unsigned>> 19
2539 # asm 1: vpsrlq $19,<W2=reg256#6%128,>W2right19=reg256#9%128
2540 # asm 2: vpsrlq $19,<W2=%xmm5,>W2right19=%xmm8
2541 vpsrlq $19,%xmm5,%xmm8
2542
2543 # qhasm: 4x X5left56 = X5 << 56
2544 # asm 1: vpsllq $56,<X5=reg256#8,>X5left56=reg256#10
2545 # asm 2: vpsllq $56,<X5=%ymm7,>X5left56=%ymm9
2546 vpsllq $56,%ymm7,%ymm9
2547
2548 # qhasm: r7 += r3
2549 # asm 1: add <r3=int64#6,<r7=int64#11
2550 # asm 2: add <r3=%r9,<r7=%r13
2551 add %r9,%r13
2552
2553 # qhasm: r4Sigma0 ^= r439
2554 # asm 1: xor <r439=int64#12,<r4Sigma0=int64#13
2555 # asm 2: xor <r439=%r14,<r4Sigma0=%r15
2556 xor %r14,%r15
2557
2558 # qhasm: r2 += wc12131415[1]
2559 # asm 1: addq <wc12131415=stack256#5,<r2=int64#3
2560 # asm 2: addq <wc12131415=136(%rsp),<r2=%rdx
2561 addq 136(%rsp),%rdx
2562
2563 # qhasm: r4andr5 = r5
2564 # asm 1: mov <r5=int64#8,>r4andr5=int64#12
2565 # asm 2: mov <r5=%r10,>r4andr5=%r14
2566 mov %r10,%r14
2567
2568 # qhasm: r4andr5 &= r4
2569 # asm 1: and <r4=int64#9,<r4andr5=int64#12
2570 # asm 2: and <r4=%r11,<r4andr5=%r14
2571 and %r11,%r14
2572
2573 # qhasm: 2x,0 W2left45 = W2 << 45
2574 # asm 1: vpsllq $45,<W2=reg256#6%128,>W2left45=reg256#11%128
2575 # asm 2: vpsllq $45,<W2=%xmm5,>W2left45=%xmm10
2576 vpsllq $45,%xmm5,%xmm10
2577
2578 # qhasm: r3 += r4Sigma0
2579 # asm 1: add <r4Sigma0=int64#13,<r3=int64#6
2580 # asm 2: add <r4Sigma0=%r15,<r3=%r9
2581 add %r15,%r9
2582
2583 # qhasm: maj3 ^= r4andr5
2584 # asm 1: xor <r4andr5=int64#12,<maj3=int64#7
2585 # asm 2: xor <r4andr5=%r14,<maj3=%rax
2586 xor %r14,%rax
2587
2588 # qhasm: ch2 = r1
2589 # asm 1: mov <r1=int64#4,>ch2=int64#13
2590 # asm 2: mov <r1=%rcx,>ch2=%r15
2591 mov %rcx,%r15
2592
2593 # qhasm: ch2 ^= r0
2594 # asm 1: xor <r0=int64#2,<ch2=int64#13
2595 # asm 2: xor <r0=%rsi,<ch2=%r15
2596 xor %rsi,%r15
2597
2598 # qhasm: 2x,0 W2right61 = W2 unsigned>> 61
2599 # asm 1: vpsrlq $61,<W2=reg256#6%128,>W2right61=reg256#12%128
2600 # asm 2: vpsrlq $61,<W2=%xmm5,>W2right61=%xmm11
2601 vpsrlq $61,%xmm5,%xmm11
2602
2603 # qhasm: r7Sigma1 = r7>>>14
2604 # asm 1: rorx $14,<r7=int64#11,>r7Sigma1=int64#15
2605 # asm 2: rorx $14,<r7=%r13,>r7Sigma1=%rbp
2606 rorx $14,%r13,%rbp
2607
2608 # qhasm: X5sigma0 = X5sigma0 ^ X5left56
2609 # asm 1: vpxor <X5sigma0=reg256#7,<X5left56=reg256#10,>X5sigma0=reg256#7
2610 # asm 2: vpxor <X5sigma0=%ymm6,<X5left56=%ymm9,>X5sigma0=%ymm6
2611 vpxor %ymm6,%ymm9,%ymm6
2612
2613 # qhasm: r3 += maj3
2614 # asm 1: add <maj3=int64#7,<r3=int64#6
2615 # asm 2: add <maj3=%rax,<r3=%r9
2616 add %rax,%r9
2617
2618 # qhasm: 4x X5right7 = X5 unsigned>> 7
2619 # asm 1: vpsrlq $7,<X5=reg256#8,>X5right7=reg256#8
2620 # asm 2: vpsrlq $7,<X5=%ymm7,>X5right7=%ymm7
2621 vpsrlq $7,%ymm7,%ymm7
2622
2623 # qhasm: r718 = r7>>>18
2624 # asm 1: rorx $18,<r7=int64#11,>r718=int64#7
2625 # asm 2: rorx $18,<r7=%r13,>r718=%rax
2626 rorx $18,%r13,%rax
2627
2628 # qhasm: 1x,0 W2sigma1 = W2right19 ^ W2left45
2629 # asm 1: vpxor <W2right19=reg256#9%128,<W2left45=reg256#11%128,>W2sigma1=reg256#9%128
2630 # asm 2: vpxor <W2right19=%xmm8,<W2left45=%xmm10,>W2sigma1=%xmm8
2631 vpxor %xmm8,%xmm10,%xmm8
2632
2633 # qhasm: ch2 &= r7
2634 # asm 1: and <r7=int64#11,<ch2=int64#13
2635 # asm 2: and <r7=%r13,<ch2=%r15
2636 and %r13,%r15
2637
2638 # qhasm: r7Sigma1 ^= r718
2639 # asm 1: xor <r718=int64#7,<r7Sigma1=int64#15
2640 # asm 2: xor <r718=%rax,<r7Sigma1=%rbp
2641 xor %rax,%rbp
2642
2643 # qhasm: X5sigma0 = X5sigma0 ^ X5right7
2644 # asm 1: vpxor <X5sigma0=reg256#7,<X5right7=reg256#8,>X5sigma0=reg256#7
2645 # asm 2: vpxor <X5sigma0=%ymm6,<X5right7=%ymm7,>X5sigma0=%ymm6
2646 vpxor %ymm6,%ymm7,%ymm6
2647
2648 # qhasm: r741 = r7>>>41
2649 # asm 1: rorx $41,<r7=int64#11,>r741=int64#7
2650 # asm 2: rorx $41,<r7=%r13,>r741=%rax
2651 rorx $41,%r13,%rax
2652
2653 # qhasm: maj2 &= r3
2654 # asm 1: and <r3=int64#6,<maj2=int64#14
2655 # asm 2: and <r3=%r9,<maj2=%rbx
2656 and %r9,%rbx
2657
2658 # qhasm: 1x,0 W2sigma1 ^= W2right61
2659 # asm 1: vpxor <W2right61=reg256#12%128,<W2sigma1=reg256#9%128,<W2sigma1=reg256#9%128
2660 # asm 2: vpxor <W2right61=%xmm11,<W2sigma1=%xmm8,<W2sigma1=%xmm8
2661 vpxor %xmm11,%xmm8,%xmm8
2662
2663 # qhasm: 4x X4 = X4 + X5sigma0
2664 # asm 1: vpaddq <X4=reg256#3,<X5sigma0=reg256#7,>X4=reg256#3
2665 # asm 2: vpaddq <X4=%ymm2,<X5sigma0=%ymm6,>X4=%ymm2
2666 vpaddq %ymm2,%ymm6,%ymm2
2667
2668 # qhasm: r7Sigma1 ^= r741
2669 # asm 1: xor <r741=int64#7,<r7Sigma1=int64#15
2670 # asm 2: xor <r741=%rax,<r7Sigma1=%rbp
2671 xor %rax,%rbp
2672
2673 # qhasm: maj2 ^= r4andr5
2674 # asm 1: xor <r4andr5=int64#12,<maj2=int64#14
2675 # asm 2: xor <r4andr5=%r14,<maj2=%rbx
2676 xor %r14,%rbx
2677
2678 # qhasm: 2x,0 W2left3 = W2 << 3
2679 # asm 1: vpsllq $3,<W2=reg256#6%128,>W2left3=reg256#7%128
2680 # asm 2: vpsllq $3,<W2=%xmm5,>W2left3=%xmm6
2681 vpsllq $3,%xmm5,%xmm6
2682
2683 # qhasm: r3Sigma0 = r3>>>28
2684 # asm 1: rorx $28,<r3=int64#6,>r3Sigma0=int64#7
2685 # asm 2: rorx $28,<r3=%r9,>r3Sigma0=%rax
2686 rorx $28,%r9,%rax
2687
2688 # qhasm: ch2 ^= r1
2689 # asm 1: xor <r1=int64#4,<ch2=int64#13
2690 # asm 2: xor <r1=%rcx,<ch2=%r15
2691 xor %rcx,%r15
2692
2693 # qhasm: r2 += r7Sigma1
2694 # asm 1: add <r7Sigma1=int64#15,<r2=int64#3
2695 # asm 2: add <r7Sigma1=%rbp,<r2=%rdx
2696 add %rbp,%rdx
2697
2698 # qhasm: r334 = r3>>>34
2699 # asm 1: rorx $34,<r3=int64#6,>r334=int64#12
2700 # asm 2: rorx $34,<r3=%r9,>r334=%r14
2701 rorx $34,%r9,%r14
2702
2703 # qhasm: 4x X4 = X4 + mem256[&w + 104]
2704 # asm 1: vpaddq <w=stack1280#1,<X4=reg256#3,>X4=reg256#3
2705 # asm 2: vpaddq <w=264(%rsp),<X4=%ymm2,>X4=%ymm2
2706 vpaddq 264(%rsp),%ymm2,%ymm2
2707
2708 # qhasm: r1 += wc12131415[2]
2709 # asm 1: addq <wc12131415=stack256#5,<r1=int64#4
2710 # asm 2: addq <wc12131415=144(%rsp),<r1=%rcx
2711 addq 144(%rsp),%rcx
2712
2713 # qhasm: r2 += ch2
2714 # asm 1: add <ch2=int64#13,<r2=int64#3
2715 # asm 2: add <ch2=%r15,<r2=%rdx
2716 add %r15,%rdx
2717
2718 # qhasm: r3Sigma0 ^= r334
2719 # asm 1: xor <r334=int64#12,<r3Sigma0=int64#7
2720 # asm 2: xor <r334=%r14,<r3Sigma0=%rax
2721 xor %r14,%rax
2722
2723 # qhasm: r339 = r3>>>39
2724 # asm 1: rorx $39,<r3=int64#6,>r339=int64#12
2725 # asm 2: rorx $39,<r3=%r9,>r339=%r14
2726 rorx $39,%r9,%r14
2727
2728 # qhasm: r6 += r2
2729 # asm 1: add <r2=int64#3,<r6=int64#10
2730 # asm 2: add <r2=%rdx,<r6=%r12
2731 add %rdx,%r12
2732
2733 # qhasm: 1x,0 W2sigma1 ^= W2left3
2734 # asm 1: vpxor <W2left3=reg256#7%128,<W2sigma1=reg256#9%128,<W2sigma1=reg256#9%128
2735 # asm 2: vpxor <W2left3=%xmm6,<W2sigma1=%xmm8,<W2sigma1=%xmm8
2736 vpxor %xmm6,%xmm8,%xmm8
2737
2738 # qhasm: 2x,0 W2right6 = W2 unsigned>> 6
2739 # asm 1: vpsrlq $6,<W2=reg256#6%128,>W2right6=reg256#6%128
2740 # asm 2: vpsrlq $6,<W2=%xmm5,>W2right6=%xmm5
2741 vpsrlq $6,%xmm5,%xmm5
2742
2743 # qhasm: r3Sigma0 ^= r339
2744 # asm 1: xor <r339=int64#12,<r3Sigma0=int64#7
2745 # asm 2: xor <r339=%r14,<r3Sigma0=%rax
2746 xor %r14,%rax
2747
2748 # qhasm: r2 += maj2
2749 # asm 1: add <maj2=int64#14,<r2=int64#3
2750 # asm 2: add <maj2=%rbx,<r2=%rdx
2751 add %rbx,%rdx
2752
2753 # qhasm: ch1 = r0
2754 # asm 1: mov <r0=int64#2,>ch1=int64#12
2755 # asm 2: mov <r0=%rsi,>ch1=%r14
2756 mov %rsi,%r14
2757
2758 # qhasm: ch1 ^= r7
2759 # asm 1: xor <r7=int64#11,<ch1=int64#12
2760 # asm 2: xor <r7=%r13,<ch1=%r14
2761 xor %r13,%r14
2762
2763 # qhasm: r2 += r3Sigma0
2764 # asm 1: add <r3Sigma0=int64#7,<r2=int64#3
2765 # asm 2: add <r3Sigma0=%rax,<r2=%rdx
2766 add %rax,%rdx
2767
2768 # qhasm: r6Sigma1 = r6>>>14
2769 # asm 1: rorx $14,<r6=int64#10,>r6Sigma1=int64#7
2770 # asm 2: rorx $14,<r6=%r12,>r6Sigma1=%rax
2771 rorx $14,%r12,%rax
2772
2773 # qhasm: 1x,0 W2sigma1 ^= W2right6
2774 # asm 1: vpxor <W2right6=reg256#6%128,<W2sigma1=reg256#9%128,<W2sigma1=reg256#9%128
2775 # asm 2: vpxor <W2right6=%xmm5,<W2sigma1=%xmm8,<W2sigma1=%xmm8
2776 vpxor %xmm5,%xmm8,%xmm8
2777
2778 # qhasm: ch1 &= r6
2779 # asm 1: and <r6=int64#10,<ch1=int64#12
2780 # asm 2: and <r6=%r12,<ch1=%r14
2781 and %r12,%r14
2782
2783 # qhasm: r618 = r6>>>18
2784 # asm 1: rorx $18,<r6=int64#10,>r618=int64#13
2785 # asm 2: rorx $18,<r6=%r12,>r618=%r15
2786 rorx $18,%r12,%r15
2787
2788 # qhasm: r641 = r6>>>41
2789 # asm 1: rorx $41,<r6=int64#10,>r641=int64#14
2790 # asm 2: rorx $41,<r6=%r12,>r641=%rbx
2791 rorx $41,%r12,%rbx
2792
2793 # qhasm: 4x X4 = W2sigma1 + X4
2794 # asm 1: vpaddq <W2sigma1=reg256#9,<X4=reg256#3,>X4=reg256#3
2795 # asm 2: vpaddq <W2sigma1=%ymm8,<X4=%ymm2,>X4=%ymm2
2796 vpaddq %ymm8,%ymm2,%ymm2
2797
2798 # qhasm: ch1 ^= r0
2799 # asm 1: xor <r0=int64#2,<ch1=int64#12
2800 # asm 2: xor <r0=%rsi,<ch1=%r14
2801 xor %rsi,%r14
2802
2803 # qhasm: r6Sigma1 ^= r618
2804 # asm 1: xor <r618=int64#13,<r6Sigma1=int64#7
2805 # asm 2: xor <r618=%r15,<r6Sigma1=%rax
2806 xor %r15,%rax
2807
2808 # qhasm: 2x,0 W4right19 = X4 unsigned>> 19
2809 # asm 1: vpsrlq $19,<X4=reg256#3%128,>W4right19=reg256#6%128
2810 # asm 2: vpsrlq $19,<X4=%xmm2,>W4right19=%xmm5
2811 vpsrlq $19,%xmm2,%xmm5
2812
2813 # qhasm: r1 += ch1
2814 # asm 1: add <ch1=int64#12,<r1=int64#4
2815 # asm 2: add <ch1=%r14,<r1=%rcx
2816 add %r14,%rcx
2817
2818 # qhasm: r2Sigma0 = r2>>>28
2819 # asm 1: rorx $28,<r2=int64#3,>r2Sigma0=int64#12
2820 # asm 2: rorx $28,<r2=%rdx,>r2Sigma0=%r14
2821 rorx $28,%rdx,%r14
2822
2823 # qhasm: 2x,0 W4left45 = X4 << 45
2824 # asm 1: vpsllq $45,<X4=reg256#3%128,>W4left45=reg256#7%128
2825 # asm 2: vpsllq $45,<X4=%xmm2,>W4left45=%xmm6
2826 vpsllq $45,%xmm2,%xmm6
2827
2828 # qhasm: r6Sigma1 ^= r641
2829 # asm 1: xor <r641=int64#14,<r6Sigma1=int64#7
2830 # asm 2: xor <r641=%rbx,<r6Sigma1=%rax
2831 xor %rbx,%rax
2832
2833 # qhasm: r234 = r2>>>34
2834 # asm 1: rorx $34,<r2=int64#3,>r234=int64#13
2835 # asm 2: rorx $34,<r2=%rdx,>r234=%r15
2836 rorx $34,%rdx,%r15
2837
2838 # qhasm: maj0 = r3
2839 # asm 1: mov <r3=int64#6,>maj0=int64#14
2840 # asm 2: mov <r3=%r9,>maj0=%rbx
2841 mov %r9,%rbx
2842
2843 # qhasm: maj0 ^= r2
2844 # asm 1: xor <r2=int64#3,<maj0=int64#14
2845 # asm 2: xor <r2=%rdx,<maj0=%rbx
2846 xor %rdx,%rbx
2847
2848 # qhasm: 2x,0 W4right61 = X4 unsigned>> 61
2849 # asm 1: vpsrlq $61,<X4=reg256#3%128,>W4right61=reg256#8%128
2850 # asm 2: vpsrlq $61,<X4=%xmm2,>W4right61=%xmm7
2851 vpsrlq $61,%xmm2,%xmm7
2852
2853 # qhasm: X9 = mem256[&w + 72]
2854 # asm 1: vmovupd <w=stack1280#1,>X9=reg256#9
2855 # asm 2: vmovupd <w=232(%rsp),>X9=%ymm8
2856 vmovupd 232(%rsp),%ymm8
2857
2858 # qhasm: r2Sigma0 ^= r234
2859 # asm 1: xor <r234=int64#13,<r2Sigma0=int64#12
2860 # asm 2: xor <r234=%r15,<r2Sigma0=%r14
2861 xor %r15,%r14
2862
2863 # qhasm: 1x,0 W4sigma1 = W4right19 ^ W4left45
2864 # asm 1: vpxor <W4right19=reg256#6%128,<W4left45=reg256#7%128,>W4sigma1=reg256#6%128
2865 # asm 2: vpxor <W4right19=%xmm5,<W4left45=%xmm6,>W4sigma1=%xmm5
2866 vpxor %xmm5,%xmm6,%xmm5
2867
2868 # qhasm: r239 = r2>>>39
2869 # asm 1: rorx $39,<r2=int64#3,>r239=int64#13
2870 # asm 2: rorx $39,<r2=%rdx,>r239=%r15
2871 rorx $39,%rdx,%r15
2872
2873 # qhasm: 2x,0 W4left3 = X4 << 3
2874 # asm 1: vpsllq $3,<X4=reg256#3%128,>W4left3=reg256#7%128
2875 # asm 2: vpsllq $3,<X4=%xmm2,>W4left3=%xmm6
2876 vpsllq $3,%xmm2,%xmm6
2877
2878 # qhasm: 1x,0 W4sigma1 ^= W4right61
2879 # asm 1: vpxor <W4right61=reg256#8%128,<W4sigma1=reg256#6%128,<W4sigma1=reg256#6%128
2880 # asm 2: vpxor <W4right61=%xmm7,<W4sigma1=%xmm5,<W4sigma1=%xmm5
2881 vpxor %xmm7,%xmm5,%xmm5
2882
2883 # qhasm: 2x,0 W4right6 = X4 unsigned>> 6
2884 # asm 1: vpsrlq $6,<X4=reg256#3%128,>W4right6=reg256#8%128
2885 # asm 2: vpsrlq $6,<X4=%xmm2,>W4right6=%xmm7
2886 vpsrlq $6,%xmm2,%xmm7
2887
2888 # qhasm: r2Sigma0 ^= r239
2889 # asm 1: xor <r239=int64#13,<r2Sigma0=int64#12
2890 # asm 2: xor <r239=%r15,<r2Sigma0=%r14
2891 xor %r15,%r14
2892
2893 # qhasm: r1 += r6Sigma1
2894 # asm 1: add <r6Sigma1=int64#7,<r1=int64#4
2895 # asm 2: add <r6Sigma1=%rax,<r1=%rcx
2896 add %rax,%rcx
2897
2898 # qhasm: 1x,0 W4sigma1 ^= W4left3
2899 # asm 1: vpxor <W4left3=reg256#7%128,<W4sigma1=reg256#6%128,<W4sigma1=reg256#6%128
