Debian Codegraph @

crypto_kem/348864/avx/vec_mul_asm.S

raw
1 // 20240504 djb: add note.GNU-stack
2 // 20221231 djb: port hidden to macos; tnx thom wiggers
3 // 20221230 djb: add linker line
4
5 // linker define vec_mul_asm
6
7 #include "crypto_asm_hidden.h"
8
9 #define vec_mul_asm CRYPTO_SHARED_NAMESPACE(vec_mul_asm)
10 #define _vec_mul_asm _CRYPTO_SHARED_NAMESPACE(vec_mul_asm)
11
12 # qhasm: int64 input_0
13
14 # qhasm: int64 input_1
15
16 # qhasm: int64 input_2
17
18 # qhasm: int64 input_3
19
20 # qhasm: int64 input_4
21
22 # qhasm: int64 input_5
23
24 # qhasm: stack64 input_6
25
26 # qhasm: stack64 input_7
27
28 # qhasm: int64 caller_r11
29
30 # qhasm: int64 caller_r12
31
32 # qhasm: int64 caller_r13
33
34 # qhasm: int64 caller_r14
35
36 # qhasm: int64 caller_r15
37
38 # qhasm: int64 caller_rbx
39
40 # qhasm: int64 caller_rbp
41
42 # qhasm: reg256 s0
43
44 # qhasm: reg256 s1
45
46 # qhasm: reg256 s2
47
48 # qhasm: reg256 s3
49
50 # qhasm: reg256 s4
51
52 # qhasm: reg256 s5
53
54 # qhasm: reg256 t0
55
56 # qhasm: reg256 t1
57
58 # qhasm: reg256 t2
59
60 # qhasm: reg256 b0
61
62 # qhasm: reg256 b1
63
64 # qhasm: reg256 b2
65
66 # qhasm: reg256 b3
67
68 # qhasm: reg256 b4
69
70 # qhasm: reg256 b5
71
72 # qhasm: reg256 a0
73
74 # qhasm: reg256 a1
75
76 # qhasm: reg256 a2
77
78 # qhasm: reg256 a3
79
80 # qhasm: reg256 a4
81
82 # qhasm: reg256 a5
83
84 # qhasm: reg256 r0
85
86 # qhasm: reg256 r1
87
88 # qhasm: reg256 r2
89
90 # qhasm: reg256 r3
91
92 # qhasm: reg256 r4
93
94 # qhasm: reg256 r5
95
96 # qhasm: reg256 r6
97
98 # qhasm: reg256 r7
99
100 # qhasm: reg256 r8
101
102 # qhasm: reg256 r9
103
104 # qhasm: reg256 r10
105
106 # qhasm: reg256 r11
107
108 # qhasm: reg256 r12
109
110 # qhasm: reg256 r13
111
112 # qhasm: reg256 r14
113
114 # qhasm: reg256 r15
115
116 # qhasm: reg256 r16
117
118 # qhasm: reg256 r17
119
120 # qhasm: reg256 r18
121
122 # qhasm: reg256 r19
123
124 # qhasm: reg256 r20
125
126 # qhasm: reg256 r21
127
128 # qhasm: reg256 r22
129
130 # qhasm: reg256 r
131
132 # qhasm: int64 h0
133
134 # qhasm: int64 h1
135
136 # qhasm: int64 h2
137
138 # qhasm: int64 h3
139
140 # qhasm: int64 h4
141
142 # qhasm: int64 h5
143
144 # qhasm: int64 h6
145
146 # qhasm: int64 h7
147
148 # qhasm: int64 h8
149
150 # qhasm: int64 h9
151
152 # qhasm: int64 h10
153
154 # qhasm: int64 h11
155
156 # qhasm: int64 h12
157
158 # qhasm: int64 h13
159
160 # qhasm: int64 h14
161
162 # qhasm: int64 h15
163
164 # qhasm: int64 h16
165
166 # qhasm: int64 h17
167
168 # qhasm: int64 h18
169
170 # qhasm: int64 h19
171
172 # qhasm: int64 h20
173
174 # qhasm: int64 h21
175
176 # qhasm: int64 h22
177
178 # qhasm: stack4864 buf
179
180 # qhasm: int64 ptr
181
182 # qhasm: int64 tmp
183
184 # qhasm: stack64 r11_stack
185
186 # qhasm: stack64 r12_stack
187
188 # qhasm: stack64 r13_stack
189
190 # qhasm: stack64 r14_stack
191
192 # qhasm: stack64 r15_stack
193
194 # qhasm: stack64 rbx_stack
195
196 # qhasm: stack64 rbp_stack
197
198 # qhasm: enter vec_mul_asm
199 .p2align 5
200 ASM_HIDDEN _vec_mul_asm
201 ASM_HIDDEN vec_mul_asm
202 .global _vec_mul_asm
203 .global vec_mul_asm
204 _vec_mul_asm:
205 vec_mul_asm:
206 mov %rsp,%r11
207 AND $31,%r11
208 add $672,%r11
209 sub %r11,%rsp
210
211 # qhasm: r11_stack = caller_r11
212 # asm 1: movq <caller_r11=int64#9,>r11_stack=stack64#1
213 # asm 2: movq <caller_r11=%r11,>r11_stack=608(%rsp)
214 movq %r11,608(%rsp)
215
216 # qhasm: r12_stack = caller_r12
217 # asm 1: movq <caller_r12=int64#10,>r12_stack=stack64#2
218 # asm 2: movq <caller_r12=%r12,>r12_stack=616(%rsp)
219 movq %r12,616(%rsp)
220
221 # qhasm: r13_stack = caller_r13
222 # asm 1: movq <caller_r13=int64#11,>r13_stack=stack64#3
223 # asm 2: movq <caller_r13=%r13,>r13_stack=624(%rsp)
224 movq %r13,624(%rsp)
225
226 # qhasm: r14_stack = caller_r14
227 # asm 1: movq <caller_r14=int64#12,>r14_stack=stack64#4
228 # asm 2: movq <caller_r14=%r14,>r14_stack=632(%rsp)
229 movq %r14,632(%rsp)
230
231 # qhasm: r15_stack = caller_r15
232 # asm 1: movq <caller_r15=int64#13,>r15_stack=stack64#5
