Debian Codegraph @

crypto_kem/6688128f/avx/update_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 update_asm
6
7 #include "crypto_asm_hidden.h"
8
9 #define update_asm CRYPTO_SHARED_NAMESPACE(update_asm)
10 #define _update_asm _CRYPTO_SHARED_NAMESPACE(update_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: int64 s0
43
44 # qhasm: int64 s1
45
46 # qhasm: int64 s2
47
48 # qhasm: enter update_asm
49 .p2align 5
50 ASM_HIDDEN _update_asm
51 ASM_HIDDEN update_asm
52 .global _update_asm
53 .global update_asm
54 _update_asm:
55 update_asm:
56 mov %rsp,%r11
57 AND $31,%r11
58 add $0,%r11
59 sub %r11,%rsp
60
61 # qhasm: s2 = input_1
62 # asm 1: mov <input_1=int64#2,>s2=int64#2
63 # asm 2: mov <input_1=%rsi,>s2=%rsi
64 mov %rsi,%rsi
65
66 # qhasm: s0 = mem64[ input_0 + 0 ]
67 # asm 1: movq 0(<input_0=int64#1),>s0=int64#4
68 # asm 2: movq 0(<input_0=%rdi),>s0=%rcx
69 movq 0(%rdi),%rcx
70
71 # qhasm: s1 = mem64[ input_0 + 8 ]
72 # asm 1: movq 8(<input_0=int64#1),>s1=int64#5
73 # asm 2: movq 8(<input_0=%rdi),>s1=%r8
74 movq 8(%rdi),%r8
75
76 # qhasm: s0 = (s1 s0) >> 1
77 # asm 1: shrd $1,<s1=int64#5,<s0=int64#4
78 # asm 2: shrd $1,<s1=%r8,<s0=%rcx
79 shrd $1,%r8,%rcx
80
81 # qhasm: s1 = (s2 s1) >> 1
82 # asm 1: shrd $1,<s2=int64#2,<s1=int64#5
83 # asm 2: shrd $1,<s2=%rsi,<s1=%r8
84 shrd $1,%rsi,%r8
85
86 # qhasm: (uint64) s2 >>= 1
87 # asm 1: shr $1,<s2=int64#2
88 # asm 2: shr $1,<s2=%rsi
89 shr $1,%rsi
90
91 # qhasm: mem64[ input_0 + 0 ] = s0
92 # asm 1: movq <s0=int64#4,0(<input_0=int64#1)
93 # asm 2: movq <s0=%rcx,0(<input_0=%rdi)
94 movq %rcx,0(%rdi)
95
96 # qhasm: mem64[ input_0 + 8 ] = s1
97 # asm 1: movq <s1=int64#5,8(<input_0=int64#1)
98 # asm 2: movq <s1=%r8,8(<input_0=%rdi)
99 movq %r8,8(%rdi)
100
101 # qhasm: input_0 += input_2
102 # asm 1: add <input_2=int64#3,<input_0=int64#1
103 # asm 2: add <input_2=%rdx,<input_0=%rdi
104 add %rdx,%rdi
105
106 # qhasm: s0 = mem64[ input_0 + 0 ]
107 # asm 1: movq 0(<input_0=int64#1),>s0=int64#4
108 # asm 2: movq 0(<input_0=%rdi),>s0=%rcx
109 movq 0(%rdi),%rcx
110
111 # qhasm: s1 = mem64[ input_0 + 8 ]
112 # asm 1: movq 8(<input_0=int64#1),>s1=int64#5
113 # asm 2: movq 8(<input_0=%rdi),>s1=%r8
114 movq 8(%rdi),%r8
115
116 # qhasm: s0 = (s1 s0) >> 1
117 # asm 1: shrd $1,<s1=int64#5,<s0=int64#4