2900 # asm 2: vpxor <W4left3=%xmm6,<W4sigma1=%xmm5,<W4sigma1=%xmm5
2901 vpxor %xmm6,%xmm5,%xmm5
2902
2903 # qhasm: r5 += r1
2904 # asm 1: add <r1=int64#4,<r5=int64#8
2905 # asm 2: add <r1=%rcx,<r5=%r10
2906 add %rcx,%r10
2907
2908 # qhasm: r2andr3 = r3
2909 # asm 1: mov <r3=int64#6,>r2andr3=int64#7
2910 # asm 2: mov <r3=%r9,>r2andr3=%rax
2911 mov %r9,%rax
2912
2913 # qhasm: r5Sigma1 = r5>>>14
2914 # asm 1: rorx $14,<r5=int64#8,>r5Sigma1=int64#13
2915 # asm 2: rorx $14,<r5=%r10,>r5Sigma1=%r15
2916 rorx $14,%r10,%r15
2917
2918 # qhasm: r2andr3 &= r2
2919 # asm 1: and <r2=int64#3,<r2andr3=int64#7
2920 # asm 2: and <r2=%rdx,<r2andr3=%rax
2921 and %rdx,%rax
2922
2923 # qhasm: 1x,0 W4sigma1 ^= W4right6
2924 # asm 1: vpxor <W4right6=reg256#8%128,<W4sigma1=reg256#6%128,<W4sigma1=reg256#6%128
2925 # asm 2: vpxor <W4right6=%xmm7,<W4sigma1=%xmm5,<W4sigma1=%xmm5
2926 vpxor %xmm7,%xmm5,%xmm5
2927
2928 # qhasm: r1 += r2Sigma0
2929 # asm 1: add <r2Sigma0=int64#12,<r1=int64#4
2930 # asm 2: add <r2Sigma0=%r14,<r1=%rcx
2931 add %r14,%rcx
2932
2933 # qhasm: maj1 = r4
2934 # asm 1: mov <r4=int64#9,>maj1=int64#12
2935 # asm 2: mov <r4=%r11,>maj1=%r14
2936 mov %r11,%r14
2937
2938 # qhasm: maj1 &= maj0
2939 # asm 1: and <maj0=int64#14,<maj1=int64#12
2940 # asm 2: and <maj0=%rbx,<maj1=%r14
2941 and %rbx,%r14
2942
2943 # qhasm: W4sigma1 = W4sigma1[2,3,0,1]
2944 # asm 1: vpermq $0x4e,<W4sigma1=reg256#6,>W4sigma1=reg256#6
2945 # asm 2: vpermq $0x4e,<W4sigma1=%ymm5,>W4sigma1=%ymm5
2946 vpermq $0x4e,%ymm5,%ymm5
2947
2948 # qhasm: maj1 ^= r2andr3
2949 # asm 1: xor <r2andr3=int64#7,<maj1=int64#12
2950 # asm 2: xor <r2andr3=%rax,<maj1=%r14
2951 xor %rax,%r14
2952
2953 # qhasm: ch0 = r7
2954 # asm 1: mov <r7=int64#11,>ch0=int64#15
2955 # asm 2: mov <r7=%r13,>ch0=%rbp
2956 mov %r13,%rbp
2957
2958 # qhasm: 4x X9right1 = X9 unsigned>> 1
2959 # asm 1: vpsrlq $1,<X9=reg256#9,>X9right1=reg256#7
2960 # asm 2: vpsrlq $1,<X9=%ymm8,>X9right1=%ymm6
2961 vpsrlq $1,%ymm8,%ymm6
2962
2963 # qhasm: ch0 ^= r6
2964 # asm 1: xor <r6=int64#10,<ch0=int64#15
2965 # asm 2: xor <r6=%r12,<ch0=%rbp
2966 xor %r12,%rbp
2967
2968 # qhasm: r1 += maj1
2969 # asm 1: add <maj1=int64#12,<r1=int64#4
2970 # asm 2: add <maj1=%r14,<r1=%rcx
2971 add %r14,%rcx
2972
2973 # qhasm: ch0 &= r5
2974 # asm 1: and <r5=int64#8,<ch0=int64#15
2975 # asm 2: and <r5=%r10,<ch0=%rbp
2976 and %r10,%rbp
2977
2978 # qhasm: r518 = r5>>>18
2979 # asm 1: rorx $18,<r5=int64#8,>r518=int64#12
2980 # asm 2: rorx $18,<r5=%r10,>r518=%r14
2981 rorx $18,%r10,%r14
2982
2983 # qhasm: maj0 &= r1
2984 # asm 1: and <r1=int64#4,<maj0=int64#14
2985 # asm 2: and <r1=%rcx,<maj0=%rbx
2986 and %rcx,%rbx
2987
2988 # qhasm: r0 += wc12131415[3]
2989 # asm 1: addq <wc12131415=stack256#5,<r0=int64#2
2990 # asm 2: addq <wc12131415=152(%rsp),<r0=%rsi
2991 addq 152(%rsp),%rsi
2992
2993 # qhasm: ch0 ^= r7
2994 # asm 1: xor <r7=int64#11,<ch0=int64#15
2995 # asm 2: xor <r7=%r13,<ch0=%rbp
2996 xor %r13,%rbp
2997
2998 # qhasm: r5Sigma1 ^= r518
2999 # asm 1: xor <r518=int64#12,<r5Sigma1=int64#13
3000 # asm 2: xor <r518=%r14,<r5Sigma1=%r15
3001 xor %r14,%r15
3002
3003 # qhasm: 4x X4 = X4 + W4sigma1
3004 # asm 1: vpaddq <X4=reg256#3,<W4sigma1=reg256#6,>X4=reg256#3
3005 # asm 2: vpaddq <X4=%ymm2,<W4sigma1=%ymm5,>X4=%ymm2
3006 vpaddq %ymm2,%ymm5,%ymm2
3007
3008 # qhasm: r541 = r5>>>41
3009 # asm 1: rorx $41,<r5=int64#8,>r541=int64#12
3010 # asm 2: rorx $41,<r5=%r10,>r541=%r14
3011 rorx $41,%r10,%r14
3012
3013 # qhasm: mem256[&w + 32] = X4
3014 # asm 1: vmovupd <X4=reg256#3,<w=stack1280#1
3015 # asm 2: vmovupd <X4=%ymm2,<w=192(%rsp)
3016 vmovupd %ymm2,192(%rsp)
3017
3018 # qhasm: r0 += ch0
3019 # asm 1: add <ch0=int64#15,<r0=int64#2
3020 # asm 2: add <ch0=%rbp,<r0=%rsi
3021 add %rbp,%rsi
3022
3023 # qhasm: maj0 ^= r2andr3
3024 # asm 1: xor <r2andr3=int64#7,<maj0=int64#14
3025 # asm 2: xor <r2andr3=%rax,<maj0=%rbx
3026 xor %rax,%rbx
3027
3028 # qhasm: r1Sigma0 = r1>>>28
3029 # asm 1: rorx $28,<r1=int64#4,>r1Sigma0=int64#7
3030 # asm 2: rorx $28,<r1=%rcx,>r1Sigma0=%rax
3031 rorx $28,%rcx,%rax
3032
3033 # qhasm: 4x D4 = X4 + mem256[constants + 32]
3034 # asm 1: vpaddq 32(<constants=int64#5),<X4=reg256#3,>D4=reg256#6
3035 # asm 2: vpaddq 32(<constants=%r8),<X4=%ymm2,>D4=%ymm5
3036 vpaddq 32(%r8),%ymm2,%ymm5
3037
3038 # qhasm: wc4567 = D4
3039 # asm 1: vmovapd <D4=reg256#6,>wc4567=stack256#4
3040 # asm 2: vmovapd <D4=%ymm5,>wc4567=96(%rsp)
3041 vmovapd %ymm5,96(%rsp)
3042
3043 # qhasm: r5Sigma1 ^= r541
3044 # asm 1: xor <r541=int64#12,<r5Sigma1=int64#13
3045 # asm 2: xor <r541=%r14,<r5Sigma1=%r15
3046 xor %r14,%r15
3047
3048 # qhasm: r134 = r1>>>34
3049 # asm 1: rorx $34,<r1=int64#4,>r134=int64#12
3050 # asm 2: rorx $34,<r1=%rcx,>r134=%r14
3051 rorx $34,%rcx,%r14
3052
3053 # qhasm: r0 += r5Sigma1
3054 # asm 1: add <r5Sigma1=int64#13,<r0=int64#2
3055 # asm 2: add <r5Sigma1=%r15,<r0=%rsi
3056 add %r15,%rsi
3057
3058 # qhasm: r1Sigma0 ^= r134
3059 # asm 1: xor <r134=int64#12,<r1Sigma0=int64#7
3060 # asm 2: xor <r134=%r14,<r1Sigma0=%rax
3061 xor %r14,%rax
3062
3063 # qhasm: r7 += wc0123[0]
3064 # asm 1: addq <wc0123=stack256#3,<r7=int64#11
3065 # asm 2: addq <wc0123=64(%rsp),<r7=%r13
3066 addq 64(%rsp),%r13
3067
3068 # qhasm: r139 = r1>>>39
3069 # asm 1: rorx $39,<r1=int64#4,>r139=int64#12
3070 # asm 2: rorx $39,<r1=%rcx,>r139=%r14
3071 rorx $39,%rcx,%r14
3072
3073 # qhasm: r4 += r0
3074 # asm 1: add <r0=int64#2,<r4=int64#9
3075 # asm 2: add <r0=%rsi,<r4=%r11
3076 add %rsi,%r11
3077
3078 # qhasm: r0 += maj0
3079 # asm 1: add <maj0=int64#14,<r0=int64#2
3080 # asm 2: add <maj0=%rbx,<r0=%rsi
3081 add %rbx,%rsi
3082
3083 # qhasm: r1Sigma0 ^= r139
3084 # asm 1: xor <r139=int64#12,<r1Sigma0=int64#7
3085 # asm 2: xor <r139=%r14,<r1Sigma0=%rax
3086 xor %r14,%rax
3087
3088 # qhasm: r4Sigma1 = r4>>>14
3089 # asm 1: rorx $14,<r4=int64#9,>r4Sigma1=int64#12
3090 # asm 2: rorx $14,<r4=%r11,>r4Sigma1=%r14
3091 rorx $14,%r11,%r14
3092
3093 # qhasm: W6 = mem128[&w + 48],0
3094 # asm 1: vmovupd <w=stack1280#1,>W6=reg256#6%128
3095 # asm 2: vmovupd <w=208(%rsp),>W6=%xmm5
3096 vmovupd 208(%rsp),%xmm5
3097
3098 # qhasm: r0 += r1Sigma0
3099 # asm 1: add <r1Sigma0=int64#7,<r0=int64#2
3100 # asm 2: add <r1Sigma0=%rax,<r0=%rsi
3101 add %rax,%rsi
3102
3103 # qhasm: ch7 = r6
3104 # asm 1: mov <r6=int64#10,>ch7=int64#7
3105 # asm 2: mov <r6=%r12,>ch7=%rax
3106 mov %r12,%rax
3107
3108 # qhasm: ch7 ^= r5
3109 # asm 1: xor <r5=int64#8,<ch7=int64#7
3110 # asm 2: xor <r5=%r10,<ch7=%rax
3111 xor %r10,%rax
3112
3113 # qhasm: r418 = r4>>>18
3114 # asm 1: rorx $18,<r4=int64#9,>r418=int64#13
3115 # asm 2: rorx $18,<r4=%r11,>r418=%r15
3116 rorx $18,%r11,%r15
3117
3118 # qhasm: 4x X9left63 = X9 << 63
3119 # asm 1: vpsllq $63,<X9=reg256#9,>X9left63=reg256#8
3120 # asm 2: vpsllq $63,<X9=%ymm8,>X9left63=%ymm7
3121 vpsllq $63,%ymm8,%ymm7
3122
3123 # qhasm: ch7 &= r4
3124 # asm 1: and <r4=int64#9,<ch7=int64#7
3125 # asm 2: and <r4=%r11,<ch7=%rax
3126 and %r11,%rax
3127
3128 # qhasm: maj6 = r1
3129 # asm 1: mov <r1=int64#4,>maj6=int64#14
3130 # asm 2: mov <r1=%rcx,>maj6=%rbx
3131 mov %rcx,%rbx
3132
3133 # qhasm: maj6 ^= r0
3134 # asm 1: xor <r0=int64#2,<maj6=int64#14
3135 # asm 2: xor <r0=%rsi,<maj6=%rbx
3136 xor %rsi,%rbx
3137
3138 # qhasm: r441 = r4>>>41
3139 # asm 1: rorx $41,<r4=int64#9,>r441=int64#15
3140 # asm 2: rorx $41,<r4=%r11,>r441=%rbp
3141 rorx $41,%r11,%rbp
3142
3143 # qhasm: X9sigma0 = X9right1 ^ X9left63
3144 # asm 1: vpxor <X9right1=reg256#7,<X9left63=reg256#8,>X9sigma0=reg256#7
3145 # asm 2: vpxor <X9right1=%ymm6,<X9left63=%ymm7,>X9sigma0=%ymm6
3146 vpxor %ymm6,%ymm7,%ymm6
3147
3148 # qhasm: r4Sigma1 ^= r418
3149 # asm 1: xor <r418=int64#13,<r4Sigma1=int64#12
3150 # asm 2: xor <r418=%r15,<r4Sigma1=%r14
3151 xor %r15,%r14
3152
3153 # qhasm: 4x X9right8 = X9 unsigned>> 8
3154 # asm 1: vpsrlq $8,<X9=reg256#9,>X9right8=reg256#8
3155 # asm 2: vpsrlq $8,<X9=%ymm8,>X9right8=%ymm7
3156 vpsrlq $8,%ymm8,%ymm7
3157
3158 # qhasm: ch7 ^= r6
3159 # asm 1: xor <r6=int64#10,<ch7=int64#7
3160 # asm 2: xor <r6=%r12,<ch7=%rax
3161 xor %r12,%rax
3162
3163 # qhasm: r4Sigma1 ^= r441
3164 # asm 1: xor <r441=int64#15,<r4Sigma1=int64#12
3165 # asm 2: xor <r441=%rbp,<r4Sigma1=%r14
3166 xor %rbp,%r14
3167
3168 # qhasm: r0Sigma0 = r0>>>28
3169 # asm 1: rorx $28,<r0=int64#2,>r0Sigma0=int64#13
3170 # asm 2: rorx $28,<r0=%rsi,>r0Sigma0=%r15
3171 rorx $28,%rsi,%r15
3172
3173 # qhasm: r7 += ch7
3174 # asm 1: add <ch7=int64#7,<r7=int64#11
3175 # asm 2: add <ch7=%rax,<r7=%r13
3176 add %rax,%r13
3177
3178 # qhasm: X9sigma0 = X9sigma0 ^ X9right8
3179 # asm 1: vpxor <X9sigma0=reg256#7,<X9right8=reg256#8,>X9sigma0=reg256#7
3180 # asm 2: vpxor <X9sigma0=%ymm6,<X9right8=%ymm7,>X9sigma0=%ymm6
3181 vpxor %ymm6,%ymm7,%ymm6
3182
3183 # qhasm: r7 += r4Sigma1
3184 # asm 1: add <r4Sigma1=int64#12,<r7=int64#11
3185 # asm 2: add <r4Sigma1=%r14,<r7=%r13
3186 add %r14,%r13
3187
3188 # qhasm: r034 = r0>>>34
3189 # asm 1: rorx $34,<r0=int64#2,>r034=int64#7
3190 # asm 2: rorx $34,<r0=%rsi,>r034=%rax
3191 rorx $34,%rsi,%rax
3192
3193 # qhasm: maj7 = r2
3194 # asm 1: mov <r2=int64#3,>maj7=int64#12
3195 # asm 2: mov <r2=%rdx,>maj7=%r14
3196 mov %rdx,%r14
3197
3198 # qhasm: maj7 &= maj6
3199 # asm 1: and <maj6=int64#14,<maj7=int64#12
3200 # asm 2: and <maj6=%rbx,<maj7=%r14
3201 and %rbx,%r14
3202
3203 # qhasm: r0Sigma0 ^= r034
3204 # asm 1: xor <r034=int64#7,<r0Sigma0=int64#13
3205 # asm 2: xor <r034=%rax,<r0Sigma0=%r15
3206 xor %rax,%r15
3207
3208 # qhasm: 2x,0 W6right19 = W6 unsigned>> 19
3209 # asm 1: vpsrlq $19,<W6=reg256#6%128,>W6right19=reg256#8%128
3210 # asm 2: vpsrlq $19,<W6=%xmm5,>W6right19=%xmm7
3211 vpsrlq $19,%xmm5,%xmm7
3212
3213 # qhasm: r039 = r0>>>39
3214 # asm 1: rorx $39,<r0=int64#2,>r039=int64#7
3215 # asm 2: rorx $39,<r0=%rsi,>r039=%rax
3216 rorx $39,%rsi,%rax
3217
3218 # qhasm: r3 += r7
3219 # asm 1: add <r7=int64#11,<r3=int64#6
3220 # asm 2: add <r7=%r13,<r3=%r9
3221 add %r13,%r9
3222
3223 # qhasm: r0Sigma0 ^= r039
3224 # asm 1: xor <r039=int64#7,<r0Sigma0=int64#13
3225 # asm 2: xor <r039=%rax,<r0Sigma0=%r15
3226 xor %rax,%r15
3227
3228 # qhasm: 4x X9left56 = X9 << 56
3229 # asm 1: vpsllq $56,<X9=reg256#9,>X9left56=reg256#10
3230 # asm 2: vpsllq $56,<X9=%ymm8,>X9left56=%ymm9
3231 vpsllq $56,%ymm8,%ymm9
3232
3233 # qhasm: r6 += wc0123[1]
3234 # asm 1: addq <wc0123=stack256#3,<r6=int64#10
3235 # asm 2: addq <wc0123=72(%rsp),<r6=%r12
3236 addq 72(%rsp),%r12
3237
3238 # qhasm: r0andr1 = r1
3239 # asm 1: mov <r1=int64#4,>r0andr1=int64#7
3240 # asm 2: mov <r1=%rcx,>r0andr1=%rax
3241 mov %rcx,%rax
3242
3243 # qhasm: r0andr1 &= r0
3244 # asm 1: and <r0=int64#2,<r0andr1=int64#7
3245 # asm 2: and <r0=%rsi,<r0andr1=%rax
3246 and %rsi,%rax
3247
3248 # qhasm: r7 += r0Sigma0
3249 # asm 1: add <r0Sigma0=int64#13,<r7=int64#11
3250 # asm 2: add <r0Sigma0=%r15,<r7=%r13
3251 add %r15,%r13
3252
3253 # qhasm: ch6 = r5
3254 # asm 1: mov <r5=int64#8,>ch6=int64#13
3255 # asm 2: mov <r5=%r10,>ch6=%r15
3256 mov %r10,%r15
3257
3258 # qhasm: 2x,0 W6left45 = W6 << 45
3259 # asm 1: vpsllq $45,<W6=reg256#6%128,>W6left45=reg256#11%128
3260 # asm 2: vpsllq $45,<W6=%xmm5,>W6left45=%xmm10
3261 vpsllq $45,%xmm5,%xmm10
3262
3263 # qhasm: maj7 ^= r0andr1
3264 # asm 1: xor <r0andr1=int64#7,<maj7=int64#12
3265 # asm 2: xor <r0andr1=%rax,<maj7=%r14
3266 xor %rax,%r14
3267
3268 # qhasm: ch6 ^= r4
3269 # asm 1: xor <r4=int64#9,<ch6=int64#13
3270 # asm 2: xor <r4=%r11,<ch6=%r15
3271 xor %r11,%r15
3272
3273 # qhasm: 2x,0 W6right61 = W6 unsigned>> 61
3274 # asm 1: vpsrlq $61,<W6=reg256#6%128,>W6right61=reg256#12%128
3275 # asm 2: vpsrlq $61,<W6=%xmm5,>W6right61=%xmm11
3276 vpsrlq $61,%xmm5,%xmm11
3277
3278 # qhasm: r3Sigma1 = r3>>>14
3279 # asm 1: rorx $14,<r3=int64#6,>r3Sigma1=int64#15
3280 # asm 2: rorx $14,<r3=%r9,>r3Sigma1=%rbp
3281 rorx $14,%r9,%rbp
3282
3283 # qhasm: r7 += maj7
3284 # asm 1: add <maj7=int64#12,<r7=int64#11
3285 # asm 2: add <maj7=%r14,<r7=%r13
3286 add %r14,%r13
3287