233 # asm 2: movq <caller_r15=%r15,>r15_stack=640(%rsp)
234 movq %r15,640(%rsp)
235
236 # qhasm: rbx_stack = caller_rbx
237 # asm 1: movq <caller_rbx=int64#14,>rbx_stack=stack64#6
238 # asm 2: movq <caller_rbx=%rbx,>rbx_stack=648(%rsp)
239 movq %rbx,648(%rsp)
240
241 # qhasm: ptr = &buf
242 # asm 1: leaq <buf=stack4864#1,>ptr=int64#4
243 # asm 2: leaq <buf=0(%rsp),>ptr=%rcx
244 leaq 0(%rsp),%rcx
245
246 # qhasm: s0 = mem256[ input_1 + 0 ]
247 # asm 1: vmovupd 0(<input_1=int64#2),>s0=reg256#1
248 # asm 2: vmovupd 0(<input_1=%rsi),>s0=%ymm0
249 vmovupd 0(%rsi),%ymm0
250
251 # qhasm: s1 = mem256[ input_1 + 32 ]
252 # asm 1: vmovupd 32(<input_1=int64#2),>s1=reg256#2
253 # asm 2: vmovupd 32(<input_1=%rsi),>s1=%ymm1
254 vmovupd 32(%rsi),%ymm1
255
256 # qhasm: s2 = mem256[ input_1 + 64 ]
257 # asm 1: vmovupd 64(<input_1=int64#2),>s2=reg256#3
258 # asm 2: vmovupd 64(<input_1=%rsi),>s2=%ymm2
259 vmovupd 64(%rsi),%ymm2
260
261 # qhasm: t0 = mem256[ input_2 + 0 ]
262 # asm 1: vmovupd 0(<input_2=int64#3),>t0=reg256#4
263 # asm 2: vmovupd 0(<input_2=%rdx),>t0=%ymm3
264 vmovupd 0(%rdx),%ymm3
265
266 # qhasm: t1 = mem256[ input_2 + 32 ]
267 # asm 1: vmovupd 32(<input_2=int64#3),>t1=reg256#5
268 # asm 2: vmovupd 32(<input_2=%rdx),>t1=%ymm4
269 vmovupd 32(%rdx),%ymm4
270
271 # qhasm: t2 = mem256[ input_2 + 64 ]
272 # asm 1: vmovupd 64(<input_2=int64#3),>t2=reg256#6
273 # asm 2: vmovupd 64(<input_2=%rdx),>t2=%ymm5
274 vmovupd 64(%rdx),%ymm5
275
276 # qhasm: a5[0,1,2,3] = s2[2,2,3,3]
277 # asm 1: vpermq $0xfa,<s2=reg256#3,>a5=reg256#7
278 # asm 2: vpermq $0xfa,<s2=%ymm2,>a5=%ymm6
279 vpermq $0xfa,%ymm2,%ymm6
280
281 # qhasm: b5[0,1,2,3] = t2[2,3,2,3]
282 # asm 1: vpermq $0xee,<t2=reg256#6,>b5=reg256#8
283 # asm 2: vpermq $0xee,<t2=%ymm5,>b5=%ymm7
284 vpermq $0xee,%ymm5,%ymm7
285
286 # qhasm: r10 = a5 & b5
287 # asm 1: vpand <a5=reg256#7,<b5=reg256#8,>r10=reg256#9
288 # asm 2: vpand <a5=%ymm6,<b5=%ymm7,>r10=%ymm8
289 vpand %ymm6,%ymm7,%ymm8
290
291 # qhasm: mem256[ ptr + 320 ] = r10
292 # asm 1: vmovupd <r10=reg256#9,320(<ptr=int64#4)
293 # asm 2: vmovupd <r10=%ymm8,320(<ptr=%rcx)
294 vmovupd %ymm8,320(%rcx)
295
296 # qhasm: b4[0,1,2,3] = t2[0,1,0,1]
297 # asm 1: vpermq $0x44,<t2=reg256#6,>b4=reg256#6
298 # asm 2: vpermq $0x44,<t2=%ymm5,>b4=%ymm5
299 vpermq $0x44,%ymm5,%ymm5
300
301 # qhasm: r9 = a5 & b4
302 # asm 1: vpand <a5=reg256#7,<b4=reg256#6,>r9=reg256#9
303 # asm 2: vpand <a5=%ymm6,<b4=%ymm5,>r9=%ymm8
304 vpand %ymm6,%ymm5,%ymm8
305
306 # qhasm: b3[0,1,2,3] = t1[2,3,2,3]
307 # asm 1: vpermq $0xee,<t1=reg256#5,>b3=reg256#10
308 # asm 2: vpermq $0xee,<t1=%ymm4,>b3=%ymm9
309 vpermq $0xee,%ymm4,%ymm9
310
311 # qhasm: r8 = a5 & b3
312 # asm 1: vpand <a5=reg256#7,<b3=reg256#10,>r8=reg256#11
313 # asm 2: vpand <a5=%ymm6,<b3=%ymm9,>r8=%ymm10
314 vpand %ymm6,%ymm9,%ymm10
315
316 # qhasm: b2[0,1,2,3] = t1[0,1,0,1]
317 # asm 1: vpermq $0x44,<t1=reg256#5,>b2=reg256#5
318 # asm 2: vpermq $0x44,<t1=%ymm4,>b2=%ymm4
319 vpermq $0x44,%ymm4,%ymm4
320
321 # qhasm: r7 = a5 & b2
322 # asm 1: vpand <a5=reg256#7,<b2=reg256#5,>r7=reg256#12
323 # asm 2: vpand <a5=%ymm6,<b2=%ymm4,>r7=%ymm11
324 vpand %ymm6,%ymm4,%ymm11
325
326 # qhasm: b1[0,1,2,3] = t0[2,3,2,3]
327 # asm 1: vpermq $0xee,<t0=reg256#4,>b1=reg256#13
328 # asm 2: vpermq $0xee,<t0=%ymm3,>b1=%ymm12
329 vpermq $0xee,%ymm3,%ymm12
330
331 # qhasm: r6 = a5 & b1
332 # asm 1: vpand <a5=reg256#7,<b1=reg256#13,>r6=reg256#14
333 # asm 2: vpand <a5=%ymm6,<b1=%ymm12,>r6=%ymm13
334 vpand %ymm6,%ymm12,%ymm13
335
336 # qhasm: b0[0,1,2,3] = t0[0,1,0,1]
337 # asm 1: vpermq $0x44,<t0=reg256#4,>b0=reg256#4
338 # asm 2: vpermq $0x44,<t0=%ymm3,>b0=%ymm3
339 vpermq $0x44,%ymm3,%ymm3
340
341 # qhasm: r5 = a5 & b0
342 # asm 1: vpand <a5=reg256#7,<b0=reg256#4,>r5=reg256#7
343 # asm 2: vpand <a5=%ymm6,<b0=%ymm3,>r5=%ymm6
344 vpand %ymm6,%ymm3,%ymm6
345
346 # qhasm: a4[0,1,2,3] = s2[0,0,1,1]
347 # asm 1: vpermq $0x50,<s2=reg256#3,>a4=reg256#3
348 # asm 2: vpermq $0x50,<s2=%ymm2,>a4=%ymm2
349 vpermq $0x50,%ymm2,%ymm2
350
351 # qhasm: r = a4 & b5
352 # asm 1: vpand <a4=reg256#3,<b5=reg256#8,>r=reg256#15