118 # asm 2: shrd $1,<s1=%r8,<s0=%rcx
119 shrd $1,%r8,%rcx
120
121 # qhasm: s1 = (s2 s1) >> 1
122 # asm 1: shrd $1,<s2=int64#2,<s1=int64#5
123 # asm 2: shrd $1,<s2=%rsi,<s1=%r8
124 shrd $1,%rsi,%r8
125
126 # qhasm: (uint64) s2 >>= 1
127 # asm 1: shr $1,<s2=int64#2
128 # asm 2: shr $1,<s2=%rsi
129 shr $1,%rsi
130
131 # qhasm: mem64[ input_0 + 0 ] = s0
132 # asm 1: movq <s0=int64#4,0(<input_0=int64#1)
133 # asm 2: movq <s0=%rcx,0(<input_0=%rdi)
134 movq %rcx,0(%rdi)
135
136 # qhasm: mem64[ input_0 + 8 ] = s1
137 # asm 1: movq <s1=int64#5,8(<input_0=int64#1)
138 # asm 2: movq <s1=%r8,8(<input_0=%rdi)
139 movq %r8,8(%rdi)
140
141 # qhasm: input_0 += input_2
142 # asm 1: add <input_2=int64#3,<input_0=int64#1
143 # asm 2: add <input_2=%rdx,<input_0=%rdi
144 add %rdx,%rdi
145
146 # qhasm: s0 = mem64[ input_0 + 0 ]
147 # asm 1: movq 0(<input_0=int64#1),>s0=int64#4
148 # asm 2: movq 0(<input_0=%rdi),>s0=%rcx
149 movq 0(%rdi),%rcx
150
151 # qhasm: s1 = mem64[ input_0 + 8 ]
152 # asm 1: movq 8(<input_0=int64#1),>s1=int64#5
153 # asm 2: movq 8(<input_0=%rdi),>s1=%r8
154 movq 8(%rdi),%r8
155
156 # qhasm: s0 = (s1 s0) >> 1
157 # asm 1: shrd $1,<s1=int64#5,<s0=int64#4
158 # asm 2: shrd $1,<s1=%r8,<s0=%rcx
159 shrd $1,%r8,%rcx
160
161 # qhasm: s1 = (s2 s1) >> 1
162 # asm 1: shrd $1,<s2=int64#2,<s1=int64#5
163 # asm 2: shrd $1,<s2=%rsi,<s1=%r8
164 shrd $1,%rsi,%r8
165
166 # qhasm: (uint64) s2 >>= 1
167 # asm 1: shr $1,<s2=int64#2
168 # asm 2: shr $1,<s2=%rsi
169 shr $1,%rsi
170
171 # qhasm: mem64[ input_0 + 0 ] = s0
172 # asm 1: movq <s0=int64#4,0(<input_0=int64#1)
173 # asm 2: movq <s0=%rcx,0(<input_0=%rdi)
174 movq %rcx,0(%rdi)
175
176 # qhasm: mem64[ input_0 + 8 ] = s1
177 # asm 1: movq <s1=int64#5,8(<input_0=int64#1)
178 # asm 2: movq <s1=%r8,8(<input_0=%rdi)
179 movq %r8,8(%rdi)
180
181 # qhasm: input_0 += input_2
182 # asm 1: add <input_2=int64#3,<input_0=int64#1
183 # asm 2: add <input_2=%rdx,<input_0=%rdi
184 add %rdx,%rdi
185
186 # qhasm: s0 = mem64[ input_0 + 0 ]
187 # asm 1: movq 0(<input_0=int64#1),>s0=int64#4
188 # asm 2: movq 0(<input_0=%rdi),>s0=%rcx
189 movq 0(%rdi),%rcx
190
191 # qhasm: s1 = mem64[ input_0 + 8 ]
192 # asm 1: movq 8(<input_0=int64#1),>s1=int64#5
193 # asm 2: movq 8(<input_0=%rdi),>s1=%r8
194 movq 8(%rdi),%r8
195
196 # qhasm: s0 = (s1 s0) >> 1