3288 # qhasm: X9sigma0 = X9sigma0 ^ X9left56
3289 # asm 1: vpxor <X9sigma0=reg256#7,<X9left56=reg256#10,>X9sigma0=reg256#7
3290 # asm 2: vpxor <X9sigma0=%ymm6,<X9left56=%ymm9,>X9sigma0=%ymm6
3291 vpxor %ymm6,%ymm9,%ymm6
3292
3293 # qhasm: 4x X9right7 = X9 unsigned>> 7
3294 # asm 1: vpsrlq $7,<X9=reg256#9,>X9right7=reg256#9
3295 # asm 2: vpsrlq $7,<X9=%ymm8,>X9right7=%ymm8
3296 vpsrlq $7,%ymm8,%ymm8
3297
3298 # qhasm: r318 = r3>>>18
3299 # asm 1: rorx $18,<r3=int64#6,>r318=int64#12
3300 # asm 2: rorx $18,<r3=%r9,>r318=%r14
3301 rorx $18,%r9,%r14
3302
3303 # qhasm: 1x,0 W6sigma1 = W6right19 ^ W6left45
3304 # asm 1: vpxor <W6right19=reg256#8%128,<W6left45=reg256#11%128,>W6sigma1=reg256#8%128
3305 # asm 2: vpxor <W6right19=%xmm7,<W6left45=%xmm10,>W6sigma1=%xmm7
3306 vpxor %xmm7,%xmm10,%xmm7
3307
3308 # qhasm: ch6 &= r3
3309 # asm 1: and <r3=int64#6,<ch6=int64#13
3310 # asm 2: and <r3=%r9,<ch6=%r15
3311 and %r9,%r15
3312
3313 # qhasm: r3Sigma1 ^= r318
3314 # asm 1: xor <r318=int64#12,<r3Sigma1=int64#15
3315 # asm 2: xor <r318=%r14,<r3Sigma1=%rbp
3316 xor %r14,%rbp
3317
3318 # qhasm: r341 = r3>>>41
3319 # asm 1: rorx $41,<r3=int64#6,>r341=int64#12
3320 # asm 2: rorx $41,<r3=%r9,>r341=%r14
3321 rorx $41,%r9,%r14
3322
3323 # qhasm: X9sigma0 = X9sigma0 ^ X9right7
3324 # asm 1: vpxor <X9sigma0=reg256#7,<X9right7=reg256#9,>X9sigma0=reg256#7
3325 # asm 2: vpxor <X9sigma0=%ymm6,<X9right7=%ymm8,>X9sigma0=%ymm6
3326 vpxor %ymm6,%ymm8,%ymm6
3327
3328 # qhasm: maj6 &= r7
3329 # asm 1: and <r7=int64#11,<maj6=int64#14
3330 # asm 2: and <r7=%r13,<maj6=%rbx
3331 and %r13,%rbx
3332
3333 # qhasm: 1x,0 W6sigma1 ^= W6right61
3334 # asm 1: vpxor <W6right61=reg256#12%128,<W6sigma1=reg256#8%128,<W6sigma1=reg256#8%128
3335 # asm 2: vpxor <W6right61=%xmm11,<W6sigma1=%xmm7,<W6sigma1=%xmm7
3336 vpxor %xmm11,%xmm7,%xmm7
3337
3338 # qhasm: 4x X8 = X8 + X9sigma0
3339 # asm 1: vpaddq <X8=reg256#4,<X9sigma0=reg256#7,>X8=reg256#4
3340 # asm 2: vpaddq <X8=%ymm3,<X9sigma0=%ymm6,>X8=%ymm3
3341 vpaddq %ymm3,%ymm6,%ymm3
3342
3343 # qhasm: r3Sigma1 ^= r341
3344 # asm 1: xor <r341=int64#12,<r3Sigma1=int64#15
3345 # asm 2: xor <r341=%r14,<r3Sigma1=%rbp
3346 xor %r14,%rbp
3347
3348 # qhasm: maj6 ^= r0andr1
3349 # asm 1: xor <r0andr1=int64#7,<maj6=int64#14
3350 # asm 2: xor <r0andr1=%rax,<maj6=%rbx
3351 xor %rax,%rbx
3352
3353 # qhasm: 2x,0 W6left3 = W6 << 3
3354 # asm 1: vpsllq $3,<W6=reg256#6%128,>W6left3=reg256#7%128
3355 # asm 2: vpsllq $3,<W6=%xmm5,>W6left3=%xmm6
3356 vpsllq $3,%xmm5,%xmm6
3357
3358 # qhasm: r7Sigma0 = r7>>>28
3359 # asm 1: rorx $28,<r7=int64#11,>r7Sigma0=int64#7
3360 # asm 2: rorx $28,<r7=%r13,>r7Sigma0=%rax
3361 rorx $28,%r13,%rax
3362
3363 # qhasm: ch6 ^= r5
3364 # asm 1: xor <r5=int64#8,<ch6=int64#13
3365 # asm 2: xor <r5=%r10,<ch6=%r15
3366 xor %r10,%r15
3367
3368 # qhasm: r6 += r3Sigma1
3369 # asm 1: add <r3Sigma1=int64#15,<r6=int64#10
3370 # asm 2: add <r3Sigma1=%rbp,<r6=%r12
3371 add %rbp,%r12
3372
3373 # qhasm: 4x X8 = X8 + mem256[&w + 8]
3374 # asm 1: vpaddq <w=stack1280#1,<X8=reg256#4,>X8=reg256#4
3375 # asm 2: vpaddq <w=168(%rsp),<X8=%ymm3,>X8=%ymm3
3376 vpaddq 168(%rsp),%ymm3,%ymm3
3377
3378 # qhasm: r734 = r7>>>34
3379 # asm 1: rorx $34,<r7=int64#11,>r734=int64#12
3380 # asm 2: rorx $34,<r7=%r13,>r734=%r14
3381 rorx $34,%r13,%r14
3382
3383 # qhasm: r5 += wc0123[2]
3384 # asm 1: addq <wc0123=stack256#3,<r5=int64#8
3385 # asm 2: addq <wc0123=80(%rsp),<r5=%r10
3386 addq 80(%rsp),%r10
3387
3388 # qhasm: r6 += ch6
3389 # asm 1: add <ch6=int64#13,<r6=int64#10
3390 # asm 2: add <ch6=%r15,<r6=%r12
3391 add %r15,%r12
3392
3393 # qhasm: r7Sigma0 ^= r734
3394 # asm 1: xor <r734=int64#12,<r7Sigma0=int64#7
3395 # asm 2: xor <r734=%r14,<r7Sigma0=%rax
3396 xor %r14,%rax
3397
3398 # qhasm: 1x,0 W6sigma1 ^= W6left3
3399 # asm 1: vpxor <W6left3=reg256#7%128,<W6sigma1=reg256#8%128,<W6sigma1=reg256#8%128
3400 # asm 2: vpxor <W6left3=%xmm6,<W6sigma1=%xmm7,<W6sigma1=%xmm7
3401 vpxor %xmm6,%xmm7,%xmm7
3402
3403 # qhasm: r739 = r7>>>39
3404 # asm 1: rorx $39,<r7=int64#11,>r739=int64#12
3405 # asm 2: rorx $39,<r7=%r13,>r739=%r14
3406 rorx $39,%r13,%r14
3407
3408 # qhasm: r2 += r6
3409 # asm 1: add <r6=int64#10,<r2=int64#3
3410 # asm 2: add <r6=%r12,<r2=%rdx
3411 add %r12,%rdx
3412
3413 # qhasm: r7Sigma0 ^= r739
3414 # asm 1: xor <r739=int64#12,<r7Sigma0=int64#7
3415 # asm 2: xor <r739=%r14,<r7Sigma0=%rax
3416 xor %r14,%rax
3417
3418 # qhasm: ch5 = r4
3419 # asm 1: mov <r4=int64#9,>ch5=int64#12
3420 # asm 2: mov <r4=%r11,>ch5=%r14
3421 mov %r11,%r14
3422
3423 # qhasm: 2x,0 W6right6 = W6 unsigned>> 6
3424 # asm 1: vpsrlq $6,<W6=reg256#6%128,>W6right6=reg256#6%128
3425 # asm 2: vpsrlq $6,<W6=%xmm5,>W6right6=%xmm5
3426 vpsrlq $6,%xmm5,%xmm5
3427
3428 # qhasm: r6 += maj6
3429 # asm 1: add <maj6=int64#14,<r6=int64#10
3430 # asm 2: add <maj6=%rbx,<r6=%r12
3431 add %rbx,%r12
3432
3433 # qhasm: ch5 ^= r3
3434 # asm 1: xor <r3=int64#6,<ch5=int64#12
3435 # asm 2: xor <r3=%r9,<ch5=%r14
3436 xor %r9,%r14
3437
3438 # qhasm: r6 += r7Sigma0
3439 # asm 1: add <r7Sigma0=int64#7,<r6=int64#10
3440 # asm 2: add <r7Sigma0=%rax,<r6=%r12
3441 add %rax,%r12
3442
3443 # qhasm: r2Sigma1 = r2>>>14
3444 # asm 1: rorx $14,<r2=int64#3,>r2Sigma1=int64#7
3445 # asm 2: rorx $14,<r2=%rdx,>r2Sigma1=%rax
3446 rorx $14,%rdx,%rax
3447
3448 # qhasm: 1x,0 W6sigma1 ^= W6right6
3449 # asm 1: vpxor <W6right6=reg256#6%128,<W6sigma1=reg256#8%128,<W6sigma1=reg256#8%128
3450 # asm 2: vpxor <W6right6=%xmm5,<W6sigma1=%xmm7,<W6sigma1=%xmm7
3451 vpxor %xmm5,%xmm7,%xmm7
3452
3453 # qhasm: ch5 &= r2
3454 # asm 1: and <r2=int64#3,<ch5=int64#12
3455 # asm 2: and <r2=%rdx,<ch5=%r14
3456 and %rdx,%r14
3457
3458 # qhasm: r218 = r2>>>18
3459 # asm 1: rorx $18,<r2=int64#3,>r218=int64#13
3460 # asm 2: rorx $18,<r2=%rdx,>r218=%r15
3461 rorx $18,%rdx,%r15
3462
3463 # qhasm: r241 = r2>>>41
3464 # asm 1: rorx $41,<r2=int64#3,>r241=int64#14
3465 # asm 2: rorx $41,<r2=%rdx,>r241=%rbx
3466 rorx $41,%rdx,%rbx
3467
3468 # qhasm: 4x X8 = W6sigma1 + X8
3469 # asm 1: vpaddq <W6sigma1=reg256#8,<X8=reg256#4,>X8=reg256#4
3470 # asm 2: vpaddq <W6sigma1=%ymm7,<X8=%ymm3,>X8=%ymm3
3471 vpaddq %ymm7,%ymm3,%ymm3
3472
3473 # qhasm: 2x,0 W8right19 = X8 unsigned>> 19
3474 # asm 1: vpsrlq $19,<X8=reg256#4%128,>W8right19=reg256#6%128
3475 # asm 2: vpsrlq $19,<X8=%xmm3,>W8right19=%xmm5
3476 vpsrlq $19,%xmm3,%xmm5
3477
3478 # qhasm: ch5 ^= r4
3479 # asm 1: xor <r4=int64#9,<ch5=int64#12
3480 # asm 2: xor <r4=%r11,<ch5=%r14
3481 xor %r11,%r14
3482
3483 # qhasm: r2Sigma1 ^= r218
3484 # asm 1: xor <r218=int64#13,<r2Sigma1=int64#7
3485 # asm 2: xor <r218=%r15,<r2Sigma1=%rax
3486 xor %r15,%rax
3487
3488 # qhasm: r6Sigma0 = r6>>>28
3489 # asm 1: rorx $28,<r6=int64#10,>r6Sigma0=int64#13
3490 # asm 2: rorx $28,<r6=%r12,>r6Sigma0=%r15
3491 rorx $28,%r12,%r15
3492
3493 # qhasm: r5 += ch5
3494 # asm 1: add <ch5=int64#12,<r5=int64#8
3495 # asm 2: add <ch5=%r14,<r5=%r10
3496 add %r14,%r10
3497
3498 # qhasm: 2x,0 W8left45 = X8 << 45
3499 # asm 1: vpsllq $45,<X8=reg256#4%128,>W8left45=reg256#7%128
3500 # asm 2: vpsllq $45,<X8=%xmm3,>W8left45=%xmm6
3501 vpsllq $45,%xmm3,%xmm6
3502
3503 # qhasm: r634 = r6>>>34
3504 # asm 1: rorx $34,<r6=int64#10,>r634=int64#12
3505 # asm 2: rorx $34,<r6=%r12,>r634=%r14
3506 rorx $34,%r12,%r14
3507
3508 # qhasm: r2Sigma1 ^= r241
3509 # asm 1: xor <r241=int64#14,<r2Sigma1=int64#7
3510 # asm 2: xor <r241=%rbx,<r2Sigma1=%rax
3511 xor %rbx,%rax
3512
3513 # qhasm: maj4 = r7
3514 # asm 1: mov <r7=int64#11,>maj4=int64#14
3515 # asm 2: mov <r7=%r13,>maj4=%rbx
3516 mov %r13,%rbx
3517
3518 # qhasm: 2x,0 W8right61 = X8 unsigned>> 61
3519 # asm 1: vpsrlq $61,<X8=reg256#4%128,>W8right61=reg256#8%128
3520 # asm 2: vpsrlq $61,<X8=%xmm3,>W8right61=%xmm7
3521 vpsrlq $61,%xmm3,%xmm7
3522
3523 # qhasm: maj4 ^= r6
3524 # asm 1: xor <r6=int64#10,<maj4=int64#14
3525 # asm 2: xor <r6=%r12,<maj4=%rbx
3526 xor %r12,%rbx
3527
3528 # qhasm: r6Sigma0 ^= r634
3529 # asm 1: xor <r634=int64#12,<r6Sigma0=int64#13
3530 # asm 2: xor <r634=%r14,<r6Sigma0=%r15
3531 xor %r14,%r15
3532
3533 # qhasm: 1x,0 W8sigma1 = W8right19 ^ W8left45
3534 # asm 1: vpxor <W8right19=reg256#6%128,<W8left45=reg256#7%128,>W8sigma1=reg256#6%128
3535 # asm 2: vpxor <W8right19=%xmm5,<W8left45=%xmm6,>W8sigma1=%xmm5
3536 vpxor %xmm5,%xmm6,%xmm5
3537
3538 # qhasm: r639 = r6>>>39
3539 # asm 1: rorx $39,<r6=int64#10,>r639=int64#12
3540 # asm 2: rorx $39,<r6=%r12,>r639=%r14
3541 rorx $39,%r12,%r14
3542
3543 # qhasm: 1x,0 W8sigma1 ^= W8right61
3544 # asm 1: vpxor <W8right61=reg256#8%128,<W8sigma1=reg256#6%128,<W8sigma1=reg256#6%128
3545 # asm 2: vpxor <W8right61=%xmm7,<W8sigma1=%xmm5,<W8sigma1=%xmm5
3546 vpxor %xmm7,%xmm5,%xmm5
3547
3548 # qhasm: 2x,0 W8left3 = X8 << 3
3549 # asm 1: vpsllq $3,<X8=reg256#4%128,>W8left3=reg256#7%128
3550 # asm 2: vpsllq $3,<X8=%xmm3,>W8left3=%xmm6
3551 vpsllq $3,%xmm3,%xmm6
3552
3553 # qhasm: r6Sigma0 ^= r639
3554 # asm 1: xor <r639=int64#12,<r6Sigma0=int64#13
3555 # asm 2: xor <r639=%r14,<r6Sigma0=%r15
3556 xor %r14,%r15
3557
3558 # qhasm: r5 += r2Sigma1
3559 # asm 1: add <r2Sigma1=int64#7,<r5=int64#8
3560 # asm 2: add <r2Sigma1=%rax,<r5=%r10
3561 add %rax,%r10
3562
3563 # qhasm: r1 += r5
3564 # asm 1: add <r5=int64#8,<r1=int64#4
3565 # asm 2: add <r5=%r10,<r1=%rcx
3566 add %r10,%rcx
3567
3568 # qhasm: 2x,0 W8right6 = X8 unsigned>> 6
3569 # asm 1: vpsrlq $6,<X8=reg256#4%128,>W8right6=reg256#8%128
3570 # asm 2: vpsrlq $6,<X8=%xmm3,>W8right6=%xmm7
3571 vpsrlq $6,%xmm3,%xmm7
3572
3573 # qhasm: 1x,0 W8sigma1 ^= W8left3
3574 # asm 1: vpxor <W8left3=reg256#7%128,<W8sigma1=reg256#6%128,<W8sigma1=reg256#6%128
3575 # asm 2: vpxor <W8left3=%xmm6,<W8sigma1=%xmm5,<W8sigma1=%xmm5
3576 vpxor %xmm6,%xmm5,%xmm5
3577
3578 # qhasm: r6andr7 = r7
3579 # asm 1: mov <r7=int64#11,>r6andr7=int64#7
3580 # asm 2: mov <r7=%r13,>r6andr7=%rax
3581 mov %r13,%rax
3582
3583 # qhasm: r6andr7 &= r6
3584 # asm 1: and <r6=int64#10,<r6andr7=int64#7
3585 # asm 2: and <r6=%r12,<r6andr7=%rax
3586 and %r12,%rax
3587
3588 # qhasm: r1Sigma1 = r1>>>14
3589 # asm 1: rorx $14,<r1=int64#4,>r1Sigma1=int64#12
3590 # asm 2: rorx $14,<r1=%rcx,>r1Sigma1=%r14
3591 rorx $14,%rcx,%r14
3592
3593 # qhasm: 1x,0 W8sigma1 ^= W8right6
3594 # asm 1: vpxor <W8right6=reg256#8%128,<W8sigma1=reg256#6%128,<W8sigma1=reg256#6%128
3595 # asm 2: vpxor <W8right6=%xmm7,<W8sigma1=%xmm5,<W8sigma1=%xmm5
3596 vpxor %xmm7,%xmm5,%xmm5
3597
3598 # qhasm: r5 += r6Sigma0
3599 # asm 1: add <r6Sigma0=int64#13,<r5=int64#8
3600 # asm 2: add <r6Sigma0=%r15,<r5=%r10
3601 add %r15,%r10
3602
3603 # qhasm: maj5 = r0
3604 # asm 1: mov <r0=int64#2,>maj5=int64#13
3605 # asm 2: mov <r0=%rsi,>maj5=%r15
3606 mov %rsi,%r15
3607
3608 # qhasm: maj5 &= maj4
3609 # asm 1: and <maj4=int64#14,<maj5=int64#13
3610 # asm 2: and <maj4=%rbx,<maj5=%r15
3611 and %rbx,%r15
3612
3613 # qhasm: W8sigma1 = W8sigma1[2,3,0,1]
3614 # asm 1: vpermq $0x4e,<W8sigma1=reg256#6,>W8sigma1=reg256#6
3615 # asm 2: vpermq $0x4e,<W8sigma1=%ymm5,>W8sigma1=%ymm5
3616 vpermq $0x4e,%ymm5,%ymm5
3617
3618 # qhasm: maj5 ^= r6andr7
3619 # asm 1: xor <r6andr7=int64#7,<maj5=int64#13
3620 # asm 2: xor <r6andr7=%rax,<maj5=%r15
3621 xor %rax,%r15
3622
3623 # qhasm: ch4 = r3
3624 # asm 1: mov <r3=int64#6,>ch4=int64#15
3625 # asm 2: mov <r3=%r9,>ch4=%rbp
3626 mov %r9,%rbp
3627
3628 # qhasm: ch4 ^= r2
3629 # asm 1: xor <r2=int64#3,<ch4=int64#15
3630 # asm 2: xor <r2=%rdx,<ch4=%rbp
3631 xor %rdx,%rbp
3632
3633 # qhasm: X13 = mem256[&w + 104]
3634 # asm 1: vmovupd <w=stack1280#1,>X13=reg256#7
3635 # asm 2: vmovupd <w=264(%rsp),>X13=%ymm6
3636 vmovupd 264(%rsp),%ymm6
3637
3638 # qhasm: 4x X13right1 = X13 unsigned>> 1
3639 # asm 1: vpsrlq $1,<X13=reg256#7,>X13right1=reg256#8
3640 # asm 2: vpsrlq $1,<X13=%ymm6,>X13right1=%ymm7
3641 vpsrlq $1,%ymm6,%ymm7
3642
3643 # qhasm: r5 += maj5
3644 # asm 1: add <maj5=int64#13,<r5=int64#8
3645 # asm 2: add <maj5=%r15,<r5=%r10
3646 add %r15,%r10
3647
3648 # qhasm: r118 = r1>>>18
3649 # asm 1: rorx $18,<r1=int64#4,>r118=int64#13