353 # asm 2: vpand <a4=%ymm2,<b5=%ymm7,>r=%ymm14
354 vpand %ymm2,%ymm7,%ymm14
355
356 # qhasm: r9 ^= r
357 # asm 1: vpxor <r=reg256#15,<r9=reg256#9,<r9=reg256#9
358 # asm 2: vpxor <r=%ymm14,<r9=%ymm8,<r9=%ymm8
359 vpxor %ymm14,%ymm8,%ymm8
360
361 # qhasm: mem256[ ptr + 288 ] = r9
362 # asm 1: vmovupd <r9=reg256#9,288(<ptr=int64#4)
363 # asm 2: vmovupd <r9=%ymm8,288(<ptr=%rcx)
364 vmovupd %ymm8,288(%rcx)
365
366 # qhasm: r = a4 & b4
367 # asm 1: vpand <a4=reg256#3,<b4=reg256#6,>r=reg256#9
368 # asm 2: vpand <a4=%ymm2,<b4=%ymm5,>r=%ymm8
369 vpand %ymm2,%ymm5,%ymm8
370
371 # qhasm: r8 ^= r
372 # asm 1: vpxor <r=reg256#9,<r8=reg256#11,<r8=reg256#11
373 # asm 2: vpxor <r=%ymm8,<r8=%ymm10,<r8=%ymm10
374 vpxor %ymm8,%ymm10,%ymm10
375
376 # qhasm: r = a4 & b3
377 # asm 1: vpand <a4=reg256#3,<b3=reg256#10,>r=reg256#9
378 # asm 2: vpand <a4=%ymm2,<b3=%ymm9,>r=%ymm8
379 vpand %ymm2,%ymm9,%ymm8
380
381 # qhasm: r7 ^= r
382 # asm 1: vpxor <r=reg256#9,<r7=reg256#12,<r7=reg256#12
383 # asm 2: vpxor <r=%ymm8,<r7=%ymm11,<r7=%ymm11
384 vpxor %ymm8,%ymm11,%ymm11
385
386 # qhasm: r = a4 & b2
387 # asm 1: vpand <a4=reg256#3,<b2=reg256#5,>r=reg256#9
388 # asm 2: vpand <a4=%ymm2,<b2=%ymm4,>r=%ymm8
389 vpand %ymm2,%ymm4,%ymm8
390
391 # qhasm: r6 ^= r
392 # asm 1: vpxor <r=reg256#9,<r6=reg256#14,<r6=reg256#14
393 # asm 2: vpxor <r=%ymm8,<r6=%ymm13,<r6=%ymm13
394 vpxor %ymm8,%ymm13,%ymm13
395
396 # qhasm: r = a4 & b1
397 # asm 1: vpand <a4=reg256#3,<b1=reg256#13,>r=reg256#9
398 # asm 2: vpand <a4=%ymm2,<b1=%ymm12,>r=%ymm8
399 vpand %ymm2,%ymm12,%ymm8
400
401 # qhasm: r5 ^= r
402 # asm 1: vpxor <r=reg256#9,<r5=reg256#7,<r5=reg256#7
403 # asm 2: vpxor <r=%ymm8,<r5=%ymm6,<r5=%ymm6
404 vpxor %ymm8,%ymm6,%ymm6
405
406 # qhasm: r4 = a4 & b0
407 # asm 1: vpand <a4=reg256#3,<b0=reg256#4,>r4=reg256#3
408 # asm 2: vpand <a4=%ymm2,<b0=%ymm3,>r4=%ymm2
409 vpand %ymm2,%ymm3,%ymm2
410
411 # qhasm: a3[0,1,2,3] = s1[2,2,3,3]
412 # asm 1: vpermq $0xfa,<s1=reg256#2,>a3=reg256#9
413 # asm 2: vpermq $0xfa,<s1=%ymm1,>a3=%ymm8
414 vpermq $0xfa,%ymm1,%ymm8
415
416 # qhasm: r = a3 & b5
417 # asm 1: vpand <a3=reg256#9,<b5=reg256#8,>r=reg256#15
418 # asm 2: vpand <a3=%ymm8,<b5=%ymm7,>r=%ymm14
419 vpand %ymm8,%ymm7,%ymm14
420
421 # qhasm: r8 ^= r
422 # asm 1: vpxor <r=reg256#15,<r8=reg256#11,<r8=reg256#11
423 # asm 2: vpxor <r=%ymm14,<r8=%ymm10,<r8=%ymm10
424 vpxor %ymm14,%ymm10,%ymm10
425
426 # qhasm: mem256[ ptr + 256 ] = r8
427 # asm 1: vmovupd <r8=reg256#11,256(<ptr=int64#4)
428 # asm 2: vmovupd <r8=%ymm10,256(<ptr=%rcx)
429 vmovupd %ymm10,256(%rcx)
430
431 # qhasm: r = a3 & b4
432 # asm 1: vpand <a3=reg256#9,<b4=reg256#6,>r=reg256#11
433 # asm 2: vpand <a3=%ymm8,<b4=%ymm5,>r=%ymm10
434 vpand %ymm8,%ymm5,%ymm10
435
436 # qhasm: r7 ^= r
437 # asm 1: vpxor <r=reg256#11,<r7=reg256#12,<r7=reg256#12
438 # asm 2: vpxor <r=%ymm10,<r7=%ymm11,<r7=%ymm11
439 vpxor %ymm10,%ymm11,%ymm11
440
441 # qhasm: r = a3 & b3
442 # asm 1: vpand <a3=reg256#9,<b3=reg256#10,>r=reg256#11
443 # asm 2: vpand <a3=%ymm8,<b3=%ymm9,>r=%ymm10
444 vpand %ymm8,%ymm9,%ymm10
445
446 # qhasm: r6 ^= r
447 # asm 1: vpxor <r=reg256#11,<r6=reg256#14,<r6=reg256#14
448 # asm 2: vpxor <r=%ymm10,<r6=%ymm13,<r6=%ymm13
449 vpxor %ymm10,%ymm13,%ymm13
450
451 # qhasm: r = a3 & b2
452 # asm 1: vpand <a3=reg256#9,<b2=reg256#5,>r=reg256#11
453 # asm 2: vpand <a3=%ymm8,<b2=%ymm4,>r=%ymm10
454 vpand %ymm8,%ymm4,%ymm10
455
456 # qhasm: r5 ^= r
457 # asm 1: vpxor <r=reg256#11,<r5=reg256#7,<r5=reg256#7
458 # asm 2: vpxor <r=%ymm10,<r5=%ymm6,<r5=%ymm6
459 vpxor %ymm10,%ymm6,%ymm6
460
461 # qhasm: r = a3 & b1
462 # asm 1: vpand <a3=reg256#9,<b1=reg256#13,>r=reg256#11
463 # asm 2: vpand <a3=%ymm8,<b1=%ymm12,>r=%ymm10
464 vpand %ymm8,%ymm12,%ymm10
465
466 # qhasm: r4 ^= r
467 # asm 1: vpxor <r=reg256#11,<r4=reg256#3,<r4=reg256#3
468 # asm 2: vpxor <r=%ymm10,<r4=%ymm2,<r4=%ymm2
469 vpxor %ymm10,%ymm2,%ymm2
470
471 # qhasm: r3 = a3 & b0
472 # asm 1: vpand <a3=reg256#9,<b0=reg256#4,>r3=reg256#9
473 # asm 2: vpand <a3=%ymm8,<b0=%ymm3,>r3=%ymm8
474 vpand %ymm8,%ymm3,%ymm8
475
476 # qhasm: a2[0,1,2,3] = s1[0,0,1,1]