197 # asm 1: shrd $1,<s1=int64#5,<s0=int64#4
198 # asm 2: shrd $1,<s1=%r8,<s0=%rcx
199 shrd $1,%r8,%rcx
200
201 # qhasm: s1 = (s2 s1) >> 1
202 # asm 1: shrd $1,<s2=int64#2,<s1=int64#5
203 # asm 2: shrd $1,<s2=%rsi,<s1=%r8
204 shrd $1,%rsi,%r8
205
206 # qhasm: (uint64) s2 >>= 1
207 # asm 1: shr $1,<s2=int64#2
208 # asm 2: shr $1,<s2=%rsi
209 shr $1,%rsi
210
211 # qhasm: mem64[ input_0 + 0 ] = s0
212 # asm 1: movq <s0=int64#4,0(<input_0=int64#1)
213 # asm 2: movq <s0=%rcx,0(<input_0=%rdi)
214 movq %rcx,0(%rdi)
215
216 # qhasm: mem64[ input_0 + 8 ] = s1
217 # asm 1: movq <s1=int64#5,8(<input_0=int64#1)
218 # asm 2: movq <s1=%r8,8(<input_0=%rdi)
219 movq %r8,8(%rdi)
220
221 # qhasm: input_0 += input_2
222 # asm 1: add <input_2=int64#3,<input_0=int64#1
223 # asm 2: add <input_2=%rdx,<input_0=%rdi
224 add %rdx,%rdi
225
226 # qhasm: s0 = mem64[ input_0 + 0 ]
227 # asm 1: movq 0(<input_0=int64#1),>s0=int64#4
228 # asm 2: movq 0(<input_0=%rdi),>s0=%rcx
229 movq 0(%rdi),%rcx
230
231 # qhasm: s1 = mem64[ input_0 + 8 ]
232 # asm 1: movq 8(<input_0=int64#1),>s1=int64#5
233 # asm 2: movq 8(<input_0=%rdi),>s1=%r8
234 movq 8(%rdi),%r8
235
236 # qhasm: s0 = (s1 s0) >> 1
237 # asm 1: shrd $1,<s1=int64#5,<s0=int64#4
238 # asm 2: shrd $1,<s1=%r8,<s0=%rcx
239 shrd $1,%r8,%rcx
240
241 # qhasm: s1 = (s2 s1) >> 1
242 # asm 1: shrd $1,<s2=int64#2,<s1=int64#5
243 # asm 2: shrd $1,<s2=%rsi,<s1=%r8
244 shrd $1,%rsi,%r8
245
246 # qhasm: (uint64) s2 >>= 1
247 # asm 1: shr $1,<s2=int64#2
248 # asm 2: shr $1,<s2=%rsi
249 shr $1,%rsi
250
251 # qhasm: mem64[ input_0 + 0 ] = s0
252 # asm 1: movq <s0=int64#4,0(<input_0=int64#1)
253 # asm 2: movq <s0=%rcx,0(<input_0=%rdi)
254 movq %rcx,0(%rdi)
255
256 # qhasm: mem64[ input_0 + 8 ] = s1
257 # asm 1: movq <s1=int64#5,8(<input_0=int64#1)
258 # asm 2: movq <s1=%r8,8(<input_0=%rdi)
259 movq %r8,8(%rdi)
260
261 # qhasm: input_0 += input_2
262 # asm 1: add <input_2=int64#3,<input_0=int64#1
263 # asm 2: add <input_2=%rdx,<input_0=%rdi
264 add %rdx,%rdi
265
266 # qhasm: s0 = mem64[ input_0 + 0 ]
267 # asm 1: movq 0(<input_0=int64#1),>s0=int64#4
268 # asm 2: movq 0(<input_0=%rdi),>s0=%rcx
269 movq 0(%rdi),%rcx
270
271 # qhasm: s1 = mem64[ input_0 + 8 ]
272 # asm 1: movq 8(<input_0=int64#1),>s1=int64#5
273 # asm 2: movq 8(<input_0=%rdi),>s1=%r8