3650 # asm 2: rorx $18,<r1=%rcx,>r118=%r15
3651 rorx $18,%rcx,%r15
3652
3653 # qhasm: ch4 &= r1
3654 # asm 1: and <r1=int64#4,<ch4=int64#15
3655 # asm 2: and <r1=%rcx,<ch4=%rbp
3656 and %rcx,%rbp
3657
3658 # qhasm: maj4 &= r5
3659 # asm 1: and <r5=int64#8,<maj4=int64#14
3660 # asm 2: and <r5=%r10,<maj4=%rbx
3661 and %r10,%rbx
3662
3663 # qhasm: ch4 ^= r3
3664 # asm 1: xor <r3=int64#6,<ch4=int64#15
3665 # asm 2: xor <r3=%r9,<ch4=%rbp
3666 xor %r9,%rbp
3667
3668 # qhasm: r4 += wc0123[3]
3669 # asm 1: addq <wc0123=stack256#3,<r4=int64#9
3670 # asm 2: addq <wc0123=88(%rsp),<r4=%r11
3671 addq 88(%rsp),%r11
3672
3673 # qhasm: r1Sigma1 ^= r118
3674 # asm 1: xor <r118=int64#13,<r1Sigma1=int64#12
3675 # asm 2: xor <r118=%r15,<r1Sigma1=%r14
3676 xor %r15,%r14
3677
3678 # qhasm: r141 = r1>>>41
3679 # asm 1: rorx $41,<r1=int64#4,>r141=int64#13
3680 # asm 2: rorx $41,<r1=%rcx,>r141=%r15
3681 rorx $41,%rcx,%r15
3682
3683 # qhasm: 4x X8 = X8 + W8sigma1
3684 # asm 1: vpaddq <X8=reg256#4,<W8sigma1=reg256#6,>X8=reg256#4
3685 # asm 2: vpaddq <X8=%ymm3,<W8sigma1=%ymm5,>X8=%ymm3
3686 vpaddq %ymm3,%ymm5,%ymm3
3687
3688 # qhasm: mem256[&w + 64] = X8
3689 # asm 1: vmovupd <X8=reg256#4,<w=stack1280#1
3690 # asm 2: vmovupd <X8=%ymm3,<w=224(%rsp)
3691 vmovupd %ymm3,224(%rsp)
3692
3693 # qhasm: r4 += ch4
3694 # asm 1: add <ch4=int64#15,<r4=int64#9
3695 # asm 2: add <ch4=%rbp,<r4=%r11
3696 add %rbp,%r11
3697
3698 # qhasm: maj4 ^= r6andr7
3699 # asm 1: xor <r6andr7=int64#7,<maj4=int64#14
3700 # asm 2: xor <r6andr7=%rax,<maj4=%rbx
3701 xor %rax,%rbx
3702
3703 # qhasm: r5Sigma0 = r5>>>28
3704 # asm 1: rorx $28,<r5=int64#8,>r5Sigma0=int64#7
3705 # asm 2: rorx $28,<r5=%r10,>r5Sigma0=%rax
3706 rorx $28,%r10,%rax
3707
3708 # qhasm: 4x D8 = X8 + mem256[constants + 64]
3709 # asm 1: vpaddq 64(<constants=int64#5),<X8=reg256#4,>D8=reg256#6
3710 # asm 2: vpaddq 64(<constants=%r8),<X8=%ymm3,>D8=%ymm5
3711 vpaddq 64(%r8),%ymm3,%ymm5
3712
3713 # qhasm: wc891011 = D8
3714 # asm 1: vmovapd <D8=reg256#6,>wc891011=stack256#3
3715 # asm 2: vmovapd <D8=%ymm5,>wc891011=64(%rsp)
3716 vmovapd %ymm5,64(%rsp)
3717
3718 # qhasm: r534 = r5>>>34
3719 # asm 1: rorx $34,<r5=int64#8,>r534=int64#15
3720 # asm 2: rorx $34,<r5=%r10,>r534=%rbp
3721 rorx $34,%r10,%rbp
3722
3723 # qhasm: r1Sigma1 ^= r141
3724 # asm 1: xor <r141=int64#13,<r1Sigma1=int64#12
3725 # asm 2: xor <r141=%r15,<r1Sigma1=%r14
3726 xor %r15,%r14
3727
3728 # qhasm: r4 += r1Sigma1
3729 # asm 1: add <r1Sigma1=int64#12,<r4=int64#9
3730 # asm 2: add <r1Sigma1=%r14,<r4=%r11
3731 add %r14,%r11
3732
3733 # qhasm: r5Sigma0 ^= r534
3734 # asm 1: xor <r534=int64#15,<r5Sigma0=int64#7
3735 # asm 2: xor <r534=%rbp,<r5Sigma0=%rax
3736 xor %rbp,%rax
3737
3738 # qhasm: r539 = r5>>>39
3739 # asm 1: rorx $39,<r5=int64#8,>r539=int64#12
3740 # asm 2: rorx $39,<r5=%r10,>r539=%r14
3741 rorx $39,%r10,%r14
3742
3743 # qhasm: r3 += wc4567[0]
3744 # asm 1: addq <wc4567=stack256#4,<r3=int64#6
3745 # asm 2: addq <wc4567=96(%rsp),<r3=%r9
3746 addq 96(%rsp),%r9
3747
3748 # qhasm: r0 += r4
3749 # asm 1: add <r4=int64#9,<r0=int64#2
3750 # asm 2: add <r4=%r11,<r0=%rsi
3751 add %r11,%rsi
3752
3753 # qhasm: r4 += maj4
3754 # asm 1: add <maj4=int64#14,<r4=int64#9
3755 # asm 2: add <maj4=%rbx,<r4=%r11
3756 add %rbx,%r11
3757
3758 # qhasm: W10 = mem128[&w + 80],0
3759 # asm 1: vmovupd <w=stack1280#1,>W10=reg256#6%128
3760 # asm 2: vmovupd <w=240(%rsp),>W10=%xmm5
3761 vmovupd 240(%rsp),%xmm5
3762
3763 # qhasm: r5Sigma0 ^= r539
3764 # asm 1: xor <r539=int64#12,<r5Sigma0=int64#7
3765 # asm 2: xor <r539=%r14,<r5Sigma0=%rax
3766 xor %r14,%rax
3767
3768 # qhasm: r0Sigma1 = r0>>>14
3769 # asm 1: rorx $14,<r0=int64#2,>r0Sigma1=int64#12
3770 # asm 2: rorx $14,<r0=%rsi,>r0Sigma1=%r14
3771 rorx $14,%rsi,%r14
3772
3773 # qhasm: r4 += r5Sigma0
3774 # asm 1: add <r5Sigma0=int64#7,<r4=int64#9
3775 # asm 2: add <r5Sigma0=%rax,<r4=%r11
3776 add %rax,%r11
3777
3778 # qhasm: ch3 = r2
3779 # asm 1: mov <r2=int64#3,>ch3=int64#7
3780 # asm 2: mov <r2=%rdx,>ch3=%rax
3781 mov %rdx,%rax
3782
3783 # qhasm: ch3 ^= r1
3784 # asm 1: xor <r1=int64#4,<ch3=int64#7
3785 # asm 2: xor <r1=%rcx,<ch3=%rax
3786 xor %rcx,%rax
3787
3788 # qhasm: ch3 &= r0
3789 # asm 1: and <r0=int64#2,<ch3=int64#7
3790 # asm 2: and <r0=%rsi,<ch3=%rax
3791 and %rsi,%rax
3792
3793 # qhasm: 4x X13left63 = X13 << 63
3794 # asm 1: vpsllq $63,<X13=reg256#7,>X13left63=reg256#9
3795 # asm 2: vpsllq $63,<X13=%ymm6,>X13left63=%ymm8
3796 vpsllq $63,%ymm6,%ymm8
3797
3798 # qhasm: r018 = r0>>>18
3799 # asm 1: rorx $18,<r0=int64#2,>r018=int64#13
3800 # asm 2: rorx $18,<r0=%rsi,>r018=%r15
3801 rorx $18,%rsi,%r15
3802
3803 # qhasm: maj2 = r5
3804 # asm 1: mov <r5=int64#8,>maj2=int64#14
3805 # asm 2: mov <r5=%r10,>maj2=%rbx
3806 mov %r10,%rbx
3807
3808 # qhasm: maj2 ^= r4
3809 # asm 1: xor <r4=int64#9,<maj2=int64#14
3810 # asm 2: xor <r4=%r11,<maj2=%rbx
3811 xor %r11,%rbx
3812
3813 # qhasm: X13sigma0 = X13right1 ^ X13left63
3814 # asm 1: vpxor <X13right1=reg256#8,<X13left63=reg256#9,>X13sigma0=reg256#8
3815 # asm 2: vpxor <X13right1=%ymm7,<X13left63=%ymm8,>X13sigma0=%ymm7
3816 vpxor %ymm7,%ymm8,%ymm7
3817
3818 # qhasm: r041 = r0>>>41
3819 # asm 1: rorx $41,<r0=int64#2,>r041=int64#15
3820 # asm 2: rorx $41,<r0=%rsi,>r041=%rbp
3821 rorx $41,%rsi,%rbp
3822
3823 # qhasm: r0Sigma1 ^= r018
3824 # asm 1: xor <r018=int64#13,<r0Sigma1=int64#12
3825 # asm 2: xor <r018=%r15,<r0Sigma1=%r14
3826 xor %r15,%r14
3827
3828 # qhasm: ch3 ^= r2
3829 # asm 1: xor <r2=int64#3,<ch3=int64#7
3830 # asm 2: xor <r2=%rdx,<ch3=%rax
3831 xor %rdx,%rax
3832
3833 # qhasm: 4x X13right8 = X13 unsigned>> 8
3834 # asm 1: vpsrlq $8,<X13=reg256#7,>X13right8=reg256#9
3835 # asm 2: vpsrlq $8,<X13=%ymm6,>X13right8=%ymm8
3836 vpsrlq $8,%ymm6,%ymm8
3837
3838 # qhasm: r4Sigma0 = r4>>>28
3839 # asm 1: rorx $28,<r4=int64#9,>r4Sigma0=int64#13
3840 # asm 2: rorx $28,<r4=%r11,>r4Sigma0=%r15
3841 rorx $28,%r11,%r15
3842
3843 # qhasm: r0Sigma1 ^= r041
3844 # asm 1: xor <r041=int64#15,<r0Sigma1=int64#12
3845 # asm 2: xor <r041=%rbp,<r0Sigma1=%r14
3846 xor %rbp,%r14
3847
3848 # qhasm: r3 += ch3
3849 # asm 1: add <ch3=int64#7,<r3=int64#6
3850 # asm 2: add <ch3=%rax,<r3=%r9
3851 add %rax,%r9
3852
3853 # qhasm: r434 = r4>>>34
3854 # asm 1: rorx $34,<r4=int64#9,>r434=int64#7
3855 # asm 2: rorx $34,<r4=%r11,>r434=%rax
3856 rorx $34,%r11,%rax
3857
3858 # qhasm: X13sigma0 = X13sigma0 ^ X13right8
3859 # asm 1: vpxor <X13sigma0=reg256#8,<X13right8=reg256#9,>X13sigma0=reg256#8
3860 # asm 2: vpxor <X13sigma0=%ymm7,<X13right8=%ymm8,>X13sigma0=%ymm7
3861 vpxor %ymm7,%ymm8,%ymm7
3862
3863 # qhasm: r3 += r0Sigma1
3864 # asm 1: add <r0Sigma1=int64#12,<r3=int64#6
3865 # asm 2: add <r0Sigma1=%r14,<r3=%r9
3866 add %r14,%r9
3867
3868 # qhasm: maj3 = r6
3869 # asm 1: mov <r6=int64#10,>maj3=int64#12
3870 # asm 2: mov <r6=%r12,>maj3=%r14
3871 mov %r12,%r14
3872
3873 # qhasm: maj3 &= maj2
3874 # asm 1: and <maj2=int64#14,<maj3=int64#12
3875 # asm 2: and <maj2=%rbx,<maj3=%r14
3876 and %rbx,%r14
3877
3878 # qhasm: 2x,0 W10right19 = W10 unsigned>> 19
3879 # asm 1: vpsrlq $19,<W10=reg256#6%128,>W10right19=reg256#9%128
3880 # asm 2: vpsrlq $19,<W10=%xmm5,>W10right19=%xmm8
3881 vpsrlq $19,%xmm5,%xmm8
3882
3883 # qhasm: r4Sigma0 ^= r434
3884 # asm 1: xor <r434=int64#7,<r4Sigma0=int64#13
3885 # asm 2: xor <r434=%rax,<r4Sigma0=%r15
3886 xor %rax,%r15
3887
3888 # qhasm: r7 += r3
3889 # asm 1: add <r3=int64#6,<r7=int64#11
3890 # asm 2: add <r3=%r9,<r7=%r13
3891 add %r9,%r13
3892
3893 # qhasm: r439 = r4>>>39
3894 # asm 1: rorx $39,<r4=int64#9,>r439=int64#7
3895 # asm 2: rorx $39,<r4=%r11,>r439=%rax
3896 rorx $39,%r11,%rax
3897
3898 # qhasm: 4x X13left56 = X13 << 56
3899 # asm 1: vpsllq $56,<X13=reg256#7,>X13left56=reg256#10
3900 # asm 2: vpsllq $56,<X13=%ymm6,>X13left56=%ymm9
3901 vpsllq $56,%ymm6,%ymm9
3902
3903 # qhasm: r4Sigma0 ^= r439
3904 # asm 1: xor <r439=int64#7,<r4Sigma0=int64#13
3905 # asm 2: xor <r439=%rax,<r4Sigma0=%r15
3906 xor %rax,%r15
3907
3908 # qhasm: r2 += wc4567[1]
3909 # asm 1: addq <wc4567=stack256#4,<r2=int64#3
3910 # asm 2: addq <wc4567=104(%rsp),<r2=%rdx
3911 addq 104(%rsp),%rdx
3912
3913 # qhasm: r4andr5 = r5
3914 # asm 1: mov <r5=int64#8,>r4andr5=int64#7
3915 # asm 2: mov <r5=%r10,>r4andr5=%rax
3916 mov %r10,%rax
3917
3918 # qhasm: 2x,0 W10left45 = W10 << 45
3919 # asm 1: vpsllq $45,<W10=reg256#6%128,>W10left45=reg256#11%128
3920 # asm 2: vpsllq $45,<W10=%xmm5,>W10left45=%xmm10
3921 vpsllq $45,%xmm5,%xmm10
3922
3923 # qhasm: r4andr5 &= r4
3924 # asm 1: and <r4=int64#9,<r4andr5=int64#7
3925 # asm 2: and <r4=%r11,<r4andr5=%rax
3926 and %r11,%rax
3927
3928 # qhasm: r3 += r4Sigma0
3929 # asm 1: add <r4Sigma0=int64#13,<r3=int64#6
3930 # asm 2: add <r4Sigma0=%r15,<r3=%r9
3931 add %r15,%r9
3932
3933 # qhasm: maj3 ^= r4andr5
3934 # asm 1: xor <r4andr5=int64#7,<maj3=int64#12
3935 # asm 2: xor <r4andr5=%rax,<maj3=%r14
3936 xor %rax,%r14
3937
3938 # qhasm: ch2 = r1
3939 # asm 1: mov <r1=int64#4,>ch2=int64#13
3940 # asm 2: mov <r1=%rcx,>ch2=%r15
3941 mov %rcx,%r15
3942
3943 # qhasm: 2x,0 W10right61 = W10 unsigned>> 61
3944 # asm 1: vpsrlq $61,<W10=reg256#6%128,>W10right61=reg256#12%128
3945 # asm 2: vpsrlq $61,<W10=%xmm5,>W10right61=%xmm11
3946 vpsrlq $61,%xmm5,%xmm11
3947
3948 # qhasm: ch2 ^= r0
3949 # asm 1: xor <r0=int64#2,<ch2=int64#13
3950 # asm 2: xor <r0=%rsi,<ch2=%r15
3951 xor %rsi,%r15
3952
3953 # qhasm: X13sigma0 = X13sigma0 ^ X13left56
3954 # asm 1: vpxor <X13sigma0=reg256#8,<X13left56=reg256#10,>X13sigma0=reg256#8
3955 # asm 2: vpxor <X13sigma0=%ymm7,<X13left56=%ymm9,>X13sigma0=%ymm7
3956 vpxor %ymm7,%ymm9,%ymm7
3957
3958 # qhasm: r7Sigma1 = r7>>>14
3959 # asm 1: rorx $14,<r7=int64#11,>r7Sigma1=int64#15
3960 # asm 2: rorx $14,<r7=%r13,>r7Sigma1=%rbp
3961 rorx $14,%r13,%rbp
3962
3963 # qhasm: 4x X13right7 = X13 unsigned>> 7
3964 # asm 1: vpsrlq $7,<X13=reg256#7,>X13right7=reg256#7
3965 # asm 2: vpsrlq $7,<X13=%ymm6,>X13right7=%ymm6
3966 vpsrlq $7,%ymm6,%ymm6
3967
3968 # qhasm: r3 += maj3
3969 # asm 1: add <maj3=int64#12,<r3=int64#6
3970 # asm 2: add <maj3=%r14,<r3=%r9
3971 add %r14,%r9
3972
3973 # qhasm: r718 = r7>>>18
3974 # asm 1: rorx $18,<r7=int64#11,>r718=int64#12
3975 # asm 2: rorx $18,<r7=%r13,>r718=%r14
3976 rorx $18,%r13,%r14
3977
3978 # qhasm: 1x,0 W10sigma1 = W10right19 ^ W10left45
3979 # asm 1: vpxor <W10right19=reg256#9%128,<W10left45=reg256#11%128,>W10sigma1=reg256#9%128
3980 # asm 2: vpxor <W10right19=%xmm8,<W10left45=%xmm10,>W10sigma1=%xmm8
3981 vpxor %xmm8,%xmm10,%xmm8
3982
3983 # qhasm: ch2 &= r7
3984 # asm 1: and <r7=int64#11,<ch2=int64#13
3985 # asm 2: and <r7=%r13,<ch2=%r15
3986 and %r13,%r15
3987
3988 # qhasm: r7Sigma1 ^= r718
3989 # asm 1: xor <r718=int64#12,<r7Sigma1=int64#15
3990 # asm 2: xor <r718=%r14,<r7Sigma1=%rbp
3991 xor %r14,%rbp
3992
3993 # qhasm: r741 = r7>>>41
3994 # asm 1: rorx $41,<r7=int64#11,>r741=int64#12
3995 # asm 2: rorx $41,<r7=%r13,>r741=%r14
3996 rorx $41,%r13,%r14
3997
3998 # qhasm: X13sigma0 = X13sigma0 ^ X13right7
3999 # asm 1: vpxor <X13sigma0=reg256#8,<X13right7=reg256#7,>X13sigma0=reg256#7
4000 # asm 2: vpxor <X13sigma0=%ymm7,<X13right7=%ymm6,>X13sigma0=%ymm6
4001 vpxor %ymm7,%ymm6,%ymm6
4002
4003 # qhasm: maj2 &= r3
4004 # asm 1: and <r3=int64#6,<maj2=int64#14
4005 # asm 2: and <r3=%r9,<maj2=%rbx
4006 and %r9,%rbx
4007
4008 # qhasm: 1x,0 W10sigma1 ^= W10right61
4009 # asm 1: vpxor <W10right61=reg256#12%128,<W10sigma1=reg256#9%128,<W10sigma1=reg256#9%128
4010 # asm 2: vpxor <W10right61=%xmm11,<W10sigma1=%xmm8,<W10sigma1=%xmm8
4011 vpxor %xmm11,%xmm8,%xmm8
4012
4013 # qhasm: 4x X12 = X12 + X13sigma0
4014 # asm 1: vpaddq <X12=reg256#5,<X13sigma0=reg256#7,>X12=reg256#5
4015 # asm 2: vpaddq <X12=%ymm4,<X13sigma0=%ymm6,>X12=%ymm4
4016 vpaddq %ymm4,%ymm6,%ymm4
4017
4018 # qhasm: r7Sigma1 ^= r741
4019 # asm 1: xor <r741=int64#12,<r7Sigma1=int64#15
4020 # asm 2: xor <r741=%r14,<r7Sigma1=%rbp
4021 xor %r14,%rbp
4022
4023 # qhasm: maj2 ^= r4andr5