477 # asm 1: vpermq $0x50,<s1=reg256#2,>a2=reg256#2
478 # asm 2: vpermq $0x50,<s1=%ymm1,>a2=%ymm1
479 vpermq $0x50,%ymm1,%ymm1
480
481 # qhasm: r = a2 & b5
482 # asm 1: vpand <a2=reg256#2,<b5=reg256#8,>r=reg256#11
483 # asm 2: vpand <a2=%ymm1,<b5=%ymm7,>r=%ymm10
484 vpand %ymm1,%ymm7,%ymm10
485
486 # qhasm: r7 ^= r
487 # asm 1: vpxor <r=reg256#11,<r7=reg256#12,<r7=reg256#12
488 # asm 2: vpxor <r=%ymm10,<r7=%ymm11,<r7=%ymm11
489 vpxor %ymm10,%ymm11,%ymm11
490
491 # qhasm: mem256[ ptr + 224 ] = r7
492 # asm 1: vmovupd <r7=reg256#12,224(<ptr=int64#4)
493 # asm 2: vmovupd <r7=%ymm11,224(<ptr=%rcx)
494 vmovupd %ymm11,224(%rcx)
495
496 # qhasm: r = a2 & b4
497 # asm 1: vpand <a2=reg256#2,<b4=reg256#6,>r=reg256#11
498 # asm 2: vpand <a2=%ymm1,<b4=%ymm5,>r=%ymm10
499 vpand %ymm1,%ymm5,%ymm10
500
501 # qhasm: r6 ^= r
502 # asm 1: vpxor <r=reg256#11,<r6=reg256#14,<r6=reg256#14
503 # asm 2: vpxor <r=%ymm10,<r6=%ymm13,<r6=%ymm13
504 vpxor %ymm10,%ymm13,%ymm13
505
506 # qhasm: r = a2 & b3
507 # asm 1: vpand <a2=reg256#2,<b3=reg256#10,>r=reg256#11
508 # asm 2: vpand <a2=%ymm1,<b3=%ymm9,>r=%ymm10
509 vpand %ymm1,%ymm9,%ymm10
510
511 # qhasm: r5 ^= r
512 # asm 1: vpxor <r=reg256#11,<r5=reg256#7,<r5=reg256#7
513 # asm 2: vpxor <r=%ymm10,<r5=%ymm6,<r5=%ymm6
514 vpxor %ymm10,%ymm6,%ymm6
515
516 # qhasm: r = a2 & b2
517 # asm 1: vpand <a2=reg256#2,<b2=reg256#5,>r=reg256#11
518 # asm 2: vpand <a2=%ymm1,<b2=%ymm4,>r=%ymm10
519 vpand %ymm1,%ymm4,%ymm10
520
521 # qhasm: r4 ^= r
522 # asm 1: vpxor <r=reg256#11,<r4=reg256#3,<r4=reg256#3
523 # asm 2: vpxor <r=%ymm10,<r4=%ymm2,<r4=%ymm2
524 vpxor %ymm10,%ymm2,%ymm2
525
526 # qhasm: r = a2 & b1
527 # asm 1: vpand <a2=reg256#2,<b1=reg256#13,>r=reg256#11
528 # asm 2: vpand <a2=%ymm1,<b1=%ymm12,>r=%ymm10
529 vpand %ymm1,%ymm12,%ymm10
530
531 # qhasm: r3 ^= r
532 # asm 1: vpxor <r=reg256#11,<r3=reg256#9,<r3=reg256#9
533 # asm 2: vpxor <r=%ymm10,<r3=%ymm8,<r3=%ymm8
534 vpxor %ymm10,%ymm8,%ymm8
535
536 # qhasm: r2 = a2 & b0
537 # asm 1: vpand <a2=reg256#2,<b0=reg256#4,>r2=reg256#2
538 # asm 2: vpand <a2=%ymm1,<b0=%ymm3,>r2=%ymm1
539 vpand %ymm1,%ymm3,%ymm1
540
541 # qhasm: a1[0,1,2,3] = s0[2,2,3,3]
542 # asm 1: vpermq $0xfa,<s0=reg256#1,>a1=reg256#11
543 # asm 2: vpermq $0xfa,<s0=%ymm0,>a1=%ymm10
544 vpermq $0xfa,%ymm0,%ymm10
545
546 # qhasm: r = a1 & b5
547 # asm 1: vpand <a1=reg256#11,<b5=reg256#8,>r=reg256#12
548 # asm 2: vpand <a1=%ymm10,<b5=%ymm7,>r=%ymm11
549 vpand %ymm10,%ymm7,%ymm11
550
551 # qhasm: r6 ^= r
552 # asm 1: vpxor <r=reg256#12,<r6=reg256#14,<r6=reg256#14
553 # asm 2: vpxor <r=%ymm11,<r6=%ymm13,<r6=%ymm13
554 vpxor %ymm11,%ymm13,%ymm13
555
556 # qhasm: mem256[ ptr + 192 ] = r6
557 # asm 1: vmovupd <r6=reg256#14,192(<ptr=int64#4)
558 # asm 2: vmovupd <r6=%ymm13,192(<ptr=%rcx)
559 vmovupd %ymm13,192(%rcx)
560
561 # qhasm: r = a1 & b4
562 # asm 1: vpand <a1=reg256#11,<b4=reg256#6,>r=reg256#12
563 # asm 2: vpand <a1=%ymm10,<b4=%ymm5,>r=%ymm11
564 vpand %ymm10,%ymm5,%ymm11
565
566 # qhasm: r5 ^= r
567 # asm 1: vpxor <r=reg256#12,<r5=reg256#7,<r5=reg256#7
568 # asm 2: vpxor <r=%ymm11,<r5=%ymm6,<r5=%ymm6
569 vpxor %ymm11,%ymm6,%ymm6
570
571 # qhasm: r = a1 & b3
572 # asm 1: vpand <a1=reg256#11,<b3=reg256#10,>r=reg256#12
573 # asm 2: vpand <a1=%ymm10,<b3=%ymm9,>r=%ymm11
574 vpand %ymm10,%ymm9,%ymm11
575
576 # qhasm: r4 ^= r
577 # asm 1: vpxor <r=reg256#12,<r4=reg256#3,<r4=reg256#3
578 # asm 2: vpxor <r=%ymm11,<r4=%ymm2,<r4=%ymm2
579 vpxor %ymm11,%ymm2,%ymm2
580
581 # qhasm: r = a1 & b2
582 # asm 1: vpand <a1=reg256#11,<b2=reg256#5,>r=reg256#12
583 # asm 2: vpand <a1=%ymm10,<b2=%ymm4,>r=%ymm11
584 vpand %ymm10,%ymm4,%ymm11
585
586 # qhasm: r3 ^= r
587 # asm 1: vpxor <r=reg256#12,<r3=reg256#9,<r3=reg256#9
588 # asm 2: vpxor <r=%ymm11,<r3=%ymm8,<r3=%ymm8
589 vpxor %ymm11,%ymm8,%ymm8
590
591 # qhasm: r = a1 & b1
592 # asm 1: vpand <a1=reg256#11,<b1=reg256#13,>r=reg256#12
593 # asm 2: vpand <a1=%ymm10,<b1=%ymm12,>r=%ymm11
594 vpand %ymm10,%ymm12,%ymm11
595
596 # qhasm: r2 ^= r
597 # asm 1: vpxor <r=reg256#12,<r2=reg256#2,<r2=reg256#2
598 # asm 2: vpxor <r=%ymm11,<r2=%ymm1,<r2=%ymm1