274 movq 8(%rdi),%r8
275
276 # qhasm: s0 = (s1 s0) >> 1
277 # asm 1: shrd $1,<s1=int64#5,<s0=int64#4
278 # asm 2: shrd $1,<s1=%r8,<s0=%rcx
279 shrd $1,%r8,%rcx
280
281 # qhasm: s1 = (s2 s1) >> 1
282 # asm 1: shrd $1,<s2=int64#2,<s1=int64#5
283 # asm 2: shrd $1,<s2=%rsi,<s1=%r8
284 shrd $1,%rsi,%r8
285
286 # qhasm: (uint64) s2 >>= 1
287 # asm 1: shr $1,<s2=int64#2
288 # asm 2: shr $1,<s2=%rsi
289 shr $1,%rsi
290
291 # qhasm: mem64[ input_0 + 0 ] = s0
292 # asm 1: movq <s0=int64#4,0(<input_0=int64#1)
293 # asm 2: movq <s0=%rcx,0(<input_0=%rdi)
294 movq %rcx,0(%rdi)
295
296 # qhasm: mem64[ input_0 + 8 ] = s1
297 # asm 1: movq <s1=int64#5,8(<input_0=int64#1)
298 # asm 2: movq <s1=%r8,8(<input_0=%rdi)
299 movq %r8,8(%rdi)
300
301 # qhasm: input_0 += input_2
302 # asm 1: add <input_2=int64#3,<input_0=int64#1
303 # asm 2: add <input_2=%rdx,<input_0=%rdi
304 add %rdx,%rdi
305
306 # qhasm: s0 = mem64[ input_0 + 0 ]
307 # asm 1: movq 0(<input_0=int64#1),>s0=int64#4
308 # asm 2: movq 0(<input_0=%rdi),>s0=%rcx
309 movq 0(%rdi),%rcx
310
311 # qhasm: s1 = mem64[ input_0 + 8 ]
312 # asm 1: movq 8(<input_0=int64#1),>s1=int64#5
313 # asm 2: movq 8(<input_0=%rdi),>s1=%r8
314 movq 8(%rdi),%r8
315
316 # qhasm: s0 = (s1 s0) >> 1
317 # asm 1: shrd $1,<s1=int64#5,<s0=int64#4
318 # asm 2: shrd $1,<s1=%r8,<s0=%rcx
319 shrd $1,%r8,%rcx
320
321 # qhasm: s1 = (s2 s1) >> 1
322 # asm 1: shrd $1,<s2=int64#2,<s1=int64#5
323 # asm 2: shrd $1,<s2=%rsi,<s1=%r8
324 shrd $1,%rsi,%r8
325
326 # qhasm: (uint64) s2 >>= 1
327 # asm 1: shr $1,<s2=int64#2
328 # asm 2: shr $1,<s2=%rsi
329 shr $1,%rsi
330
331 # qhasm: mem64[ input_0 + 0 ] = s0
332 # asm 1: movq <s0=int64#4,0(<input_0=int64#1)
333 # asm 2: movq <s0=%rcx,0(<input_0=%rdi)
334 movq %rcx,0(%rdi)
335
336 # qhasm: mem64[ input_0 + 8 ] = s1
337 # asm 1: movq <s1=int64#5,8(<input_0=int64#1)
338 # asm 2: movq <s1=%r8,8(<input_0=%rdi)
339 movq %r8,8(%rdi)
340
341 # qhasm: input_0 += input_2
342 # asm 1: add <input_2=int64#3,<input_0=int64#1
343 # asm 2: add <input_2=%rdx,<input_0=%rdi
344 add %rdx,%rdi
345
346 # qhasm: s0 = mem64[ input_0 + 0 ]
347 # asm 1: movq 0(<input_0=int64#1),>s0=int64#4
348 # asm 2: movq 0(<input_0=%rdi),>s0=%rcx
349 movq 0(%rdi),%rcx
350
351 # qhasm: s1 = mem64[ input_0 + 8 ]
352 # asm 1: movq 8(<input_0=int64#1),>s1=int64#5