4024 # asm 1: xor <r4andr5=int64#7,<maj2=int64#14
4025 # asm 2: xor <r4andr5=%rax,<maj2=%rbx
4026 xor %rax,%rbx
4027
4028 # qhasm: 2x,0 W10left3 = W10 << 3
4029 # asm 1: vpsllq $3,<W10=reg256#6%128,>W10left3=reg256#7%128
4030 # asm 2: vpsllq $3,<W10=%xmm5,>W10left3=%xmm6
4031 vpsllq $3,%xmm5,%xmm6
4032
4033 # qhasm: r3Sigma0 = r3>>>28
4034 # asm 1: rorx $28,<r3=int64#6,>r3Sigma0=int64#7
4035 # asm 2: rorx $28,<r3=%r9,>r3Sigma0=%rax
4036 rorx $28,%r9,%rax
4037
4038 # qhasm: ch2 ^= r1
4039 # asm 1: xor <r1=int64#4,<ch2=int64#13
4040 # asm 2: xor <r1=%rcx,<ch2=%r15
4041 xor %rcx,%r15
4042
4043 # qhasm: r2 += r7Sigma1
4044 # asm 1: add <r7Sigma1=int64#15,<r2=int64#3
4045 # asm 2: add <r7Sigma1=%rbp,<r2=%rdx
4046 add %rbp,%rdx
4047
4048 # qhasm: r334 = r3>>>34
4049 # asm 1: rorx $34,<r3=int64#6,>r334=int64#12
4050 # asm 2: rorx $34,<r3=%r9,>r334=%r14
4051 rorx $34,%r9,%r14
4052
4053 # qhasm: 4x X12 = X12 + mem256[&w + 40]
4054 # asm 1: vpaddq <w=stack1280#1,<X12=reg256#5,>X12=reg256#5
4055 # asm 2: vpaddq <w=200(%rsp),<X12=%ymm4,>X12=%ymm4
4056 vpaddq 200(%rsp),%ymm4,%ymm4
4057
4058 # qhasm: r1 += wc4567[2]
4059 # asm 1: addq <wc4567=stack256#4,<r1=int64#4
4060 # asm 2: addq <wc4567=112(%rsp),<r1=%rcx
4061 addq 112(%rsp),%rcx
4062
4063 # qhasm: r2 += ch2
4064 # asm 1: add <ch2=int64#13,<r2=int64#3
4065 # asm 2: add <ch2=%r15,<r2=%rdx
4066 add %r15,%rdx
4067
4068 # qhasm: 1x,0 W10sigma1 ^= W10left3
4069 # asm 1: vpxor <W10left3=reg256#7%128,<W10sigma1=reg256#9%128,<W10sigma1=reg256#9%128
4070 # asm 2: vpxor <W10left3=%xmm6,<W10sigma1=%xmm8,<W10sigma1=%xmm8
4071 vpxor %xmm6,%xmm8,%xmm8
4072
4073 # qhasm: r3Sigma0 ^= r334
4074 # asm 1: xor <r334=int64#12,<r3Sigma0=int64#7
4075 # asm 2: xor <r334=%r14,<r3Sigma0=%rax
4076 xor %r14,%rax
4077
4078 # qhasm: r339 = r3>>>39
4079 # asm 1: rorx $39,<r3=int64#6,>r339=int64#12
4080 # asm 2: rorx $39,<r3=%r9,>r339=%r14
4081 rorx $39,%r9,%r14
4082
4083 # qhasm: 2x,0 W10right6 = W10 unsigned>> 6
4084 # asm 1: vpsrlq $6,<W10=reg256#6%128,>W10right6=reg256#6%128
4085 # asm 2: vpsrlq $6,<W10=%xmm5,>W10right6=%xmm5
4086 vpsrlq $6,%xmm5,%xmm5
4087
4088 # qhasm: r6 += r2
4089 # asm 1: add <r2=int64#3,<r6=int64#10
4090 # asm 2: add <r2=%rdx,<r6=%r12
4091 add %rdx,%r12
4092
4093 # qhasm: r3Sigma0 ^= r339
4094 # asm 1: xor <r339=int64#12,<r3Sigma0=int64#7
4095 # asm 2: xor <r339=%r14,<r3Sigma0=%rax
4096 xor %r14,%rax
4097
4098 # qhasm: ch1 = r0
4099 # asm 1: mov <r0=int64#2,>ch1=int64#12
4100 # asm 2: mov <r0=%rsi,>ch1=%r14
4101 mov %rsi,%r14
4102
4103 # qhasm: r2 += maj2
4104 # asm 1: add <maj2=int64#14,<r2=int64#3
4105 # asm 2: add <maj2=%rbx,<r2=%rdx
4106 add %rbx,%rdx
4107
4108 # qhasm: ch1 ^= r7
4109 # asm 1: xor <r7=int64#11,<ch1=int64#12
4110 # asm 2: xor <r7=%r13,<ch1=%r14
4111 xor %r13,%r14
4112
4113 # qhasm: r2 += r3Sigma0
4114 # asm 1: add <r3Sigma0=int64#7,<r2=int64#3
4115 # asm 2: add <r3Sigma0=%rax,<r2=%rdx
4116 add %rax,%rdx
4117
4118 # qhasm: r6Sigma1 = r6>>>14
4119 # asm 1: rorx $14,<r6=int64#10,>r6Sigma1=int64#7
4120 # asm 2: rorx $14,<r6=%r12,>r6Sigma1=%rax
4121 rorx $14,%r12,%rax
4122
4123 # qhasm: 1x,0 W10sigma1 ^= W10right6
4124 # asm 1: vpxor <W10right6=reg256#6%128,<W10sigma1=reg256#9%128,<W10sigma1=reg256#9%128
4125 # asm 2: vpxor <W10right6=%xmm5,<W10sigma1=%xmm8,<W10sigma1=%xmm8
4126 vpxor %xmm5,%xmm8,%xmm8
4127
4128 # qhasm: ch1 &= r6
4129 # asm 1: and <r6=int64#10,<ch1=int64#12
4130 # asm 2: and <r6=%r12,<ch1=%r14
4131 and %r12,%r14
4132
4133 # qhasm: r618 = r6>>>18
4134 # asm 1: rorx $18,<r6=int64#10,>r618=int64#13
4135 # asm 2: rorx $18,<r6=%r12,>r618=%r15
4136 rorx $18,%r12,%r15
4137
4138 # qhasm: ch1 ^= r0
4139 # asm 1: xor <r0=int64#2,<ch1=int64#12
4140 # asm 2: xor <r0=%rsi,<ch1=%r14
4141 xor %rsi,%r14
4142
4143 # qhasm: 4x X12 = W10sigma1 + X12
4144 # asm 1: vpaddq <W10sigma1=reg256#9,<X12=reg256#5,>X12=reg256#5
4145 # asm 2: vpaddq <W10sigma1=%ymm8,<X12=%ymm4,>X12=%ymm4
4146 vpaddq %ymm8,%ymm4,%ymm4
4147
4148 # qhasm: r641 = r6>>>41
4149 # asm 1: rorx $41,<r6=int64#10,>r641=int64#14
4150 # asm 2: rorx $41,<r6=%r12,>r641=%rbx
4151 rorx $41,%r12,%rbx
4152
4153 # qhasm: 2x,0 W12right19 = X12 unsigned>> 19
4154 # asm 1: vpsrlq $19,<X12=reg256#5%128,>W12right19=reg256#6%128
4155 # asm 2: vpsrlq $19,<X12=%xmm4,>W12right19=%xmm5
4156 vpsrlq $19,%xmm4,%xmm5
4157
4158 # qhasm: r6Sigma1 ^= r618
4159 # asm 1: xor <r618=int64#13,<r6Sigma1=int64#7
4160 # asm 2: xor <r618=%r15,<r6Sigma1=%rax
4161 xor %r15,%rax
4162
4163 # qhasm: r1 += ch1
4164 # asm 1: add <ch1=int64#12,<r1=int64#4
4165 # asm 2: add <ch1=%r14,<r1=%rcx
4166 add %r14,%rcx
4167
4168 # qhasm: r2Sigma0 = r2>>>28
4169 # asm 1: rorx $28,<r2=int64#3,>r2Sigma0=int64#12
4170 # asm 2: rorx $28,<r2=%rdx,>r2Sigma0=%r14
4171 rorx $28,%rdx,%r14
4172
4173 # qhasm: 2x,0 W12left45 = X12 << 45
4174 # asm 1: vpsllq $45,<X12=reg256#5%128,>W12left45=reg256#7%128
4175 # asm 2: vpsllq $45,<X12=%xmm4,>W12left45=%xmm6
4176 vpsllq $45,%xmm4,%xmm6
4177
4178 # qhasm: r6Sigma1 ^= r641
4179 # asm 1: xor <r641=int64#14,<r6Sigma1=int64#7
4180 # asm 2: xor <r641=%rbx,<r6Sigma1=%rax
4181 xor %rbx,%rax
4182
4183 # qhasm: r234 = r2>>>34
4184 # asm 1: rorx $34,<r2=int64#3,>r234=int64#13
4185 # asm 2: rorx $34,<r2=%rdx,>r234=%r15
4186 rorx $34,%rdx,%r15
4187
4188 # qhasm: maj0 = r3
4189 # asm 1: mov <r3=int64#6,>maj0=int64#14
4190 # asm 2: mov <r3=%r9,>maj0=%rbx
4191 mov %r9,%rbx
4192
4193 # qhasm: maj0 ^= r2
4194 # asm 1: xor <r2=int64#3,<maj0=int64#14
4195 # asm 2: xor <r2=%rdx,<maj0=%rbx
4196 xor %rdx,%rbx
4197
4198 # qhasm: r2Sigma0 ^= r234
4199 # asm 1: xor <r234=int64#13,<r2Sigma0=int64#12
4200 # asm 2: xor <r234=%r15,<r2Sigma0=%r14
4201 xor %r15,%r14
4202
4203 # qhasm: 2x,0 W12right61 = X12 unsigned>> 61
4204 # asm 1: vpsrlq $61,<X12=reg256#5%128,>W12right61=reg256#8%128
4205 # asm 2: vpsrlq $61,<X12=%xmm4,>W12right61=%xmm7
4206 vpsrlq $61,%xmm4,%xmm7
4207
4208 # qhasm: r239 = r2>>>39
4209 # asm 1: rorx $39,<r2=int64#3,>r239=int64#13
4210 # asm 2: rorx $39,<r2=%rdx,>r239=%r15
4211 rorx $39,%rdx,%r15
4212
4213 # qhasm: 1x,0 W12sigma1 = W12right19 ^ W12left45
4214 # asm 1: vpxor <W12right19=reg256#6%128,<W12left45=reg256#7%128,>W12sigma1=reg256#6%128
4215 # asm 2: vpxor <W12right19=%xmm5,<W12left45=%xmm6,>W12sigma1=%xmm5
4216 vpxor %xmm5,%xmm6,%xmm5
4217
4218 # qhasm: 2x,0 W12left3 = X12 << 3
4219 # asm 1: vpsllq $3,<X12=reg256#5%128,>W12left3=reg256#7%128
4220 # asm 2: vpsllq $3,<X12=%xmm4,>W12left3=%xmm6
4221 vpsllq $3,%xmm4,%xmm6
4222
4223 # qhasm: 1x,0 W12sigma1 ^= W12right61
4224 # asm 1: vpxor <W12right61=reg256#8%128,<W12sigma1=reg256#6%128,<W12sigma1=reg256#6%128
4225 # asm 2: vpxor <W12right61=%xmm7,<W12sigma1=%xmm5,<W12sigma1=%xmm5
4226 vpxor %xmm7,%xmm5,%xmm5
4227
4228 # qhasm: r2Sigma0 ^= r239
4229 # asm 1: xor <r239=int64#13,<r2Sigma0=int64#12
4230 # asm 2: xor <r239=%r15,<r2Sigma0=%r14
4231 xor %r15,%r14
4232
4233 # qhasm: r1 += r6Sigma1
4234 # asm 1: add <r6Sigma1=int64#7,<r1=int64#4
4235 # asm 2: add <r6Sigma1=%rax,<r1=%rcx
4236 add %rax,%rcx
4237
4238 # qhasm: 2x,0 W12right6 = X12 unsigned>> 6
4239 # asm 1: vpsrlq $6,<X12=reg256#5%128,>W12right6=reg256#8%128
4240 # asm 2: vpsrlq $6,<X12=%xmm4,>W12right6=%xmm7
4241 vpsrlq $6,%xmm4,%xmm7
4242
4243 # qhasm: r5 += r1
4244 # asm 1: add <r1=int64#4,<r5=int64#8
4245 # asm 2: add <r1=%rcx,<r5=%r10
4246 add %rcx,%r10
4247
4248 # qhasm: 1x,0 W12sigma1 ^= W12left3
4249 # asm 1: vpxor <W12left3=reg256#7%128,<W12sigma1=reg256#6%128,<W12sigma1=reg256#6%128
4250 # asm 2: vpxor <W12left3=%xmm6,<W12sigma1=%xmm5,<W12sigma1=%xmm5
4251 vpxor %xmm6,%xmm5,%xmm5
4252
4253 # qhasm: r2andr3 = r3
4254 # asm 1: mov <r3=int64#6,>r2andr3=int64#7
4255 # asm 2: mov <r3=%r9,>r2andr3=%rax
4256 mov %r9,%rax
4257
4258 # qhasm: r2andr3 &= r2
4259 # asm 1: and <r2=int64#3,<r2andr3=int64#7
4260 # asm 2: and <r2=%rdx,<r2andr3=%rax
4261 and %rdx,%rax
4262
4263 # qhasm: r5Sigma1 = r5>>>14
4264 # asm 1: rorx $14,<r5=int64#8,>r5Sigma1=int64#13
4265 # asm 2: rorx $14,<r5=%r10,>r5Sigma1=%r15
4266 rorx $14,%r10,%r15
4267
4268 # qhasm: r1 += r2Sigma0
4269 # asm 1: add <r2Sigma0=int64#12,<r1=int64#4
4270 # asm 2: add <r2Sigma0=%r14,<r1=%rcx
4271 add %r14,%rcx
4272
4273 # qhasm: 1x,0 W12sigma1 ^= W12right6
4274 # asm 1: vpxor <W12right6=reg256#8%128,<W12sigma1=reg256#6%128,<W12sigma1=reg256#6%128
4275 # asm 2: vpxor <W12right6=%xmm7,<W12sigma1=%xmm5,<W12sigma1=%xmm5
4276 vpxor %xmm7,%xmm5,%xmm5
4277
4278 # qhasm: maj1 = r4
4279 # asm 1: mov <r4=int64#9,>maj1=int64#12
4280 # asm 2: mov <r4=%r11,>maj1=%r14
4281 mov %r11,%r14
4282
4283 # qhasm: maj1 &= maj0
4284 # asm 1: and <maj0=int64#14,<maj1=int64#12
4285 # asm 2: and <maj0=%rbx,<maj1=%r14
4286 and %rbx,%r14
4287
4288 # qhasm: W12sigma1 = W12sigma1[2,3,0,1]
4289 # asm 1: vpermq $0x4e,<W12sigma1=reg256#6,>W12sigma1=reg256#6
4290 # asm 2: vpermq $0x4e,<W12sigma1=%ymm5,>W12sigma1=%ymm5
4291 vpermq $0x4e,%ymm5,%ymm5
4292
4293 # qhasm: maj1 ^= r2andr3
4294 # asm 1: xor <r2andr3=int64#7,<maj1=int64#12
4295 # asm 2: xor <r2andr3=%rax,<maj1=%r14
4296 xor %rax,%r14
4297
4298 # qhasm: ch0 = r7
4299 # asm 1: mov <r7=int64#11,>ch0=int64#15
4300 # asm 2: mov <r7=%r13,>ch0=%rbp
4301 mov %r13,%rbp
4302
4303 # qhasm: ch0 ^= r6
4304 # asm 1: xor <r6=int64#10,<ch0=int64#15
4305 # asm 2: xor <r6=%r12,<ch0=%rbp
4306 xor %r12,%rbp
4307
4308 # qhasm: r1 += maj1
4309 # asm 1: add <maj1=int64#12,<r1=int64#4
4310 # asm 2: add <maj1=%r14,<r1=%rcx
4311 add %r14,%rcx
4312
4313 # qhasm: ch0 &= r5
4314 # asm 1: and <r5=int64#8,<ch0=int64#15
4315 # asm 2: and <r5=%r10,<ch0=%rbp
4316 and %r10,%rbp
4317
4318 # qhasm: r518 = r5>>>18
4319 # asm 1: rorx $18,<r5=int64#8,>r518=int64#12
4320 # asm 2: rorx $18,<r5=%r10,>r518=%r14
4321 rorx $18,%r10,%r14
4322
4323 # qhasm: maj0 &= r1
4324 # asm 1: and <r1=int64#4,<maj0=int64#14
4325 # asm 2: and <r1=%rcx,<maj0=%rbx
4326 and %rcx,%rbx
4327
4328 # qhasm: r0 += wc4567[3]
4329 # asm 1: addq <wc4567=stack256#4,<r0=int64#2
4330 # asm 2: addq <wc4567=120(%rsp),<r0=%rsi
4331 addq 120(%rsp),%rsi
4332
4333 # qhasm: ch0 ^= r7
4334 # asm 1: xor <r7=int64#11,<ch0=int64#15
4335 # asm 2: xor <r7=%r13,<ch0=%rbp
4336 xor %r13,%rbp
4337
4338 # qhasm: r5Sigma1 ^= r518
4339 # asm 1: xor <r518=int64#12,<r5Sigma1=int64#13
4340 # asm 2: xor <r518=%r14,<r5Sigma1=%r15
4341 xor %r14,%r15
4342
4343 # qhasm: r541 = r5>>>41
4344 # asm 1: rorx $41,<r5=int64#8,>r541=int64#12
4345 # asm 2: rorx $41,<r5=%r10,>r541=%r14
4346 rorx $41,%r10,%r14
4347
4348 # qhasm: 4x X12 = X12 + W12sigma1
4349 # asm 1: vpaddq <X12=reg256#5,<W12sigma1=reg256#6,>X12=reg256#5
4350 # asm 2: vpaddq <X12=%ymm4,<W12sigma1=%ymm5,>X12=%ymm4
4351 vpaddq %ymm4,%ymm5,%ymm4
4352
4353 # qhasm: r0 += ch0
4354 # asm 1: add <ch0=int64#15,<r0=int64#2
4355 # asm 2: add <ch0=%rbp,<r0=%rsi
4356 add %rbp,%rsi
4357
4358 # qhasm: mem256[&w + 96] = X12
4359 # asm 1: vmovupd <X12=reg256#5,<w=stack1280#1
4360 # asm 2: vmovupd <X12=%ymm4,<w=256(%rsp)
4361 vmovupd %ymm4,256(%rsp)
4362
4363 # qhasm: maj0 ^= r2andr3
4364 # asm 1: xor <r2andr3=int64#7,<maj0=int64#14
4365 # asm 2: xor <r2andr3=%rax,<maj0=%rbx
4366 xor %rax,%rbx
4367
4368 # qhasm: r1Sigma0 = r1>>>28
4369 # asm 1: rorx $28,<r1=int64#4,>r1Sigma0=int64#7
4370 # asm 2: rorx $28,<r1=%rcx,>r1Sigma0=%rax
4371 rorx $28,%rcx,%rax
4372
4373 # qhasm: 4x D12 = X12 + mem256[constants + 96]
4374 # asm 1: vpaddq 96(<constants=int64#5),<X12=reg256#5,>D12=reg256#6
4375 # asm 2: vpaddq 96(<constants=%r8),<X12=%ymm4,>D12=%ymm5
4376 vpaddq 96(%r8),%ymm4,%ymm5
4377
4378 # qhasm: wc12131415 = D12
4379 # asm 1: vmovapd <D12=reg256#6,>wc12131415=stack256#5
4380 # asm 2: vmovapd <D12=%ymm5,>wc12131415=128(%rsp)
4381 vmovapd %ymm5,128(%rsp)
4382
4383 # qhasm: r5Sigma1 ^= r541
4384 # asm 1: xor <r541=int64#12,<r5Sigma1=int64#13
4385 # asm 2: xor <r541=%r14,<r5Sigma1=%r15
4386 xor %r14,%r15
4387
4388 # qhasm: r134 = r1>>>34
4389 # asm 1: rorx $34,<r1=int64#4,>r134=int64#12
4390 # asm 2: rorx $34,<r1=%rcx,>r134=%r14
4391 rorx $34,%rcx,%r14
4392