599 vpxor %ymm11,%ymm1,%ymm1
600
601 # qhasm: r1 = a1 & b0
602 # asm 1: vpand <a1=reg256#11,<b0=reg256#4,>r1=reg256#11
603 # asm 2: vpand <a1=%ymm10,<b0=%ymm3,>r1=%ymm10
604 vpand %ymm10,%ymm3,%ymm10
605
606 # qhasm: a0[0,1,2,3] = s0[0,0,1,1]
607 # asm 1: vpermq $0x50,<s0=reg256#1,>a0=reg256#1
608 # asm 2: vpermq $0x50,<s0=%ymm0,>a0=%ymm0
609 vpermq $0x50,%ymm0,%ymm0
610
611 # qhasm: r = a0 & b5
612 # asm 1: vpand <a0=reg256#1,<b5=reg256#8,>r=reg256#8
613 # asm 2: vpand <a0=%ymm0,<b5=%ymm7,>r=%ymm7
614 vpand %ymm0,%ymm7,%ymm7
615
616 # qhasm: r5 ^= r
617 # asm 1: vpxor <r=reg256#8,<r5=reg256#7,<r5=reg256#7
618 # asm 2: vpxor <r=%ymm7,<r5=%ymm6,<r5=%ymm6
619 vpxor %ymm7,%ymm6,%ymm6
620
621 # qhasm: mem256[ ptr + 160 ] = r5
622 # asm 1: vmovupd <r5=reg256#7,160(<ptr=int64#4)
623 # asm 2: vmovupd <r5=%ymm6,160(<ptr=%rcx)
624 vmovupd %ymm6,160(%rcx)
625
626 # qhasm: r = a0 & b4
627 # asm 1: vpand <a0=reg256#1,<b4=reg256#6,>r=reg256#6
628 # asm 2: vpand <a0=%ymm0,<b4=%ymm5,>r=%ymm5
629 vpand %ymm0,%ymm5,%ymm5
630
631 # qhasm: r4 ^= r
632 # asm 1: vpxor <r=reg256#6,<r4=reg256#3,<r4=reg256#3
633 # asm 2: vpxor <r=%ymm5,<r4=%ymm2,<r4=%ymm2
634 vpxor %ymm5,%ymm2,%ymm2
635
636 # qhasm: r = a0 & b3
637 # asm 1: vpand <a0=reg256#1,<b3=reg256#10,>r=reg256#6
638 # asm 2: vpand <a0=%ymm0,<b3=%ymm9,>r=%ymm5
639 vpand %ymm0,%ymm9,%ymm5
640
641 # qhasm: r3 ^= r
642 # asm 1: vpxor <r=reg256#6,<r3=reg256#9,<r3=reg256#9
643 # asm 2: vpxor <r=%ymm5,<r3=%ymm8,<r3=%ymm8
644 vpxor %ymm5,%ymm8,%ymm8
645
646 # qhasm: r = a0 & b2
647 # asm 1: vpand <a0=reg256#1,<b2=reg256#5,>r=reg256#5
648 # asm 2: vpand <a0=%ymm0,<b2=%ymm4,>r=%ymm4
649 vpand %ymm0,%ymm4,%ymm4
650
651 # qhasm: r2 ^= r
652 # asm 1: vpxor <r=reg256#5,<r2=reg256#2,<r2=reg256#2
653 # asm 2: vpxor <r=%ymm4,<r2=%ymm1,<r2=%ymm1
654 vpxor %ymm4,%ymm1,%ymm1
655
656 # qhasm: r = a0 & b1
657 # asm 1: vpand <a0=reg256#1,<b1=reg256#13,>r=reg256#5
658 # asm 2: vpand <a0=%ymm0,<b1=%ymm12,>r=%ymm4
659 vpand %ymm0,%ymm12,%ymm4
660
661 # qhasm: r1 ^= r
662 # asm 1: vpxor <r=reg256#5,<r1=reg256#11,<r1=reg256#11
663 # asm 2: vpxor <r=%ymm4,<r1=%ymm10,<r1=%ymm10
664 vpxor %ymm4,%ymm10,%ymm10
665
666 # qhasm: r0 = a0 & b0
667 # asm 1: vpand <a0=reg256#1,<b0=reg256#4,>r0=reg256#1
668 # asm 2: vpand <a0=%ymm0,<b0=%ymm3,>r0=%ymm0
669 vpand %ymm0,%ymm3,%ymm0
670
671 # qhasm: mem256[ ptr + 128 ] = r4
672 # asm 1: vmovupd <r4=reg256#3,128(<ptr=int64#4)
673 # asm 2: vmovupd <r4=%ymm2,128(<ptr=%rcx)
674 vmovupd %ymm2,128(%rcx)
675
676 # qhasm: mem256[ ptr + 96 ] = r3
677 # asm 1: vmovupd <r3=reg256#9,96(<ptr=int64#4)
678 # asm 2: vmovupd <r3=%ymm8,96(<ptr=%rcx)
679 vmovupd %ymm8,96(%rcx)
680
681 # qhasm: mem256[ ptr + 64 ] = r2
682 # asm 1: vmovupd <r2=reg256#2,64(<ptr=int64#4)
683 # asm 2: vmovupd <r2=%ymm1,64(<ptr=%rcx)
684 vmovupd %ymm1,64(%rcx)
685
686 # qhasm: mem256[ ptr + 32 ] = r1
687 # asm 1: vmovupd <r1=reg256#11,32(<ptr=int64#4)
688 # asm 2: vmovupd <r1=%ymm10,32(<ptr=%rcx)
689 vmovupd %ymm10,32(%rcx)
690
691 # qhasm: mem256[ ptr + 0 ] = r0
692 # asm 1: vmovupd <r0=reg256#1,0(<ptr=int64#4)
693 # asm 2: vmovupd <r0=%ymm0,0(<ptr=%rcx)
694 vmovupd %ymm0,0(%rcx)
695
696 # qhasm: h22 = mem64[ ptr + 344 ]
697 # asm 1: movq 344(<ptr=int64#4),>h22=int64#2
698 # asm 2: movq 344(<ptr=%rcx),>h22=%rsi
699 movq 344(%rcx),%rsi
700
701 # qhasm: h13 = h22
702 # asm 1: mov <h22=int64#2,>h13=int64#3
703 # asm 2: mov <h22=%rsi,>h13=%rdx
704 mov %rsi,%rdx
705
706 # qhasm: h10 = h22
707 # asm 1: mov <h22=int64#2,>h10=int64#2
708 # asm 2: mov <h22=%rsi,>h10=%rsi
709 mov %rsi,%rsi
710
711 # qhasm: h21 = mem64[ ptr + 336 ]
712 # asm 1: movq 336(<ptr=int64#4),>h21=int64#5
713 # asm 2: movq 336(<ptr=%rcx),>h21=%r8
714 movq 336(%rcx),%r8
715
716 # qhasm: h21 ^= *(uint64 *) ( ptr + 328 )
717 # asm 1: xorq 328(<ptr=int64#4),<h21=int64#5
718 # asm 2: xorq 328(<ptr=%rcx),<h21=%r8
719 xorq 328(%rcx),%r8
720
721 # qhasm: h12 = h21
722 # asm 1: mov <h21=int64#5,>h12=int64#6
723 # asm 2: mov <h21=%r8,>h12=%r9
724 mov %r8,%r9
725
726 # qhasm: h9 = h21
727 # asm 1: mov <h21=int64#5,>h9=int64#5