353 # asm 2: movq 8(<input_0=%rdi),>s1=%r8
354 movq 8(%rdi),%r8
355
356 # qhasm: s0 = (s1 s0) >> 1
357 # asm 1: shrd $1,<s1=int64#5,<s0=int64#4
358 # asm 2: shrd $1,<s1=%r8,<s0=%rcx
359 shrd $1,%r8,%rcx
360
361 # qhasm: s1 = (s2 s1) >> 1
362 # asm 1: shrd $1,<s2=int64#2,<s1=int64#5
363 # asm 2: shrd $1,<s2=%rsi,<s1=%r8
364 shrd $1,%rsi,%r8
365
366 # qhasm: (uint64) s2 >>= 1
367 # asm 1: shr $1,<s2=int64#2
368 # asm 2: shr $1,<s2=%rsi
369 shr $1,%rsi
370
371 # qhasm: mem64[ input_0 + 0 ] = s0
372 # asm 1: movq <s0=int64#4,0(<input_0=int64#1)
373 # asm 2: movq <s0=%rcx,0(<input_0=%rdi)
374 movq %rcx,0(%rdi)
375
376 # qhasm: mem64[ input_0 + 8 ] = s1
377 # asm 1: movq <s1=int64#5,8(<input_0=int64#1)
378 # asm 2: movq <s1=%r8,8(<input_0=%rdi)
379 movq %r8,8(%rdi)
380
381 # qhasm: input_0 += input_2
382 # asm 1: add <input_2=int64#3,<input_0=int64#1
383 # asm 2: add <input_2=%rdx,<input_0=%rdi
384 add %rdx,%rdi
385
386 # qhasm: s0 = mem64[ input_0 + 0 ]
387 # asm 1: movq 0(<input_0=int64#1),>s0=int64#4
388 # asm 2: movq 0(<input_0=%rdi),>s0=%rcx
389 movq 0(%rdi),%rcx
390
391 # qhasm: s1 = mem64[ input_0 + 8 ]
392 # asm 1: movq 8(<input_0=int64#1),>s1=int64#5
393 # asm 2: movq 8(<input_0=%rdi),>s1=%r8
394 movq 8(%rdi),%r8
395
396 # qhasm: s0 = (s1 s0) >> 1
397 # asm 1: shrd $1,<s1=int64#5,<s0=int64#4
398 # asm 2: shrd $1,<s1=%r8,<s0=%rcx
399 shrd $1,%r8,%rcx
400
401 # qhasm: s1 = (s2 s1) >> 1
402 # asm 1: shrd $1,<s2=int64#2,<s1=int64#5
403 # asm 2: shrd $1,<s2=%rsi,<s1=%r8
404 shrd $1,%rsi,%r8
405
406 # qhasm: (uint64) s2 >>= 1
407 # asm 1: shr $1,<s2=int64#2
408 # asm 2: shr $1,<s2=%rsi
409 shr $1,%rsi
410
411 # qhasm: mem64[ input_0 + 0 ] = s0
412 # asm 1: movq <s0=int64#4,0(<input_0=int64#1)
413 # asm 2: movq <s0=%rcx,0(<input_0=%rdi)
414 movq %rcx,0(%rdi)
415
416 # qhasm: mem64[ input_0 + 8 ] = s1
417 # asm 1: movq <s1=int64#5,8(<input_0=int64#1)
418 # asm 2: movq <s1=%r8,8(<input_0=%rdi)
419 movq %r8,8(%rdi)
420
421 # qhasm: input_0 += input_2
422 # asm 1: add <input_2=int64#3,<input_0=int64#1
423 # asm 2: add <input_2=%rdx,<input_0=%rdi
424 add %rdx,%rdi
425
426 # qhasm: s0 = mem64[ input_0 + 0 ]
427 # asm 1: movq 0(<input_0=int64#1),>s0=int64#4
428 # asm 2: movq 0(<input_0=%rdi),>s0=%rcx
429 movq 0(%rdi),%rcx
430
431 # qhasm: s1 = mem64[ input_0 + 8 ]