4393 # qhasm: r0 += r5Sigma1
4394 # asm 1: add <r5Sigma1=int64#13,<r0=int64#2
4395 # asm 2: add <r5Sigma1=%r15,<r0=%rsi
4396 add %r15,%rsi
4397
4398 # qhasm: r1Sigma0 ^= r134
4399 # asm 1: xor <r134=int64#12,<r1Sigma0=int64#7
4400 # asm 2: xor <r134=%r14,<r1Sigma0=%rax
4401 xor %r14,%rax
4402
4403 # qhasm: r139 = r1>>>39
4404 # asm 1: rorx $39,<r1=int64#4,>r139=int64#12
4405 # asm 2: rorx $39,<r1=%rcx,>r139=%r14
4406 rorx $39,%rcx,%r14
4407
4408 # qhasm: r4 += r0
4409 # asm 1: add <r0=int64#2,<r4=int64#9
4410 # asm 2: add <r0=%rsi,<r4=%r11
4411 add %rsi,%r11
4412
4413 # qhasm: r0 += maj0
4414 # asm 1: add <maj0=int64#14,<r0=int64#2
4415 # asm 2: add <maj0=%rbx,<r0=%rsi
4416 add %rbx,%rsi
4417
4418 # qhasm: r1Sigma0 ^= r139
4419 # asm 1: xor <r139=int64#12,<r1Sigma0=int64#7
4420 # asm 2: xor <r139=%r14,<r1Sigma0=%rax
4421 xor %r14,%rax
4422
4423 # qhasm: r0 += r1Sigma0
4424 # asm 1: add <r1Sigma0=int64#7,<r0=int64#2
4425 # asm 2: add <r1Sigma0=%rax,<r0=%rsi
4426 add %rax,%rsi
4427
4428 # qhasm: =? i -= 1
4429 # asm 1: dec <i=int64#1
4430 # asm 2: dec <i=%rdi
4431 dec %rdi
4432 # comment:fp stack unchanged by jump
4433
4434 # qhasm: goto innerloop if !=
4435 jne ._innerloop
4436
4437 # qhasm: r7 += wc891011[0]
4438 # asm 1: addq <wc891011=stack256#3,<r7=int64#11
4439 # asm 2: addq <wc891011=64(%rsp),<r7=%r13
4440 addq 64(%rsp),%r13
4441
4442 # qhasm: r4Sigma1 = r4>>>14
4443 # asm 1: rorx $14,<r4=int64#9,>r4Sigma1=int64#1
4444 # asm 2: rorx $14,<r4=%r11,>r4Sigma1=%rdi
4445 rorx $14,%r11,%rdi
4446
4447 # qhasm: ch7 = r6
4448 # asm 1: mov <r6=int64#10,>ch7=int64#5
4449 # asm 2: mov <r6=%r12,>ch7=%r8
4450 mov %r12,%r8
4451
4452 # qhasm: r418 = r4>>>18
4453 # asm 1: rorx $18,<r4=int64#9,>r418=int64#7
4454 # asm 2: rorx $18,<r4=%r11,>r418=%rax
4455 rorx $18,%r11,%rax
4456
4457 # qhasm: ch7 ^= r5
4458 # asm 1: xor <r5=int64#8,<ch7=int64#5
4459 # asm 2: xor <r5=%r10,<ch7=%r8
4460 xor %r10,%r8
4461
4462 # qhasm: r441 = r4>>>41
4463 # asm 1: rorx $41,<r4=int64#9,>r441=int64#12
4464 # asm 2: rorx $41,<r4=%r11,>r441=%r14
4465 rorx $41,%r11,%r14
4466
4467 # qhasm: r4Sigma1 ^= r418
4468 # asm 1: xor <r418=int64#7,<r4Sigma1=int64#1
4469 # asm 2: xor <r418=%rax,<r4Sigma1=%rdi
4470 xor %rax,%rdi
4471
4472 # qhasm: ch7 &= r4
4473 # asm 1: and <r4=int64#9,<ch7=int64#5
4474 # asm 2: and <r4=%r11,<ch7=%r8
4475 and %r11,%r8
4476
4477 # qhasm: r0Sigma0 = r0>>>28
4478 # asm 1: rorx $28,<r0=int64#2,>r0Sigma0=int64#7
4479 # asm 2: rorx $28,<r0=%rsi,>r0Sigma0=%rax
4480 rorx $28,%rsi,%rax
4481
4482 # qhasm: r4Sigma1 ^= r441
4483 # asm 1: xor <r441=int64#12,<r4Sigma1=int64#1
4484 # asm 2: xor <r441=%r14,<r4Sigma1=%rdi
4485 xor %r14,%rdi
4486
4487 # qhasm: ch7 ^= r6
4488 # asm 1: xor <r6=int64#10,<ch7=int64#5
4489 # asm 2: xor <r6=%r12,<ch7=%r8
4490 xor %r12,%r8
4491
4492 # qhasm: r034 = r0>>>34
4493 # asm 1: rorx $34,<r0=int64#2,>r034=int64#12
4494 # asm 2: rorx $34,<r0=%rsi,>r034=%r14
4495 rorx $34,%rsi,%r14
4496
4497 # qhasm: maj6 = r1
4498 # asm 1: mov <r1=int64#4,>maj6=int64#13
4499 # asm 2: mov <r1=%rcx,>maj6=%r15
4500 mov %rcx,%r15
4501
4502 # qhasm: maj6 ^= r0
4503 # asm 1: xor <r0=int64#2,<maj6=int64#13
4504 # asm 2: xor <r0=%rsi,<maj6=%r15
4505 xor %rsi,%r15
4506
4507 # qhasm: r0Sigma0 ^= r034
4508 # asm 1: xor <r034=int64#12,<r0Sigma0=int64#7
4509 # asm 2: xor <r034=%r14,<r0Sigma0=%rax
4510 xor %r14,%rax
4511
4512 # qhasm: r7 += ch7
4513 # asm 1: add <ch7=int64#5,<r7=int64#11
4514 # asm 2: add <ch7=%r8,<r7=%r13
4515 add %r8,%r13
4516
4517 # qhasm: r0andr1 = r1
4518 # asm 1: mov <r1=int64#4,>r0andr1=int64#5
4519 # asm 2: mov <r1=%rcx,>r0andr1=%r8
4520 mov %rcx,%r8
4521
4522 # qhasm: r039 = r0>>>39
4523 # asm 1: rorx $39,<r0=int64#2,>r039=int64#12
4524 # asm 2: rorx $39,<r0=%rsi,>r039=%r14
4525 rorx $39,%rsi,%r14
4526
4527 # qhasm: r0andr1 &= r0
4528 # asm 1: and <r0=int64#2,<r0andr1=int64#5
4529 # asm 2: and <r0=%rsi,<r0andr1=%r8
4530 and %rsi,%r8
4531
4532 # qhasm: r0Sigma0 ^= r039
4533 # asm 1: xor <r039=int64#12,<r0Sigma0=int64#7
4534 # asm 2: xor <r039=%r14,<r0Sigma0=%rax
4535 xor %r14,%rax
4536
4537 # qhasm: r7 += r4Sigma1
4538 # asm 1: add <r4Sigma1=int64#1,<r7=int64#11
4539 # asm 2: add <r4Sigma1=%rdi,<r7=%r13
4540 add %rdi,%r13
4541
4542 # qhasm: maj7 = r2
4543 # asm 1: mov <r2=int64#3,>maj7=int64#1
4544 # asm 2: mov <r2=%rdx,>maj7=%rdi
4545 mov %rdx,%rdi
4546
4547 # qhasm: r6 += wc891011[1]
4548 # asm 1: addq <wc891011=stack256#3,<r6=int64#10
4549 # asm 2: addq <wc891011=72(%rsp),<r6=%r12
4550 addq 72(%rsp),%r12
4551
4552 # qhasm: maj7 &= maj6
4553 # asm 1: and <maj6=int64#13,<maj7=int64#1
4554 # asm 2: and <maj6=%r15,<maj7=%rdi
4555 and %r15,%rdi
4556
4557 # qhasm: r3 += r7
4558 # asm 1: add <r7=int64#11,<r3=int64#6
4559 # asm 2: add <r7=%r13,<r3=%r9
4560 add %r13,%r9
4561
4562 # qhasm: r7 += r0Sigma0
4563 # asm 1: add <r0Sigma0=int64#7,<r7=int64#11
4564 # asm 2: add <r0Sigma0=%rax,<r7=%r13
4565 add %rax,%r13
4566
4567 # qhasm: ch6 = r5
4568 # asm 1: mov <r5=int64#8,>ch6=int64#7
4569 # asm 2: mov <r5=%r10,>ch6=%rax
4570 mov %r10,%rax
4571
4572 # qhasm: maj7 ^= r0andr1
4573 # asm 1: xor <r0andr1=int64#5,<maj7=int64#1
4574 # asm 2: xor <r0andr1=%r8,<maj7=%rdi
4575 xor %r8,%rdi
4576
4577 # qhasm: ch6 ^= r4
4578 # asm 1: xor <r4=int64#9,<ch6=int64#7
4579 # asm 2: xor <r4=%r11,<ch6=%rax
4580 xor %r11,%rax
4581
4582 # qhasm: r3Sigma1 = r3>>>14
4583 # asm 1: rorx $14,<r3=int64#6,>r3Sigma1=int64#12
4584 # asm 2: rorx $14,<r3=%r9,>r3Sigma1=%r14
4585 rorx $14,%r9,%r14
4586
4587 # qhasm: r7 += maj7
4588 # asm 1: add <maj7=int64#1,<r7=int64#11
4589 # asm 2: add <maj7=%rdi,<r7=%r13
4590 add %rdi,%r13
4591
4592 # qhasm: ch6 &= r3
4593 # asm 1: and <r3=int64#6,<ch6=int64#7
4594 # asm 2: and <r3=%r9,<ch6=%rax
4595 and %r9,%rax
4596
4597 # qhasm: r318 = r3>>>18
4598 # asm 1: rorx $18,<r3=int64#6,>r318=int64#1
4599 # asm 2: rorx $18,<r3=%r9,>r318=%rdi
4600 rorx $18,%r9,%rdi
4601
4602 # qhasm: r3Sigma1 ^= r318
4603 # asm 1: xor <r318=int64#1,<r3Sigma1=int64#12
4604 # asm 2: xor <r318=%rdi,<r3Sigma1=%r14
4605 xor %rdi,%r14
4606
4607 # qhasm: maj6 &= r7
4608 # asm 1: and <r7=int64#11,<maj6=int64#13
4609 # asm 2: and <r7=%r13,<maj6=%r15
4610 and %r13,%r15
4611
4612 # qhasm: ch6 ^= r5
4613 # asm 1: xor <r5=int64#8,<ch6=int64#7
4614 # asm 2: xor <r5=%r10,<ch6=%rax
4615 xor %r10,%rax
4616
4617 # qhasm: r341 = r3>>>41
4618 # asm 1: rorx $41,<r3=int64#6,>r341=int64#1
4619 # asm 2: rorx $41,<r3=%r9,>r341=%rdi
4620 rorx $41,%r9,%rdi
4621
4622 # qhasm: r3Sigma1 ^= r341
4623 # asm 1: xor <r341=int64#1,<r3Sigma1=int64#12
4624 # asm 2: xor <r341=%rdi,<r3Sigma1=%r14
4625 xor %rdi,%r14
4626
4627 # qhasm: r7Sigma0 = r7>>>28
4628 # asm 1: rorx $28,<r7=int64#11,>r7Sigma0=int64#1
4629 # asm 2: rorx $28,<r7=%r13,>r7Sigma0=%rdi
4630 rorx $28,%r13,%rdi
4631
4632 # qhasm: maj6 ^= r0andr1
4633 # asm 1: xor <r0andr1=int64#5,<maj6=int64#13
4634 # asm 2: xor <r0andr1=%r8,<maj6=%r15
4635 xor %r8,%r15
4636
4637 # qhasm: r6 += ch6
4638 # asm 1: add <ch6=int64#7,<r6=int64#10
4639 # asm 2: add <ch6=%rax,<r6=%r12
4640 add %rax,%r12
4641
4642 # qhasm: r6 += r3Sigma1
4643 # asm 1: add <r3Sigma1=int64#12,<r6=int64#10
4644 # asm 2: add <r3Sigma1=%r14,<r6=%r12
4645 add %r14,%r12
4646
4647 # qhasm: r734 = r7>>>34
4648 # asm 1: rorx $34,<r7=int64#11,>r734=int64#5
4649 # asm 2: rorx $34,<r7=%r13,>r734=%r8
4650 rorx $34,%r13,%r8
4651
4652 # qhasm: r7Sigma0 ^= r734
4653 # asm 1: xor <r734=int64#5,<r7Sigma0=int64#1
4654 # asm 2: xor <r734=%r8,<r7Sigma0=%rdi
4655 xor %r8,%rdi
4656
4657 # qhasm: r2 += r6
4658 # asm 1: add <r6=int64#10,<r2=int64#3
4659 # asm 2: add <r6=%r12,<r2=%rdx
4660 add %r12,%rdx
4661
4662 # qhasm: r6 += maj6
4663 # asm 1: add <maj6=int64#13,<r6=int64#10
4664 # asm 2: add <maj6=%r15,<r6=%r12
4665 add %r15,%r12
4666
4667 # qhasm: r739 = r7>>>39
4668 # asm 1: rorx $39,<r7=int64#11,>r739=int64#5
4669 # asm 2: rorx $39,<r7=%r13,>r739=%r8
4670 rorx $39,%r13,%r8
4671
4672 # qhasm: r7Sigma0 ^= r739
4673 # asm 1: xor <r739=int64#5,<r7Sigma0=int64#1
4674 # asm 2: xor <r739=%r8,<r7Sigma0=%rdi
4675 xor %r8,%rdi
4676
4677 # qhasm: r6 += r7Sigma0
4678 # asm 1: add <r7Sigma0=int64#1,<r6=int64#10
4679 # asm 2: add <r7Sigma0=%rdi,<r6=%r12
4680 add %rdi,%r12
4681
4682 # qhasm: in = in_stack
4683 # asm 1: movq <in_stack=stack64#11,>in=int64#1
4684 # asm 2: movq <in_stack=400(%rsp),>in=%rdi
4685 movq 400(%rsp),%rdi
4686
4687 # qhasm: r5 += wc891011[2]
4688 # asm 1: addq <wc891011=stack256#3,<r5=int64#8
4689 # asm 2: addq <wc891011=80(%rsp),<r5=%r10
4690 addq 80(%rsp),%r10
4691
4692 # qhasm: r2Sigma1 = r2>>>14
4693 # asm 1: rorx $14,<r2=int64#3,>r2Sigma1=int64#5
4694 # asm 2: rorx $14,<r2=%rdx,>r2Sigma1=%r8
4695 rorx $14,%rdx,%r8
4696
4697 # qhasm: ch5 = r4
4698 # asm 1: mov <r4=int64#9,>ch5=int64#7
4699 # asm 2: mov <r4=%r11,>ch5=%rax
4700 mov %r11,%rax
4701
4702 # qhasm: r218 = r2>>>18
4703 # asm 1: rorx $18,<r2=int64#3,>r218=int64#12
4704 # asm 2: rorx $18,<r2=%rdx,>r218=%r14
4705 rorx $18,%rdx,%r14
4706
4707 # qhasm: ch5 ^= r3
4708 # asm 1: xor <r3=int64#6,<ch5=int64#7
4709 # asm 2: xor <r3=%r9,<ch5=%rax
4710 xor %r9,%rax
4711
4712 # qhasm: r241 = r2>>>41
4713 # asm 1: rorx $41,<r2=int64#3,>r241=int64#13
4714 # asm 2: rorx $41,<r2=%rdx,>r241=%r15
4715 rorx $41,%rdx,%r15
4716
4717 # qhasm: r2Sigma1 ^= r218
4718 # asm 1: xor <r218=int64#12,<r2Sigma1=int64#5
4719 # asm 2: xor <r218=%r14,<r2Sigma1=%r8
4720 xor %r14,%r8
4721
4722 # qhasm: ch5 &= r2
4723 # asm 1: and <r2=int64#3,<ch5=int64#7
4724 # asm 2: and <r2=%rdx,<ch5=%rax
4725 and %rdx,%rax
4726
4727 # qhasm: r6Sigma0 = r6>>>28
4728 # asm 1: rorx $28,<r6=int64#10,>r6Sigma0=int64#12
4729 # asm 2: rorx $28,<r6=%r12,>r6Sigma0=%r14
4730 rorx $28,%r12,%r14
4731
4732 # qhasm: r2Sigma1 ^= r241
4733 # asm 1: xor <r241=int64#13,<r2Sigma1=int64#5
4734 # asm 2: xor <r241=%r15,<r2Sigma1=%r8
4735 xor %r15,%r8
4736
4737 # qhasm: ch5 ^= r4
4738 # asm 1: xor <r4=int64#9,<ch5=int64#7
4739 # asm 2: xor <r4=%r11,<ch5=%rax
4740 xor %r11,%rax
4741
4742 # qhasm: r634 = r6>>>34
4743 # asm 1: rorx $34,<r6=int64#10,>r634=int64#13
4744 # asm 2: rorx $34,<r6=%r12,>r634=%r15
4745 rorx $34,%r12,%r15
4746
4747 # qhasm: maj4 = r7
4748 # asm 1: mov <r7=int64#11,>maj4=int64#14
4749 # asm 2: mov <r7=%r13,>maj4=%rbx
4750 mov %r13,%rbx
4751
4752 # qhasm: maj4 ^= r6
4753 # asm 1: xor <r6=int64#10,<maj4=int64#14
4754 # asm 2: xor <r6=%r12,<maj4=%rbx
4755 xor %r12,%rbx
4756
4757 # qhasm: r6Sigma0 ^= r634
4758 # asm 1: xor <r634=int64#13,<r6Sigma0=int64#12
4759 # asm 2: xor <r634=%r15,<r6Sigma0=%r14
4760 xor %r15,%r14
4761
4762 # qhasm: r5 += ch5
4763 # asm 1: add <ch5=int64#7,<r5=int64#8
4764 # asm 2: add <ch5=%rax,<r5=%r10
4765 add %rax,%r10
4766
4767 # qhasm: r6andr7 = r7
4768 # asm 1: mov <r7=int64#11,>r6andr7=int64#7
4769 # asm 2: mov <r7=%r13,>r6andr7=%rax
4770 mov %r13,%rax
4771
4772 # qhasm: r639 = r6>>>39
4773 # asm 1: rorx $39,<r6=int64#10,>r639=int64#13
4774 # asm 2: rorx $39,<r6=%r12,>r639=%r15
4775 rorx $39,%r12,%r15
4776
4777 # qhasm: r6andr7 &= r6
4778 # asm 1: and <r6=int64#10,<r6andr7=int64#7
4779 # asm 2: and <r6=%r12,<r6andr7=%rax
4780 and %r12,%rax
4781
4782 # qhasm: r6Sigma0 ^= r639
4783 # asm 1: xor <r639=int64#13,<r6Sigma0=int64#12
4784 # asm 2: xor <r639=%r15,<r6Sigma0=%r14
4785 xor %r15,%r14
4786
4787 # qhasm: r5 += r2Sigma1
4788 # asm 1: add <r2Sigma1=int64#5,<r5=int64#8
4789 # asm 2: add <r2Sigma1=%r8,<r5=%r10
4790 add %r8,%r10
4791
4792 # qhasm: maj5 = r0
4793 # asm 1: mov <r0=int64#2,>maj5=int64#5
4794 # asm 2: mov <r0=%rsi,>maj5=%r8
4795 mov %rsi,%r8
4796
4797 # qhasm: r4 += wc891011[3]
4798 # asm 1: addq <wc891011=stack256#3,<r4=int64#9
4799 # asm 2: addq <wc891011=88(%rsp),<r4=%r11
4800 addq 88(%rsp),%r11
4801
4802 # qhasm: maj5 &= maj4
4803 # asm 1: and <maj4=int64#14,<maj5=int64#5
4804 # asm 2: and <maj4=%rbx,<maj5=%r8
4805 and %rbx,%r8
4806
4807 # qhasm: r1 += r5
4808 # asm 1: add <r5=int64#8,<r1=int64#4
4809 # asm 2: add <r5=%r10,<r1=%rcx
4810 add %r10,%rcx
4811
4812 # qhasm: r5 += r6Sigma0