728 # asm 2: mov <h21=%r8,>h9=%r8
729 mov %r8,%r8
730
731 # qhasm: h20 = mem64[ ptr + 312 ]
732 # asm 1: movq 312(<ptr=int64#4),>h20=int64#7
733 # asm 2: movq 312(<ptr=%rcx),>h20=%rax
734 movq 312(%rcx),%rax
735
736 # qhasm: h20 ^= *(uint64 *) ( ptr + 320 )
737 # asm 1: xorq 320(<ptr=int64#4),<h20=int64#7
738 # asm 2: xorq 320(<ptr=%rcx),<h20=%rax
739 xorq 320(%rcx),%rax
740
741 # qhasm: h11 = h20
742 # asm 1: mov <h20=int64#7,>h11=int64#8
743 # asm 2: mov <h20=%rax,>h11=%r10
744 mov %rax,%r10
745
746 # qhasm: h8 = h20
747 # asm 1: mov <h20=int64#7,>h8=int64#7
748 # asm 2: mov <h20=%rax,>h8=%rax
749 mov %rax,%rax
750
751 # qhasm: h19 = mem64[ ptr + 304 ]
752 # asm 1: movq 304(<ptr=int64#4),>h19=int64#9
753 # asm 2: movq 304(<ptr=%rcx),>h19=%r11
754 movq 304(%rcx),%r11
755
756 # qhasm: h19 ^= *(uint64 *) ( ptr + 296 )
757 # asm 1: xorq 296(<ptr=int64#4),<h19=int64#9
758 # asm 2: xorq 296(<ptr=%rcx),<h19=%r11
759 xorq 296(%rcx),%r11
760
761 # qhasm: h10 ^= h19
762 # asm 1: XOR <h19=int64#9,<h10=int64#2
763 # asm 2: XOR <h19=%r11,<h10=%rsi
764 XOR %r11,%rsi
765
766 # qhasm: h7 = h19
767 # asm 1: mov <h19=int64#9,>h7=int64#9
768 # asm 2: mov <h19=%r11,>h7=%r11
769 mov %r11,%r11
770
771 # qhasm: h18 = mem64[ ptr + 280 ]
772 # asm 1: movq 280(<ptr=int64#4),>h18=int64#10
773 # asm 2: movq 280(<ptr=%rcx),>h18=%r12
774 movq 280(%rcx),%r12
775
776 # qhasm: h18 ^= *(uint64 *) ( ptr + 288 )
777 # asm 1: xorq 288(<ptr=int64#4),<h18=int64#10
778 # asm 2: xorq 288(<ptr=%rcx),<h18=%r12
779 xorq 288(%rcx),%r12
780
781 # qhasm: h9 ^= h18
782 # asm 1: XOR <h18=int64#10,<h9=int64#5
783 # asm 2: XOR <h18=%r12,<h9=%r8
784 XOR %r12,%r8
785
786 # qhasm: h6 = h18
787 # asm 1: mov <h18=int64#10,>h6=int64#10
788 # asm 2: mov <h18=%r12,>h6=%r12
789 mov %r12,%r12
790
791 # qhasm: h17 = mem64[ ptr + 272 ]
792 # asm 1: movq 272(<ptr=int64#4),>h17=int64#11
793 # asm 2: movq 272(<ptr=%rcx),>h17=%r13
794 movq 272(%rcx),%r13
795
796 # qhasm: h17 ^= *(uint64 *) ( ptr + 264 )
797 # asm 1: xorq 264(<ptr=int64#4),<h17=int64#11
798 # asm 2: xorq 264(<ptr=%rcx),<h17=%r13
799 xorq 264(%rcx),%r13
800
801 # qhasm: h8 ^= h17
802 # asm 1: XOR <h17=int64#11,<h8=int64#7
803 # asm 2: XOR <h17=%r13,<h8=%rax
804 XOR %r13,%rax
805
806 # qhasm: h5 = h17
807 # asm 1: mov <h17=int64#11,>h5=int64#11
808 # asm 2: mov <h17=%r13,>h5=%r13
809 mov %r13,%r13
810
811 # qhasm: h16 = mem64[ ptr + 248 ]
812 # asm 1: movq 248(<ptr=int64#4),>h16=int64#12
813 # asm 2: movq 248(<ptr=%rcx),>h16=%r14
814 movq 248(%rcx),%r14
815
816 # qhasm: h16 ^= *(uint64 *) ( ptr + 256 )
817 # asm 1: xorq 256(<ptr=int64#4),<h16=int64#12
818 # asm 2: xorq 256(<ptr=%rcx),<h16=%r14
819 xorq 256(%rcx),%r14
820
821 # qhasm: h7 ^= h16
822 # asm 1: XOR <h16=int64#12,<h7=int64#9
823 # asm 2: XOR <h16=%r14,<h7=%r11
824 XOR %r14,%r11
825
826 # qhasm: h4 = h16
827 # asm 1: mov <h16=int64#12,>h4=int64#12
828 # asm 2: mov <h16=%r14,>h4=%r14
829 mov %r14,%r14
830
831 # qhasm: h15 = mem64[ ptr + 240 ]
832 # asm 1: movq 240(<ptr=int64#4),>h15=int64#13
833 # asm 2: movq 240(<ptr=%rcx),>h15=%r15
834 movq 240(%rcx),%r15
835
836 # qhasm: h15 ^= *(uint64 *) ( ptr + 232 )
837 # asm 1: xorq 232(<ptr=int64#4),<h15=int64#13
838 # asm 2: xorq 232(<ptr=%rcx),<h15=%r15
839 xorq 232(%rcx),%r15
840
841 # qhasm: h6 ^= h15
842 # asm 1: XOR <h15=int64#13,<h6=int64#10
843 # asm 2: XOR <h15=%r15,<h6=%r12
844 XOR %r15,%r12
845
846 # qhasm: h3 = h15
847 # asm 1: mov <h15=int64#13,>h3=int64#13
848 # asm 2: mov <h15=%r15,>h3=%r15
849 mov %r15,%r15
850
851 # qhasm: h14 = mem64[ ptr + 216 ]
852 # asm 1: movq 216(<ptr=int64#4),>h14=int64#14
853 # asm 2: movq 216(<ptr=%rcx),>h14=%rbx
854 movq 216(%rcx),%rbx
855
856 # qhasm: h14 ^= *(uint64 *) ( ptr + 224 )
857 # asm 1: xorq 224(<ptr=int64#4),<h14=int64#14
858 # asm 2: xorq 224(<ptr=%rcx),<h14=%rbx
859 xorq 224(%rcx),%rbx
860
861 # qhasm: h5 ^= h14
862 # asm 1: XOR <h14=int64#14,<h5=int64#11
863 # asm 2: XOR <h14=%rbx,<h5=%r13
864 XOR %rbx,%r13
865
866 # qhasm: h2 = h14
867 # asm 1: mov <h14=int64#14,>h2=int64#14
868 # asm 2: mov <h14=%rbx,>h2=%rbx