432 # asm 1: movq 8(<input_0=int64#1),>s1=int64#5
433 # asm 2: movq 8(<input_0=%rdi),>s1=%r8
434 movq 8(%rdi),%r8
435
436 # qhasm: s0 = (s1 s0) >> 1
437 # asm 1: shrd $1,<s1=int64#5,<s0=int64#4
438 # asm 2: shrd $1,<s1=%r8,<s0=%rcx
439 shrd $1,%r8,%rcx
440
441 # qhasm: s1 = (s2 s1) >> 1
442 # asm 1: shrd $1,<s2=int64#2,<s1=int64#5
443 # asm 2: shrd $1,<s2=%rsi,<s1=%r8
444 shrd $1,%rsi,%r8
445
446 # qhasm: (uint64) s2 >>= 1
447 # asm 1: shr $1,<s2=int64#2
448 # asm 2: shr $1,<s2=%rsi
449 shr $1,%rsi
450
451 # qhasm: mem64[ input_0 + 0 ] = s0
452 # asm 1: movq <s0=int64#4,0(<input_0=int64#1)
453 # asm 2: movq <s0=%rcx,0(<input_0=%rdi)
454 movq %rcx,0(%rdi)
455
456 # qhasm: mem64[ input_0 + 8 ] = s1
457 # asm 1: movq <s1=int64#5,8(<input_0=int64#1)
458 # asm 2: movq <s1=%r8,8(<input_0=%rdi)
459 movq %r8,8(%rdi)
460
461 # qhasm: input_0 += input_2
462 # asm 1: add <input_2=int64#3,<input_0=int64#1
463 # asm 2: add <input_2=%rdx,<input_0=%rdi
464 add %rdx,%rdi
465
466 # qhasm: s0 = mem64[ input_0 + 0 ]
467 # asm 1: movq 0(<input_0=int64#1),>s0=int64#4
468 # asm 2: movq 0(<input_0=%rdi),>s0=%rcx
469 movq 0(%rdi),%rcx
470
471 # qhasm: s1 = mem64[ input_0 + 8 ]
472 # asm 1: movq 8(<input_0=int64#1),>s1=int64#5
473 # asm 2: movq 8(<input_0=%rdi),>s1=%r8
474 movq 8(%rdi),%r8
475
476 # qhasm: s0 = (s1 s0) >> 1
477 # asm 1: shrd $1,<s1=int64#5,<s0=int64#4
478 # asm 2: shrd $1,<s1=%r8,<s0=%rcx
479 shrd $1,%r8,%rcx
480
481 # qhasm: s1 = (s2 s1) >> 1
482 # asm 1: shrd $1,<s2=int64#2,<s1=int64#5
483 # asm 2: shrd $1,<s2=%rsi,<s1=%r8
484 shrd $1,%rsi,%r8
485
486 # qhasm: (uint64) s2 >>= 1
487 # asm 1: shr $1,<s2=int64#2
488 # asm 2: shr $1,<s2=%rsi
489 shr $1,%rsi
490
491 # qhasm: mem64[ input_0 + 0 ] = s0
492 # asm 1: movq <s0=int64#4,0(<input_0=int64#1)
493 # asm 2: movq <s0=%rcx,0(<input_0=%rdi)
494 movq %rcx,0(%rdi)
495
496 # qhasm: mem64[ input_0 + 8 ] = s1
497 # asm 1: movq <s1=int64#5,8(<input_0=int64#1)
498 # asm 2: movq <s1=%r8,8(<input_0=%rdi)
499 movq %r8,8(%rdi)
500
501 # qhasm: input_0 += input_2
502 # asm 1: add <input_2=int64#3,<input_0=int64#1
503 # asm 2: add <input_2=%rdx,<input_0=%rdi
504 add %rdx,%rdi
505
506 # qhasm: s0 = mem64[ input_0 + 0 ]
507 # asm 1: movq 0(<input_0=int64#1),>s0=int64#4
508 # asm 2: movq 0(<input_0=%rdi),>s0=%rcx