4813 # asm 1: add <r6Sigma0=int64#12,<r5=int64#8
4814 # asm 2: add <r6Sigma0=%r14,<r5=%r10
4815 add %r14,%r10
4816
4817 # qhasm: ch4 = r3
4818 # asm 1: mov <r3=int64#6,>ch4=int64#12
4819 # asm 2: mov <r3=%r9,>ch4=%r14
4820 mov %r9,%r14
4821
4822 # qhasm: maj5 ^= r6andr7
4823 # asm 1: xor <r6andr7=int64#7,<maj5=int64#5
4824 # asm 2: xor <r6andr7=%rax,<maj5=%r8
4825 xor %rax,%r8
4826
4827 # qhasm: ch4 ^= r2
4828 # asm 1: xor <r2=int64#3,<ch4=int64#12
4829 # asm 2: xor <r2=%rdx,<ch4=%r14
4830 xor %rdx,%r14
4831
4832 # qhasm: r1Sigma1 = r1>>>14
4833 # asm 1: rorx $14,<r1=int64#4,>r1Sigma1=int64#13
4834 # asm 2: rorx $14,<r1=%rcx,>r1Sigma1=%r15
4835 rorx $14,%rcx,%r15
4836
4837 # qhasm: r5 += maj5
4838 # asm 1: add <maj5=int64#5,<r5=int64#8
4839 # asm 2: add <maj5=%r8,<r5=%r10
4840 add %r8,%r10
4841
4842 # qhasm: ch4 &= r1
4843 # asm 1: and <r1=int64#4,<ch4=int64#12
4844 # asm 2: and <r1=%rcx,<ch4=%r14
4845 and %rcx,%r14
4846
4847 # qhasm: r118 = r1>>>18
4848 # asm 1: rorx $18,<r1=int64#4,>r118=int64#5
4849 # asm 2: rorx $18,<r1=%rcx,>r118=%r8
4850 rorx $18,%rcx,%r8
4851
4852 # qhasm: r1Sigma1 ^= r118
4853 # asm 1: xor <r118=int64#5,<r1Sigma1=int64#13
4854 # asm 2: xor <r118=%r8,<r1Sigma1=%r15
4855 xor %r8,%r15
4856
4857 # qhasm: maj4 &= r5
4858 # asm 1: and <r5=int64#8,<maj4=int64#14
4859 # asm 2: and <r5=%r10,<maj4=%rbx
4860 and %r10,%rbx
4861
4862 # qhasm: ch4 ^= r3
4863 # asm 1: xor <r3=int64#6,<ch4=int64#12
4864 # asm 2: xor <r3=%r9,<ch4=%r14
4865 xor %r9,%r14
4866
4867 # qhasm: r141 = r1>>>41
4868 # asm 1: rorx $41,<r1=int64#4,>r141=int64#5
4869 # asm 2: rorx $41,<r1=%rcx,>r141=%r8
4870 rorx $41,%rcx,%r8
4871
4872 # qhasm: r1Sigma1 ^= r141
4873 # asm 1: xor <r141=int64#5,<r1Sigma1=int64#13
4874 # asm 2: xor <r141=%r8,<r1Sigma1=%r15
4875 xor %r8,%r15
4876
4877 # qhasm: r5Sigma0 = r5>>>28
4878 # asm 1: rorx $28,<r5=int64#8,>r5Sigma0=int64#5
4879 # asm 2: rorx $28,<r5=%r10,>r5Sigma0=%r8
4880 rorx $28,%r10,%r8
4881
4882 # qhasm: maj4 ^= r6andr7
4883 # asm 1: xor <r6andr7=int64#7,<maj4=int64#14
4884 # asm 2: xor <r6andr7=%rax,<maj4=%rbx
4885 xor %rax,%rbx
4886
4887 # qhasm: r4 += ch4
4888 # asm 1: add <ch4=int64#12,<r4=int64#9
4889 # asm 2: add <ch4=%r14,<r4=%r11
4890 add %r14,%r11
4891
4892 # qhasm: r4 += r1Sigma1
4893 # asm 1: add <r1Sigma1=int64#13,<r4=int64#9
4894 # asm 2: add <r1Sigma1=%r15,<r4=%r11
4895 add %r15,%r11
4896
4897 # qhasm: r534 = r5>>>34
4898 # asm 1: rorx $34,<r5=int64#8,>r534=int64#7
4899 # asm 2: rorx $34,<r5=%r10,>r534=%rax
4900 rorx $34,%r10,%rax
4901
4902 # qhasm: r5Sigma0 ^= r534
4903 # asm 1: xor <r534=int64#7,<r5Sigma0=int64#5
4904 # asm 2: xor <r534=%rax,<r5Sigma0=%r8
4905 xor %rax,%r8
4906
4907 # qhasm: r0 += r4
4908 # asm 1: add <r4=int64#9,<r0=int64#2
4909 # asm 2: add <r4=%r11,<r0=%rsi
4910 add %r11,%rsi
4911
4912 # qhasm: r4 += maj4
4913 # asm 1: add <maj4=int64#14,<r4=int64#9
4914 # asm 2: add <maj4=%rbx,<r4=%r11
4915 add %rbx,%r11
4916
4917 # qhasm: r539 = r5>>>39
4918 # asm 1: rorx $39,<r5=int64#8,>r539=int64#7
4919 # asm 2: rorx $39,<r5=%r10,>r539=%rax
4920 rorx $39,%r10,%rax
4921
4922 # qhasm: r5Sigma0 ^= r539
4923 # asm 1: xor <r539=int64#7,<r5Sigma0=int64#5
4924 # asm 2: xor <r539=%rax,<r5Sigma0=%r8
4925 xor %rax,%r8
4926
4927 # qhasm: r4 += r5Sigma0
4928 # asm 1: add <r5Sigma0=int64#5,<r4=int64#9
4929 # asm 2: add <r5Sigma0=%r8,<r4=%r11
4930 add %r8,%r11
4931
4932 # qhasm: inlen = inlen_stack
4933 # asm 1: movq <inlen_stack=stack64#8,>inlen=int64#7
4934 # asm 2: movq <inlen_stack=376(%rsp),>inlen=%rax
4935 movq 376(%rsp),%rax
4936
4937 # qhasm: in += 128
4938 # asm 1: add $128,<in=int64#1
4939 # asm 2: add $128,<in=%rdi
4940 add $128,%rdi
4941
4942 # qhasm: r3 += wc12131415[0]
4943 # asm 1: addq <wc12131415=stack256#5,<r3=int64#6
4944 # asm 2: addq <wc12131415=128(%rsp),<r3=%r9
4945 addq 128(%rsp),%r9
4946
4947 # qhasm: r0Sigma1 = r0>>>14
4948 # asm 1: rorx $14,<r0=int64#2,>r0Sigma1=int64#5
4949 # asm 2: rorx $14,<r0=%rsi,>r0Sigma1=%r8
4950 rorx $14,%rsi,%r8
4951
4952 # qhasm: ch3 = r2
4953 # asm 1: mov <r2=int64#3,>ch3=int64#12
4954 # asm 2: mov <r2=%rdx,>ch3=%r14
4955 mov %rdx,%r14
4956
4957 # qhasm: r018 = r0>>>18
4958 # asm 1: rorx $18,<r0=int64#2,>r018=int64#13
4959 # asm 2: rorx $18,<r0=%rsi,>r018=%r15
4960 rorx $18,%rsi,%r15
4961
4962 # qhasm: ch3 ^= r1
4963 # asm 1: xor <r1=int64#4,<ch3=int64#12
4964 # asm 2: xor <r1=%rcx,<ch3=%r14
4965 xor %rcx,%r14
4966
4967 # qhasm: r041 = r0>>>41
4968 # asm 1: rorx $41,<r0=int64#2,>r041=int64#14
4969 # asm 2: rorx $41,<r0=%rsi,>r041=%rbx
4970 rorx $41,%rsi,%rbx
4971
4972 # qhasm: r0Sigma1 ^= r018
4973 # asm 1: xor <r018=int64#13,<r0Sigma1=int64#5
4974 # asm 2: xor <r018=%r15,<r0Sigma1=%r8
4975 xor %r15,%r8
4976
4977 # qhasm: ch3 &= r0
4978 # asm 1: and <r0=int64#2,<ch3=int64#12
4979 # asm 2: and <r0=%rsi,<ch3=%r14
4980 and %rsi,%r14
4981
4982 # qhasm: r4Sigma0 = r4>>>28
4983 # asm 1: rorx $28,<r4=int64#9,>r4Sigma0=int64#13
4984 # asm 2: rorx $28,<r4=%r11,>r4Sigma0=%r15
4985 rorx $28,%r11,%r15
4986
4987 # qhasm: r0Sigma1 ^= r041
4988 # asm 1: xor <r041=int64#14,<r0Sigma1=int64#5
4989 # asm 2: xor <r041=%rbx,<r0Sigma1=%r8
4990 xor %rbx,%r8
4991
4992 # qhasm: ch3 ^= r2
4993 # asm 1: xor <r2=int64#3,<ch3=int64#12
4994 # asm 2: xor <r2=%rdx,<ch3=%r14
4995 xor %rdx,%r14
4996
4997 # qhasm: r434 = r4>>>34
4998 # asm 1: rorx $34,<r4=int64#9,>r434=int64#14
4999 # asm 2: rorx $34,<r4=%r11,>r434=%rbx
5000 rorx $34,%r11,%rbx
5001
5002 # qhasm: maj2 = r5
5003 # asm 1: mov <r5=int64#8,>maj2=int64#15
5004 # asm 2: mov <r5=%r10,>maj2=%rbp
5005 mov %r10,%rbp
5006
5007 # qhasm: maj2 ^= r4
5008 # asm 1: xor <r4=int64#9,<maj2=int64#15
5009 # asm 2: xor <r4=%r11,<maj2=%rbp
5010 xor %r11,%rbp
5011
5012 # qhasm: r4Sigma0 ^= r434
5013 # asm 1: xor <r434=int64#14,<r4Sigma0=int64#13
5014 # asm 2: xor <r434=%rbx,<r4Sigma0=%r15
5015 xor %rbx,%r15
5016
5017 # qhasm: r3 += ch3
5018 # asm 1: add <ch3=int64#12,<r3=int64#6
5019 # asm 2: add <ch3=%r14,<r3=%r9
5020 add %r14,%r9
5021
5022 # qhasm: r4andr5 = r5
5023 # asm 1: mov <r5=int64#8,>r4andr5=int64#12
5024 # asm 2: mov <r5=%r10,>r4andr5=%r14
5025 mov %r10,%r14
5026
5027 # qhasm: r439 = r4>>>39
5028 # asm 1: rorx $39,<r4=int64#9,>r439=int64#14
5029 # asm 2: rorx $39,<r4=%r11,>r439=%rbx
5030 rorx $39,%r11,%rbx
5031
5032 # qhasm: r4andr5 &= r4
5033 # asm 1: and <r4=int64#9,<r4andr5=int64#12
5034 # asm 2: and <r4=%r11,<r4andr5=%r14
5035 and %r11,%r14
5036
5037 # qhasm: r4Sigma0 ^= r439
5038 # asm 1: xor <r439=int64#14,<r4Sigma0=int64#13
5039 # asm 2: xor <r439=%rbx,<r4Sigma0=%r15
5040 xor %rbx,%r15
5041
5042 # qhasm: r3 += r0Sigma1
5043 # asm 1: add <r0Sigma1=int64#5,<r3=int64#6
5044 # asm 2: add <r0Sigma1=%r8,<r3=%r9
5045 add %r8,%r9
5046
5047 # qhasm: maj3 = r6
5048 # asm 1: mov <r6=int64#10,>maj3=int64#5
5049 # asm 2: mov <r6=%r12,>maj3=%r8
5050 mov %r12,%r8
5051
5052 # qhasm: r2 += wc12131415[1]
5053 # asm 1: addq <wc12131415=stack256#5,<r2=int64#3
5054 # asm 2: addq <wc12131415=136(%rsp),<r2=%rdx
5055 addq 136(%rsp),%rdx
5056
5057 # qhasm: maj3 &= maj2
5058 # asm 1: and <maj2=int64#15,<maj3=int64#5
5059 # asm 2: and <maj2=%rbp,<maj3=%r8
5060 and %rbp,%r8
5061
5062 # qhasm: r7 += r3
5063 # asm 1: add <r3=int64#6,<r7=int64#11
5064 # asm 2: add <r3=%r9,<r7=%r13
5065 add %r9,%r13
5066
5067 # qhasm: r3 += r4Sigma0
5068 # asm 1: add <r4Sigma0=int64#13,<r3=int64#6
5069 # asm 2: add <r4Sigma0=%r15,<r3=%r9
5070 add %r15,%r9
5071
5072 # qhasm: ch2 = r1
5073 # asm 1: mov <r1=int64#4,>ch2=int64#13
5074 # asm 2: mov <r1=%rcx,>ch2=%r15
5075 mov %rcx,%r15
5076
5077 # qhasm: maj3 ^= r4andr5
5078 # asm 1: xor <r4andr5=int64#12,<maj3=int64#5
5079 # asm 2: xor <r4andr5=%r14,<maj3=%r8
5080 xor %r14,%r8
5081
5082 # qhasm: ch2 ^= r0
5083 # asm 1: xor <r0=int64#2,<ch2=int64#13
5084 # asm 2: xor <r0=%rsi,<ch2=%r15
5085 xor %rsi,%r15
5086
5087 # qhasm: r7Sigma1 = r7>>>14
5088 # asm 1: rorx $14,<r7=int64#11,>r7Sigma1=int64#14
5089 # asm 2: rorx $14,<r7=%r13,>r7Sigma1=%rbx
5090 rorx $14,%r13,%rbx
5091
5092 # qhasm: r3 += maj3
5093 # asm 1: add <maj3=int64#5,<r3=int64#6
5094 # asm 2: add <maj3=%r8,<r3=%r9
5095 add %r8,%r9
5096
5097 # qhasm: ch2 &= r7
5098 # asm 1: and <r7=int64#11,<ch2=int64#13
5099 # asm 2: and <r7=%r13,<ch2=%r15
5100 and %r13,%r15
5101
5102 # qhasm: r718 = r7>>>18
5103 # asm 1: rorx $18,<r7=int64#11,>r718=int64#5
5104 # asm 2: rorx $18,<r7=%r13,>r718=%r8
5105 rorx $18,%r13,%r8
5106
5107 # qhasm: r7Sigma1 ^= r718
5108 # asm 1: xor <r718=int64#5,<r7Sigma1=int64#14
5109 # asm 2: xor <r718=%r8,<r7Sigma1=%rbx
5110 xor %r8,%rbx
5111
5112 # qhasm: maj2 &= r3
5113 # asm 1: and <r3=int64#6,<maj2=int64#15
5114 # asm 2: and <r3=%r9,<maj2=%rbp
5115 and %r9,%rbp
5116
5117 # qhasm: ch2 ^= r1
5118 # asm 1: xor <r1=int64#4,<ch2=int64#13
5119 # asm 2: xor <r1=%rcx,<ch2=%r15
5120 xor %rcx,%r15
5121
5122 # qhasm: r741 = r7>>>41
5123 # asm 1: rorx $41,<r7=int64#11,>r741=int64#5
5124 # asm 2: rorx $41,<r7=%r13,>r741=%r8
5125 rorx $41,%r13,%r8
5126
5127 # qhasm: r7Sigma1 ^= r741
5128 # asm 1: xor <r741=int64#5,<r7Sigma1=int64#14
5129 # asm 2: xor <r741=%r8,<r7Sigma1=%rbx
5130 xor %r8,%rbx
5131
5132 # qhasm: r3Sigma0 = r3>>>28
5133 # asm 1: rorx $28,<r3=int64#6,>r3Sigma0=int64#5
5134 # asm 2: rorx $28,<r3=%r9,>r3Sigma0=%r8
5135 rorx $28,%r9,%r8
5136
5137 # qhasm: maj2 ^= r4andr5
5138 # asm 1: xor <r4andr5=int64#12,<maj2=int64#15
5139 # asm 2: xor <r4andr5=%r14,<maj2=%rbp
5140 xor %r14,%rbp
5141
5142 # qhasm: r2 += ch2
5143 # asm 1: add <ch2=int64#13,<r2=int64#3
5144 # asm 2: add <ch2=%r15,<r2=%rdx
5145 add %r15,%rdx
5146
5147 # qhasm: r2 += r7Sigma1
5148 # asm 1: add <r7Sigma1=int64#14,<r2=int64#3
5149 # asm 2: add <r7Sigma1=%rbx,<r2=%rdx
5150 add %rbx,%rdx
5151
5152 # qhasm: r334 = r3>>>34
5153 # asm 1: rorx $34,<r3=int64#6,>r334=int64#12
5154 # asm 2: rorx $34,<r3=%r9,>r334=%r14
5155 rorx $34,%r9,%r14
5156
5157 # qhasm: r3Sigma0 ^= r334
5158 # asm 1: xor <r334=int64#12,<r3Sigma0=int64#5
5159 # asm 2: xor <r334=%r14,<r3Sigma0=%r8
5160 xor %r14,%r8
5161
5162 # qhasm: r6 += r2
5163 # asm 1: add <r2=int64#3,<r6=int64#10
5164 # asm 2: add <r2=%rdx,<r6=%r12
5165 add %rdx,%r12
5166
5167 # qhasm: r2 += maj2
5168 # asm 1: add <maj2=int64#15,<r2=int64#3
5169 # asm 2: add <maj2=%rbp,<r2=%rdx
5170 add %rbp,%rdx
5171
5172 # qhasm: r339 = r3>>>39
5173 # asm 1: rorx $39,<r3=int64#6,>r339=int64#12
5174 # asm 2: rorx $39,<r3=%r9,>r339=%r14
5175 rorx $39,%r9,%r14
5176
5177 # qhasm: r3Sigma0 ^= r339
5178 # asm 1: xor <r339=int64#12,<r3Sigma0=int64#5
5179 # asm 2: xor <r339=%r14,<r3Sigma0=%r8
5180 xor %r14,%r8
5181
5182 # qhasm: r2 += r3Sigma0
5183 # asm 1: add <r3Sigma0=int64#5,<r2=int64#3
5184 # asm 2: add <r3Sigma0=%r8,<r2=%rdx
5185 add %r8,%rdx
5186
5187 # qhasm: inlen -= 128
5188 # asm 1: sub $128,<inlen=int64#7
5189 # asm 2: sub $128,<inlen=%rax
5190 sub $128,%rax
5191
5192 # qhasm: r1 += wc12131415[2]
5193 # asm 1: addq <wc12131415=stack256#5,<r1=int64#4
5194 # asm 2: addq <wc12131415=144(%rsp),<r1=%rcx
5195 addq 144(%rsp),%rcx
5196
5197 # qhasm: r6Sigma1 = r6>>>14
5198 # asm 1: rorx $14,<r6=int64#10,>r6Sigma1=int64#5
5199 # asm 2: rorx $14,<r6=%r12,>r6Sigma1=%r8
5200 rorx $14,%r12,%r8
5201
5202 # qhasm: ch1 = r0
5203 # asm 1: mov <r0=int64#2,>ch1=int64#12
5204 # asm 2: mov <r0=%rsi,>ch1=%r14
5205 mov %rsi,%r14
5206
5207 # qhasm: r618 = r6>>>18
5208 # asm 1: rorx $18,<r6=int64#10,>r618=int64#13
5209 # asm 2: rorx $18,<r6=%r12,>r618=%r15
5210 rorx $18,%r12,%r15
5211
5212 # qhasm: ch1 ^= r7
5213 # asm 1: xor <r7=int64#11,<ch1=int64#12
5214 # asm 2: xor <r7=%r13,<ch1=%r14
5215 xor %r13,%r14
5216
5217 # qhasm: r641 = r6>>>41
5218 # asm 1: rorx $41,<r6=int64#10,>r641=int64#14
5219 # asm 2: rorx $41,<r6=%r12,>r641=%rbx
5220 rorx $41,%r12,%rbx
5221
5222 # qhasm: r6Sigma1 ^= r618
5223 # asm 1: xor <r618=int64#13,<r6Sigma1=int64#5
5224 # asm 2: xor <r618=%r15,<r6Sigma1=%r8
5225 xor %r15,%r8
5226
5227 # qhasm: ch1 &= r6
5228 # asm 1: and <r6=int64#10,<ch1=int64#12
5229 # asm 2: and <r6=%r12,<ch1=%r14