869 mov %rbx,%rbx
870
871 # qhasm: h13 ^= *(uint64 *) ( ptr + 208 )
872 # asm 1: xorq 208(<ptr=int64#4),<h13=int64#3
873 # asm 2: xorq 208(<ptr=%rcx),<h13=%rdx
874 xorq 208(%rcx),%rdx
875
876 # qhasm: h13 ^= *(uint64 *) ( ptr + 200 )
877 # asm 1: xorq 200(<ptr=int64#4),<h13=int64#3
878 # asm 2: xorq 200(<ptr=%rcx),<h13=%rdx
879 xorq 200(%rcx),%rdx
880
881 # qhasm: h4 ^= h13
882 # asm 1: XOR <h13=int64#3,<h4=int64#12
883 # asm 2: XOR <h13=%rdx,<h4=%r14
884 XOR %rdx,%r14
885
886 # qhasm: h1 = h13
887 # asm 1: mov <h13=int64#3,>h1=int64#3
888 # asm 2: mov <h13=%rdx,>h1=%rdx
889 mov %rdx,%rdx
890
891 # qhasm: h12 ^= *(uint64 *) ( ptr + 184 )
892 # asm 1: xorq 184(<ptr=int64#4),<h12=int64#6
893 # asm 2: xorq 184(<ptr=%rcx),<h12=%r9
894 xorq 184(%rcx),%r9
895
896 # qhasm: h12 ^= *(uint64 *) ( ptr + 192 )
897 # asm 1: xorq 192(<ptr=int64#4),<h12=int64#6
898 # asm 2: xorq 192(<ptr=%rcx),<h12=%r9
899 xorq 192(%rcx),%r9
900
901 # qhasm: h3 ^= h12
902 # asm 1: XOR <h12=int64#6,<h3=int64#13
903 # asm 2: XOR <h12=%r9,<h3=%r15
904 XOR %r9,%r15
905
906 # qhasm: h0 = h12
907 # asm 1: mov <h12=int64#6,>h0=int64#6
908 # asm 2: mov <h12=%r9,>h0=%r9
909 mov %r9,%r9
910
911 # qhasm: h11 ^= *(uint64 *) ( ptr + 176 )
912 # asm 1: xorq 176(<ptr=int64#4),<h11=int64#8
913 # asm 2: xorq 176(<ptr=%rcx),<h11=%r10
914 xorq 176(%rcx),%r10
915
916 # qhasm: h11 ^= *(uint64 *) ( ptr + 168 )
917 # asm 1: xorq 168(<ptr=int64#4),<h11=int64#8
918 # asm 2: xorq 168(<ptr=%rcx),<h11=%r10
919 xorq 168(%rcx),%r10
920
921 # qhasm: mem64[ input_0 + 88 ] = h11
922 # asm 1: movq <h11=int64#8,88(<input_0=int64#1)
923 # asm 2: movq <h11=%r10,88(<input_0=%rdi)
924 movq %r10,88(%rdi)
925
926 # qhasm: h10 ^= *(uint64 *) ( ptr + 152 )
927 # asm 1: xorq 152(<ptr=int64#4),<h10=int64#2
928 # asm 2: xorq 152(<ptr=%rcx),<h10=%rsi
929 xorq 152(%rcx),%rsi
930
931 # qhasm: h10 ^= *(uint64 *) ( ptr + 160 )
932 # asm 1: xorq 160(<ptr=int64#4),<h10=int64#2
933 # asm 2: xorq 160(<ptr=%rcx),<h10=%rsi
934 xorq 160(%rcx),%rsi
935
936 # qhasm: mem64[ input_0 + 80 ] = h10
937 # asm 1: movq <h10=int64#2,80(<input_0=int64#1)
938 # asm 2: movq <h10=%rsi,80(<input_0=%rdi)
939 movq %rsi,80(%rdi)
940
941 # qhasm: h9 ^= *(uint64 *) ( ptr + 144 )
942 # asm 1: xorq 144(<ptr=int64#4),<h9=int64#5
943 # asm 2: xorq 144(<ptr=%rcx),<h9=%r8
944 xorq 144(%rcx),%r8
945
946 # qhasm: h9 ^= *(uint64 *) ( ptr + 136 )
947 # asm 1: xorq 136(<ptr=int64#4),<h9=int64#5
948 # asm 2: xorq 136(<ptr=%rcx),<h9=%r8
949 xorq 136(%rcx),%r8
950
951 # qhasm: mem64[ input_0 + 72 ] = h9
952 # asm 1: movq <h9=int64#5,72(<input_0=int64#1)
953 # asm 2: movq <h9=%r8,72(<input_0=%rdi)
954 movq %r8,72(%rdi)
955
956 # qhasm: h8 ^= *(uint64 *) ( ptr + 120 )
957 # asm 1: xorq 120(<ptr=int64#4),<h8=int64#7
958 # asm 2: xorq 120(<ptr=%rcx),<h8=%rax
959 xorq 120(%rcx),%rax
960
961 # qhasm: h8 ^= *(uint64 *) ( ptr + 128 )
962 # asm 1: xorq 128(<ptr=int64#4),<h8=int64#7
963 # asm 2: xorq 128(<ptr=%rcx),<h8=%rax
964 xorq 128(%rcx),%rax
965
966 # qhasm: mem64[ input_0 + 64 ] = h8
967 # asm 1: movq <h8=int64#7,64(<input_0=int64#1)
968 # asm 2: movq <h8=%rax,64(<input_0=%rdi)
969 movq %rax,64(%rdi)
970
971 # qhasm: h7 ^= *(uint64 *) ( ptr + 112 )
972 # asm 1: xorq 112(<ptr=int64#4),<h7=int64#9
973 # asm 2: xorq 112(<ptr=%rcx),<h7=%r11
974 xorq 112(%rcx),%r11
975
976 # qhasm: h7 ^= *(uint64 *) ( ptr + 104 )
977 # asm 1: xorq 104(<ptr=int64#4),<h7=int64#9
978 # asm 2: xorq 104(<ptr=%rcx),<h7=%r11
979 xorq 104(%rcx),%r11
980
981 # qhasm: mem64[ input_0 + 56 ] = h7
982 # asm 1: movq <h7=int64#9,56(<input_0=int64#1)
983 # asm 2: movq <h7=%r11,56(<input_0=%rdi)
984 movq %r11,56(%rdi)
985
986 # qhasm: h6 ^= *(uint64 *) ( ptr + 88 )
987 # asm 1: xorq 88(<ptr=int64#4),<h6=int64#10
988 # asm 2: xorq 88(<ptr=%rcx),<h6=%r12
989 xorq 88(%rcx),%r12
990
991 # qhasm: h6 ^= *(uint64 *) ( ptr + 96 )
992 # asm 1: xorq 96(<ptr=int64#4),<h6=int64#10
993 # asm 2: xorq 96(<ptr=%rcx),<h6=%r12
994 xorq 96(%rcx),%r12
995
996 # qhasm: mem64[ input_0 + 48 ] = h6
997 # asm 1: movq <h6=int64#10,48(<input_0=int64#1)