509 movq 0(%rdi),%rcx
510
511 # qhasm: s1 = mem64[ input_0 + 8 ]
512 # asm 1: movq 8(<input_0=int64#1),>s1=int64#5
513 # asm 2: movq 8(<input_0=%rdi),>s1=%r8
514 movq 8(%rdi),%r8
515
516 # qhasm: s0 = (s1 s0) >> 1
517 # asm 1: shrd $1,<s1=int64#5,<s0=int64#4
518 # asm 2: shrd $1,<s1=%r8,<s0=%rcx
519 shrd $1,%r8,%rcx
520
521 # qhasm: s1 = (s2 s1) >> 1
522 # asm 1: shrd $1,<s2=int64#2,<s1=int64#5
523 # asm 2: shrd $1,<s2=%rsi,<s1=%r8
524 shrd $1,%rsi,%r8
525
526 # qhasm: (uint64) s2 >>= 1
527 # asm 1: shr $1,<s2=int64#2
528 # asm 2: shr $1,<s2=%rsi
529 shr $1,%rsi
530
531 # qhasm: mem64[ input_0 + 0 ] = s0
532 # asm 1: movq <s0=int64#4,0(<input_0=int64#1)
533 # asm 2: movq <s0=%rcx,0(<input_0=%rdi)
534 movq %rcx,0(%rdi)
535
536 # qhasm: mem64[ input_0 + 8 ] = s1
537 # asm 1: movq <s1=int64#5,8(<input_0=int64#1)
538 # asm 2: movq <s1=%r8,8(<input_0=%rdi)
539 movq %r8,8(%rdi)
540
541 # qhasm: input_0 += input_2
542 # asm 1: add <input_2=int64#3,<input_0=int64#1
543 # asm 2: add <input_2=%rdx,<input_0=%rdi
544 add %rdx,%rdi
545
546 # qhasm: s0 = mem64[ input_0 + 0 ]
547 # asm 1: movq 0(<input_0=int64#1),>s0=int64#4
548 # asm 2: movq 0(<input_0=%rdi),>s0=%rcx
549 movq 0(%rdi),%rcx
550
551 # qhasm: s1 = mem64[ input_0 + 8 ]
552 # asm 1: movq 8(<input_0=int64#1),>s1=int64#5
553 # asm 2: movq 8(<input_0=%rdi),>s1=%r8
554 movq 8(%rdi),%r8
555
556 # qhasm: s0 = (s1 s0) >> 1
557 # asm 1: shrd $1,<s1=int64#5,<s0=int64#4
558 # asm 2: shrd $1,<s1=%r8,<s0=%rcx
559 shrd $1,%r8,%rcx
560
561 # qhasm: s1 = (s2 s1) >> 1
562 # asm 1: shrd $1,<s2=int64#2,<s1=int64#5
563 # asm 2: shrd $1,<s2=%rsi,<s1=%r8
564 shrd $1,%rsi,%r8
565
566 # qhasm: (uint64) s2 >>= 1
567 # asm 1: shr $1,<s2=int64#2
568 # asm 2: shr $1,<s2=%rsi
569 shr $1,%rsi
570
571 # qhasm: mem64[ input_0 + 0 ] = s0
572 # asm 1: movq <s0=int64#4,0(<input_0=int64#1)
573 # asm 2: movq <s0=%rcx,0(<input_0=%rdi)
574 movq %rcx,0(%rdi)
575
576 # qhasm: mem64[ input_0 + 8 ] = s1
577 # asm 1: movq <s1=int64#5,8(<input_0=int64#1)
578 # asm 2: movq <s1=%r8,8(<input_0=%rdi)
579 movq %r8,8(%rdi)
580
581 # qhasm: input_0 += input_2
582 # asm 1: add <input_2=int64#3,<input_0=int64#1
583 # asm 2: add <input_2=%rdx,<input_0=%rdi
584 add %rdx,%rdi
585
586 # qhasm: return
587 add %r11,%rsp
588 ret
589 .section .note.GNU-stack,"",@progbits