5230 and %r12,%r14
5231
5232 # qhasm: r2Sigma0 = r2>>>28
5233 # asm 1: rorx $28,<r2=int64#3,>r2Sigma0=int64#13
5234 # asm 2: rorx $28,<r2=%rdx,>r2Sigma0=%r15
5235 rorx $28,%rdx,%r15
5236
5237 # qhasm: r6Sigma1 ^= r641
5238 # asm 1: xor <r641=int64#14,<r6Sigma1=int64#5
5239 # asm 2: xor <r641=%rbx,<r6Sigma1=%r8
5240 xor %rbx,%r8
5241
5242 # qhasm: ch1 ^= r0
5243 # asm 1: xor <r0=int64#2,<ch1=int64#12
5244 # asm 2: xor <r0=%rsi,<ch1=%r14
5245 xor %rsi,%r14
5246
5247 # qhasm: r234 = r2>>>34
5248 # asm 1: rorx $34,<r2=int64#3,>r234=int64#14
5249 # asm 2: rorx $34,<r2=%rdx,>r234=%rbx
5250 rorx $34,%rdx,%rbx
5251
5252 # qhasm: maj0 = r3
5253 # asm 1: mov <r3=int64#6,>maj0=int64#15
5254 # asm 2: mov <r3=%r9,>maj0=%rbp
5255 mov %r9,%rbp
5256
5257 # qhasm: maj0 ^= r2
5258 # asm 1: xor <r2=int64#3,<maj0=int64#15
5259 # asm 2: xor <r2=%rdx,<maj0=%rbp
5260 xor %rdx,%rbp
5261
5262 # qhasm: r2Sigma0 ^= r234
5263 # asm 1: xor <r234=int64#14,<r2Sigma0=int64#13
5264 # asm 2: xor <r234=%rbx,<r2Sigma0=%r15
5265 xor %rbx,%r15
5266
5267 # qhasm: r1 += ch1
5268 # asm 1: add <ch1=int64#12,<r1=int64#4
5269 # asm 2: add <ch1=%r14,<r1=%rcx
5270 add %r14,%rcx
5271
5272 # qhasm: r2andr3 = r3
5273 # asm 1: mov <r3=int64#6,>r2andr3=int64#12
5274 # asm 2: mov <r3=%r9,>r2andr3=%r14
5275 mov %r9,%r14
5276
5277 # qhasm: r239 = r2>>>39
5278 # asm 1: rorx $39,<r2=int64#3,>r239=int64#14
5279 # asm 2: rorx $39,<r2=%rdx,>r239=%rbx
5280 rorx $39,%rdx,%rbx
5281
5282 # qhasm: r2andr3 &= r2
5283 # asm 1: and <r2=int64#3,<r2andr3=int64#12
5284 # asm 2: and <r2=%rdx,<r2andr3=%r14
5285 and %rdx,%r14
5286
5287 # qhasm: r2Sigma0 ^= r239
5288 # asm 1: xor <r239=int64#14,<r2Sigma0=int64#13
5289 # asm 2: xor <r239=%rbx,<r2Sigma0=%r15
5290 xor %rbx,%r15
5291
5292 # qhasm: r1 += r6Sigma1
5293 # asm 1: add <r6Sigma1=int64#5,<r1=int64#4
5294 # asm 2: add <r6Sigma1=%r8,<r1=%rcx
5295 add %r8,%rcx
5296
5297 # qhasm: maj1 = r4
5298 # asm 1: mov <r4=int64#9,>maj1=int64#5
5299 # asm 2: mov <r4=%r11,>maj1=%r8
5300 mov %r11,%r8
5301
5302 # qhasm: r0 += wc12131415[3]
5303 # asm 1: addq <wc12131415=stack256#5,<r0=int64#2
5304 # asm 2: addq <wc12131415=152(%rsp),<r0=%rsi
5305 addq 152(%rsp),%rsi
5306
5307 # qhasm: maj1 &= maj0
5308 # asm 1: and <maj0=int64#15,<maj1=int64#5
5309 # asm 2: and <maj0=%rbp,<maj1=%r8
5310 and %rbp,%r8
5311
5312 # qhasm: r5 += r1
5313 # asm 1: add <r1=int64#4,<r5=int64#8
5314 # asm 2: add <r1=%rcx,<r5=%r10
5315 add %rcx,%r10
5316
5317 # qhasm: r1 += r2Sigma0
5318 # asm 1: add <r2Sigma0=int64#13,<r1=int64#4
5319 # asm 2: add <r2Sigma0=%r15,<r1=%rcx
5320 add %r15,%rcx
5321
5322 # qhasm: ch0 = r7
5323 # asm 1: mov <r7=int64#11,>ch0=int64#13
5324 # asm 2: mov <r7=%r13,>ch0=%r15
5325 mov %r13,%r15
5326
5327 # qhasm: maj1 ^= r2andr3
5328 # asm 1: xor <r2andr3=int64#12,<maj1=int64#5
5329 # asm 2: xor <r2andr3=%r14,<maj1=%r8
5330 xor %r14,%r8
5331
5332 # qhasm: ch0 ^= r6
5333 # asm 1: xor <r6=int64#10,<ch0=int64#13
5334 # asm 2: xor <r6=%r12,<ch0=%r15
5335 xor %r12,%r15
5336
5337 # qhasm: r5Sigma1 = r5>>>14
5338 # asm 1: rorx $14,<r5=int64#8,>r5Sigma1=int64#14
5339 # asm 2: rorx $14,<r5=%r10,>r5Sigma1=%rbx
5340 rorx $14,%r10,%rbx
5341
5342 # qhasm: r1 += maj1
5343 # asm 1: add <maj1=int64#5,<r1=int64#4
5344 # asm 2: add <maj1=%r8,<r1=%rcx
5345 add %r8,%rcx
5346
5347 # qhasm: ch0 &= r5
5348 # asm 1: and <r5=int64#8,<ch0=int64#13
5349 # asm 2: and <r5=%r10,<ch0=%r15
5350 and %r10,%r15
5351
5352 # qhasm: r518 = r5>>>18
5353 # asm 1: rorx $18,<r5=int64#8,>r518=int64#5
5354 # asm 2: rorx $18,<r5=%r10,>r518=%r8
5355 rorx $18,%r10,%r8
5356
5357 # qhasm: r5Sigma1 ^= r518
5358 # asm 1: xor <r518=int64#5,<r5Sigma1=int64#14
5359 # asm 2: xor <r518=%r8,<r5Sigma1=%rbx
5360 xor %r8,%rbx
5361
5362 # qhasm: maj0 &= r1
5363 # asm 1: and <r1=int64#4,<maj0=int64#15
5364 # asm 2: and <r1=%rcx,<maj0=%rbp
5365 and %rcx,%rbp
5366
5367 # qhasm: ch0 ^= r7
5368 # asm 1: xor <r7=int64#11,<ch0=int64#13
5369 # asm 2: xor <r7=%r13,<ch0=%r15
5370 xor %r13,%r15
5371
5372 # qhasm: r541 = r5>>>41
5373 # asm 1: rorx $41,<r5=int64#8,>r541=int64#5
5374 # asm 2: rorx $41,<r5=%r10,>r541=%r8
5375 rorx $41,%r10,%r8
5376
5377 # qhasm: r5Sigma1 ^= r541
5378 # asm 1: xor <r541=int64#5,<r5Sigma1=int64#14
5379 # asm 2: xor <r541=%r8,<r5Sigma1=%rbx
5380 xor %r8,%rbx
5381
5382 # qhasm: r1Sigma0 = r1>>>28
5383 # asm 1: rorx $28,<r1=int64#4,>r1Sigma0=int64#5
5384 # asm 2: rorx $28,<r1=%rcx,>r1Sigma0=%r8
5385 rorx $28,%rcx,%r8
5386
5387 # qhasm: maj0 ^= r2andr3
5388 # asm 1: xor <r2andr3=int64#12,<maj0=int64#15
5389 # asm 2: xor <r2andr3=%r14,<maj0=%rbp
5390 xor %r14,%rbp
5391
5392 # qhasm: r0 += ch0
5393 # asm 1: add <ch0=int64#13,<r0=int64#2
5394 # asm 2: add <ch0=%r15,<r0=%rsi
5395 add %r15,%rsi
5396
5397 # qhasm: r0 += r5Sigma1
5398 # asm 1: add <r5Sigma1=int64#14,<r0=int64#2
5399 # asm 2: add <r5Sigma1=%rbx,<r0=%rsi
5400 add %rbx,%rsi
5401
5402 # qhasm: r134 = r1>>>34
5403 # asm 1: rorx $34,<r1=int64#4,>r134=int64#12
5404 # asm 2: rorx $34,<r1=%rcx,>r134=%r14
5405 rorx $34,%rcx,%r14
5406
5407 # qhasm: r1Sigma0 ^= r134
5408 # asm 1: xor <r134=int64#12,<r1Sigma0=int64#5
5409 # asm 2: xor <r134=%r14,<r1Sigma0=%r8
5410 xor %r14,%r8
5411
5412 # qhasm: r4 += r0
5413 # asm 1: add <r0=int64#2,<r4=int64#9
5414 # asm 2: add <r0=%rsi,<r4=%r11
5415 add %rsi,%r11
5416
5417 # qhasm: r0 += maj0
5418 # asm 1: add <maj0=int64#15,<r0=int64#2
5419 # asm 2: add <maj0=%rbp,<r0=%rsi
5420 add %rbp,%rsi
5421
5422 # qhasm: r139 = r1>>>39
5423 # asm 1: rorx $39,<r1=int64#4,>r139=int64#12
5424 # asm 2: rorx $39,<r1=%rcx,>r139=%r14
5425 rorx $39,%rcx,%r14
5426
5427 # qhasm: r1Sigma0 ^= r139
5428 # asm 1: xor <r139=int64#12,<r1Sigma0=int64#5
5429 # asm 2: xor <r139=%r14,<r1Sigma0=%r8
5430 xor %r14,%r8
5431
5432 # qhasm: r0 += r1Sigma0
5433 # asm 1: add <r1Sigma0=int64#5,<r0=int64#2
5434 # asm 2: add <r1Sigma0=%r8,<r0=%rsi
5435 add %r8,%rsi
5436
5437 # qhasm: inlen_stack = inlen
5438 # asm 1: movq <inlen=int64#7,>inlen_stack=stack64#8
5439 # asm 2: movq <inlen=%rax,>inlen_stack=376(%rsp)
5440 movq %rax,376(%rsp)
5441
5442 # qhasm: r7 += state4567[3]
5443 # asm 1: addq <state4567=stack256#2,<r7=int64#11
5444 # asm 2: addq <state4567=56(%rsp),<r7=%r13
5445 addq 56(%rsp),%r13
5446
5447 # qhasm: r3 += state0123[3]
5448 # asm 1: addq <state0123=stack256#1,<r3=int64#6
5449 # asm 2: addq <state0123=24(%rsp),<r3=%r9
5450 addq 24(%rsp),%r9
5451
5452 # qhasm: r6 += state4567[2]
5453 # asm 1: addq <state4567=stack256#2,<r6=int64#10
5454 # asm 2: addq <state4567=48(%rsp),<r6=%r12
5455 addq 48(%rsp),%r12
5456
5457 # qhasm: r2 += state0123[2]
5458 # asm 1: addq <state0123=stack256#1,<r2=int64#3
5459 # asm 2: addq <state0123=16(%rsp),<r2=%rdx
5460 addq 16(%rsp),%rdx
5461
5462 # qhasm: r5 += state4567[1]
5463 # asm 1: addq <state4567=stack256#2,<r5=int64#8
5464 # asm 2: addq <state4567=40(%rsp),<r5=%r10
5465 addq 40(%rsp),%r10
5466
5467 # qhasm: r1 += state0123[1]
5468 # asm 1: addq <state0123=stack256#1,<r1=int64#4
5469 # asm 2: addq <state0123=8(%rsp),<r1=%rcx
5470 addq 8(%rsp),%rcx
5471
5472 # qhasm: r4 += state4567[0]
5473 # asm 1: addq <state4567=stack256#2,<r4=int64#9
5474 # asm 2: addq <state4567=32(%rsp),<r4=%r11
5475 addq 32(%rsp),%r11
5476
5477 # qhasm: constants = constants_stack
5478 # asm 1: movq <constants_stack=stack64#10,>constants=int64#5
5479 # asm 2: movq <constants_stack=392(%rsp),>constants=%r8
5480 movq 392(%rsp),%r8
5481
5482 # qhasm: r0 += state0123[0]
5483 # asm 1: addq <state0123=stack256#1,<r0=int64#2
5484 # asm 2: addq <state0123=0(%rsp),<r0=%rsi
5485 addq 0(%rsp),%rsi
5486
5487 # qhasm: unsigned<? inlen - 128
5488 # asm 1: cmp $128,<inlen=int64#7
5489 # asm 2: cmp $128,<inlen=%rax
5490 cmp $128,%rax
5491 # comment:fp stack unchanged by jump
5492
5493 # qhasm: goto outerloop if !unsigned<
5494 jae ._outerloop
5495
5496 # qhasm: inplace state4567[2] = r6
5497 # asm 1: movq <r6=int64#10,<state4567=stack256#2
5498 # asm 2: movq <r6=%r12,<state4567=48(%rsp)
5499 movq %r12,48(%rsp)
5500
5501 # qhasm: inplace state4567[3] = r7
5502 # asm 1: movq <r7=int64#11,<state4567=stack256#2
5503 # asm 2: movq <r7=%r13,<state4567=56(%rsp)
5504 movq %r13,56(%rsp)
5505
5506 # qhasm: inplace state0123[2] = r2
5507 # asm 1: movq <r2=int64#3,<state0123=stack256#1
5508 # asm 2: movq <r2=%rdx,<state0123=16(%rsp)
5509 movq %rdx,16(%rsp)
5510
5511 # qhasm: inplace state4567[0] = r4
5512 # asm 1: movq <r4=int64#9,<state4567=stack256#2
5513 # asm 2: movq <r4=%r11,<state4567=32(%rsp)
5514 movq %r11,32(%rsp)
5515
5516 # qhasm: inplace state0123[3] = r3
5517 # asm 1: movq <r3=int64#6,<state0123=stack256#1
5518 # asm 2: movq <r3=%r9,<state0123=24(%rsp)
5519 movq %r9,24(%rsp)
5520
5521 # qhasm: inplace state4567[1] = r5
5522 # asm 1: movq <r5=int64#8,<state4567=stack256#2
5523 # asm 2: movq <r5=%r10,<state4567=40(%rsp)
5524 movq %r10,40(%rsp)
5525
5526 # qhasm: inplace state0123[0] = r0
5527 # asm 1: movq <r0=int64#2,<state0123=stack256#1
5528 # asm 2: movq <r0=%rsi,<state0123=0(%rsp)
5529 movq %rsi,0(%rsp)
5530
5531 # qhasm: inplace state0123[1] = r1
5532 # asm 1: movq <r1=int64#4,<state0123=stack256#1
5533 # asm 2: movq <r1=%rcx,<state0123=8(%rsp)
5534 movq %rcx,8(%rsp)
5535
5536 # qhasm: statebytes = statebytes_stack
5537 # asm 1: movq <statebytes_stack=stack64#9,>statebytes=int64#1
5538 # asm 2: movq <statebytes_stack=384(%rsp),>statebytes=%rdi
5539 movq 384(%rsp),%rdi
5540
5541 # qhasm: X0 = state0123
5542 # asm 1: vmovapd <state0123=stack256#1,>X0=reg256#2
5543 # asm 2: vmovapd <state0123=0(%rsp),>X0=%ymm1
5544 vmovapd 0(%rsp),%ymm1
5545
5546 # qhasm: X4 = state4567
5547 # asm 1: vmovapd <state4567=stack256#2,>X4=reg256#3
5548 # asm 2: vmovapd <state4567=32(%rsp),>X4=%ymm2
5549 vmovapd 32(%rsp),%ymm2
5550
5551 # qhasm: 2x 16x X0 = X0[bigendian64]
5552 # asm 1: vpshufb <bigendian64=reg256#1,<X0=reg256#2,>X0=reg256#2
5553 # asm 2: vpshufb <bigendian64=%ymm0,<X0=%ymm1,>X0=%ymm1
5554 vpshufb %ymm0,%ymm1,%ymm1
5555
5556 # qhasm: 2x 16x X4 = X4[bigendian64]
5557 # asm 1: vpshufb <bigendian64=reg256#1,<X4=reg256#3,>X4=reg256#1
5558 # asm 2: vpshufb <bigendian64=%ymm0,<X4=%ymm2,>X4=%ymm0
5559 vpshufb %ymm0,%ymm2,%ymm0
5560
5561 # qhasm: mem256[statebytes+0] = X0
5562 # asm 1: vmovupd <X0=reg256#2,0(<statebytes=int64#1)
5563 # asm 2: vmovupd <X0=%ymm1,0(<statebytes=%rdi)
5564 vmovupd %ymm1,0(%rdi)
5565
5566 # qhasm: mem256[statebytes+32] = X4
5567 # asm 1: vmovupd <X4=reg256#1,32(<statebytes=int64#1)
5568 # asm 2: vmovupd <X4=%ymm0,32(<statebytes=%rdi)
5569 vmovupd %ymm0,32(%rdi)
5570
5571 # qhasm: vzeroupper
5572 vzeroupper
5573
5574 # qhasm: caller_r11 = r11_stack
5575 # asm 1: movq <r11_stack=stack64#1,>caller_r11=int64#9
5576 # asm 2: movq <r11_stack=320(%rsp),>caller_r11=%r11
5577 movq 320(%rsp),%r11
5578
5579 # qhasm: caller_r12 = r12_stack
5580 # asm 1: movq <r12_stack=stack64#3,>caller_r12=int64#10
5581 # asm 2: movq <r12_stack=336(%rsp),>caller_r12=%r12
5582 movq 336(%rsp),%r12
5583
5584 # qhasm: caller_r14 = r14_stack
5585 # asm 1: movq <r14_stack=stack64#4,>caller_r14=int64#12
5586 # asm 2: movq <r14_stack=344(%rsp),>caller_r14=%r14
5587 movq 344(%rsp),%r14
5588
5589 # qhasm: caller_r13 = r13_stack
5590 # asm 1: movq <r13_stack=stack64#2,>caller_r13=int64#11
5591 # asm 2: movq <r13_stack=328(%rsp),>caller_r13=%r13
5592 movq 328(%rsp),%r13
5593
5594 # qhasm: caller_r15 = r15_stack
5595 # asm 1: movq <r15_stack=stack64#6,>caller_r15=int64#13
5596 # asm 2: movq <r15_stack=360(%rsp),>caller_r15=%r15
5597 movq 360(%rsp),%r15
5598
5599 # qhasm: caller_rbx = rbx_stack
5600 # asm 1: movq <rbx_stack=stack64#5,>caller_rbx=int64#14
5601 # asm 2: movq <rbx_stack=352(%rsp),>caller_rbx=%rbx
5602 movq 352(%rsp),%rbx
5603
5604 # qhasm: caller_rbp = rbp_stack
5605 # asm 1: movq <rbp_stack=stack64#7,>caller_rbp=int64#15
5606 # asm 2: movq <rbp_stack=368(%rsp),>caller_rbp=%rbp
5607 movq 368(%rsp),%rbp
5608
5609 # qhasm: return inlen
5610 add %r11,%rsp
5611 ret
5612 .section .note.GNU-stack,"",@progbits