998 # asm 2: movq <h6=%r12,48(<input_0=%rdi)
999 movq %r12,48(%rdi)
1000
1001 # qhasm: h5 ^= *(uint64 *) ( ptr + 80 )
1002 # asm 1: xorq 80(<ptr=int64#4),<h5=int64#11
1003 # asm 2: xorq 80(<ptr=%rcx),<h5=%r13
1004 xorq 80(%rcx),%r13
1005
1006 # qhasm: h5 ^= *(uint64 *) ( ptr + 72 )
1007 # asm 1: xorq 72(<ptr=int64#4),<h5=int64#11
1008 # asm 2: xorq 72(<ptr=%rcx),<h5=%r13
1009 xorq 72(%rcx),%r13
1010
1011 # qhasm: mem64[ input_0 + 40 ] = h5
1012 # asm 1: movq <h5=int64#11,40(<input_0=int64#1)
1013 # asm 2: movq <h5=%r13,40(<input_0=%rdi)
1014 movq %r13,40(%rdi)
1015
1016 # qhasm: h4 ^= *(uint64 *) ( ptr + 56 )
1017 # asm 1: xorq 56(<ptr=int64#4),<h4=int64#12
1018 # asm 2: xorq 56(<ptr=%rcx),<h4=%r14
1019 xorq 56(%rcx),%r14
1020
1021 # qhasm: h4 ^= *(uint64 *) ( ptr + 64 )
1022 # asm 1: xorq 64(<ptr=int64#4),<h4=int64#12
1023 # asm 2: xorq 64(<ptr=%rcx),<h4=%r14
1024 xorq 64(%rcx),%r14
1025
1026 # qhasm: mem64[ input_0 + 32 ] = h4
1027 # asm 1: movq <h4=int64#12,32(<input_0=int64#1)
1028 # asm 2: movq <h4=%r14,32(<input_0=%rdi)
1029 movq %r14,32(%rdi)
1030
1031 # qhasm: h3 ^= *(uint64 *) ( ptr + 48 )
1032 # asm 1: xorq 48(<ptr=int64#4),<h3=int64#13
1033 # asm 2: xorq 48(<ptr=%rcx),<h3=%r15
1034 xorq 48(%rcx),%r15
1035
1036 # qhasm: h3 ^= *(uint64 *) ( ptr + 40 )
1037 # asm 1: xorq 40(<ptr=int64#4),<h3=int64#13
1038 # asm 2: xorq 40(<ptr=%rcx),<h3=%r15
1039 xorq 40(%rcx),%r15
1040
1041 # qhasm: mem64[ input_0 + 24 ] = h3
1042 # asm 1: movq <h3=int64#13,24(<input_0=int64#1)
1043 # asm 2: movq <h3=%r15,24(<input_0=%rdi)
1044 movq %r15,24(%rdi)
1045
1046 # qhasm: h2 ^= *(uint64 *) ( ptr + 24 )
1047 # asm 1: xorq 24(<ptr=int64#4),<h2=int64#14
1048 # asm 2: xorq 24(<ptr=%rcx),<h2=%rbx
1049 xorq 24(%rcx),%rbx
1050
1051 # qhasm: h2 ^= *(uint64 *) ( ptr + 32 )
1052 # asm 1: xorq 32(<ptr=int64#4),<h2=int64#14
1053 # asm 2: xorq 32(<ptr=%rcx),<h2=%rbx
1054 xorq 32(%rcx),%rbx
1055
1056 # qhasm: mem64[ input_0 + 16 ] = h2
1057 # asm 1: movq <h2=int64#14,16(<input_0=int64#1)
1058 # asm 2: movq <h2=%rbx,16(<input_0=%rdi)
1059 movq %rbx,16(%rdi)
1060
1061 # qhasm: h1 ^= *(uint64 *) ( ptr + 16 )
1062 # asm 1: xorq 16(<ptr=int64#4),<h1=int64#3
1063 # asm 2: xorq 16(<ptr=%rcx),<h1=%rdx
1064 xorq 16(%rcx),%rdx
1065
1066 # qhasm: h1 ^= *(uint64 *) ( ptr + 8 )
1067 # asm 1: xorq 8(<ptr=int64#4),<h1=int64#3
1068 # asm 2: xorq 8(<ptr=%rcx),<h1=%rdx
1069 xorq 8(%rcx),%rdx
1070
1071 # qhasm: mem64[ input_0 + 8 ] = h1
1072 # asm 1: movq <h1=int64#3,8(<input_0=int64#1)
1073 # asm 2: movq <h1=%rdx,8(<input_0=%rdi)
1074 movq %rdx,8(%rdi)
1075
1076 # qhasm: h0 ^= *(uint64 *) ( ptr + 0 )
1077 # asm 1: xorq 0(<ptr=int64#4),<h0=int64#6
1078 # asm 2: xorq 0(<ptr=%rcx),<h0=%r9
1079 xorq 0(%rcx),%r9
1080
1081 # qhasm: mem64[ input_0 + 0 ] = h0
1082 # asm 1: movq <h0=int64#6,0(<input_0=int64#1)
1083 # asm 2: movq <h0=%r9,0(<input_0=%rdi)
1084 movq %r9,0(%rdi)
1085
1086 # qhasm: caller_r11 = r11_stack
1087 # asm 1: movq <r11_stack=stack64#1,>caller_r11=int64#9
1088 # asm 2: movq <r11_stack=608(%rsp),>caller_r11=%r11
1089 movq 608(%rsp),%r11
1090
1091 # qhasm: caller_r12 = r12_stack
1092 # asm 1: movq <r12_stack=stack64#2,>caller_r12=int64#10
1093 # asm 2: movq <r12_stack=616(%rsp),>caller_r12=%r12
1094 movq 616(%rsp),%r12
1095
1096 # qhasm: caller_r13 = r13_stack
1097 # asm 1: movq <r13_stack=stack64#3,>caller_r13=int64#11
1098 # asm 2: movq <r13_stack=624(%rsp),>caller_r13=%r13
1099 movq 624(%rsp),%r13
1100
1101 # qhasm: caller_r14 = r14_stack
1102 # asm 1: movq <r14_stack=stack64#4,>caller_r14=int64#12
1103 # asm 2: movq <r14_stack=632(%rsp),>caller_r14=%r14
1104 movq 632(%rsp),%r14
1105
1106 # qhasm: caller_r15 = r15_stack
1107 # asm 1: movq <r15_stack=stack64#5,>caller_r15=int64#13
1108 # asm 2: movq <r15_stack=640(%rsp),>caller_r15=%r15
1109 movq 640(%rsp),%r15
1110
1111 # qhasm: caller_rbx = rbx_stack
1112 # asm 1: movq <rbx_stack=stack64#6,>caller_rbx=int64#14
1113 # asm 2: movq <rbx_stack=648(%rsp),>caller_rbx=%rbx
1114 movq 648(%rsp),%rbx
1115
1116 # qhasm: return
1117 add %r11,%rsp
1118 ret
1119 .section .note.GNU-stack,"",@progbits