forked from pascholl/SimpleOT
-
Notifications
You must be signed in to change notification settings - Fork 4
/
Copy pathfe25519_freeze.S
434 lines (329 loc) · 10 KB
/
fe25519_freeze.S
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
# qhasm: int64 rp
# qhasm: input rp
# qhasm: int64 r0
# qhasm: int64 r1
# qhasm: int64 r2
# qhasm: int64 r3
# qhasm: int64 r4
# qhasm: int64 t
# qhasm: int64 loop
# qhasm: int64 two51minus1
# qhasm: int64 two51minus19
# qhasm: int64 caller1
# qhasm: int64 caller2
# qhasm: int64 caller3
# qhasm: int64 caller4
# qhasm: int64 caller5
# qhasm: int64 caller6
# qhasm: int64 caller7
# qhasm: caller caller1
# qhasm: caller caller2
# qhasm: caller caller3
# qhasm: caller caller4
# qhasm: caller caller5
# qhasm: caller caller6
# qhasm: caller caller7
# qhasm: stack64 caller1_stack
# qhasm: stack64 caller2_stack
# qhasm: stack64 caller3_stack
# qhasm: stack64 caller4_stack
# qhasm: stack64 caller5_stack
# qhasm: stack64 caller6_stack
# qhasm: stack64 caller7_stack
# qhasm: enter fe25519_freeze
.text
.p2align 5
.globl _fe25519_freeze
.globl fe25519_freeze
_fe25519_freeze:
fe25519_freeze:
mov %rsp,%r11
and $31,%r11
add $64,%r11
sub %r11,%rsp
# qhasm: caller1_stack = caller1
# asm 1: movq <caller1=int64#9,>caller1_stack=stack64#1
# asm 2: movq <caller1=%r11,>caller1_stack=0(%rsp)
movq %r11,0(%rsp)
# qhasm: caller2_stack = caller2
# asm 1: movq <caller2=int64#10,>caller2_stack=stack64#2
# asm 2: movq <caller2=%r12,>caller2_stack=8(%rsp)
movq %r12,8(%rsp)
# qhasm: caller3_stack = caller3
# asm 1: movq <caller3=int64#11,>caller3_stack=stack64#3
# asm 2: movq <caller3=%r13,>caller3_stack=16(%rsp)
movq %r13,16(%rsp)
# qhasm: caller4_stack = caller4
# asm 1: movq <caller4=int64#12,>caller4_stack=stack64#4
# asm 2: movq <caller4=%r14,>caller4_stack=24(%rsp)
movq %r14,24(%rsp)
# qhasm: caller5_stack = caller5
# asm 1: movq <caller5=int64#13,>caller5_stack=stack64#5
# asm 2: movq <caller5=%r15,>caller5_stack=32(%rsp)
movq %r15,32(%rsp)
# qhasm: caller6_stack = caller6
# asm 1: movq <caller6=int64#14,>caller6_stack=stack64#6
# asm 2: movq <caller6=%rbx,>caller6_stack=40(%rsp)
movq %rbx,40(%rsp)
# qhasm: caller7_stack = caller7
# asm 1: movq <caller7=int64#15,>caller7_stack=stack64#7
# asm 2: movq <caller7=%rbp,>caller7_stack=48(%rsp)
movq %rbp,48(%rsp)
# qhasm: r0 = *(uint64 *) (rp + 0)
# asm 1: movq 0(<rp=int64#1),>r0=int64#2
# asm 2: movq 0(<rp=%rdi),>r0=%rsi
movq 0(%rdi),%rsi
# qhasm: r1 = *(uint64 *) (rp + 8)
# asm 1: movq 8(<rp=int64#1),>r1=int64#3
# asm 2: movq 8(<rp=%rdi),>r1=%rdx
movq 8(%rdi),%rdx
# qhasm: r2 = *(uint64 *) (rp + 16)
# asm 1: movq 16(<rp=int64#1),>r2=int64#4
# asm 2: movq 16(<rp=%rdi),>r2=%rcx
movq 16(%rdi),%rcx
# qhasm: r3 = *(uint64 *) (rp + 24)
# asm 1: movq 24(<rp=int64#1),>r3=int64#5
# asm 2: movq 24(<rp=%rdi),>r3=%r8
movq 24(%rdi),%r8
# qhasm: r4 = *(uint64 *) (rp + 32)
# asm 1: movq 32(<rp=int64#1),>r4=int64#6
# asm 2: movq 32(<rp=%rdi),>r4=%r9
movq 32(%rdi),%r9
# qhasm: two51minus1 = *(uint64 *) &CONST_REDMASK51
# asm 1: movq CONST_REDMASK51,>two51minus1=int64#7
# asm 2: movq CONST_REDMASK51,>two51minus1=%rax
movq CONST_REDMASK51(%rip),%rax
# qhasm: two51minus19 = two51minus1
# asm 1: mov <two51minus1=int64#7,>two51minus19=int64#8
# asm 2: mov <two51minus1=%rax,>two51minus19=%r10
mov %rax,%r10
# qhasm: two51minus19 -= 18
# asm 1: sub $18,<two51minus19=int64#8
# asm 2: sub $18,<two51minus19=%r10
sub $18,%r10
# qhasm: loop = 3
# asm 1: mov $3,>loop=int64#9
# asm 2: mov $3,>loop=%r11
mov $3,%r11
# qhasm: reduceloop:
._reduceloop:
# qhasm: t = r0
# asm 1: mov <r0=int64#2,>t=int64#10
# asm 2: mov <r0=%rsi,>t=%r12
mov %rsi,%r12
# qhasm: (uint64) t >>= 51
# asm 1: shr $51,<t=int64#10
# asm 2: shr $51,<t=%r12
shr $51,%r12
# qhasm: r0 &= two51minus1
# asm 1: and <two51minus1=int64#7,<r0=int64#2
# asm 2: and <two51minus1=%rax,<r0=%rsi
and %rax,%rsi
# qhasm: r1 += t
# asm 1: add <t=int64#10,<r1=int64#3
# asm 2: add <t=%r12,<r1=%rdx
add %r12,%rdx
# qhasm: t = r1
# asm 1: mov <r1=int64#3,>t=int64#10
# asm 2: mov <r1=%rdx,>t=%r12
mov %rdx,%r12
# qhasm: (uint64) t >>= 51
# asm 1: shr $51,<t=int64#10
# asm 2: shr $51,<t=%r12
shr $51,%r12
# qhasm: r1 &= two51minus1
# asm 1: and <two51minus1=int64#7,<r1=int64#3
# asm 2: and <two51minus1=%rax,<r1=%rdx
and %rax,%rdx
# qhasm: r2 += t
# asm 1: add <t=int64#10,<r2=int64#4
# asm 2: add <t=%r12,<r2=%rcx
add %r12,%rcx
# qhasm: t = r2
# asm 1: mov <r2=int64#4,>t=int64#10
# asm 2: mov <r2=%rcx,>t=%r12
mov %rcx,%r12
# qhasm: (uint64) t >>= 51
# asm 1: shr $51,<t=int64#10
# asm 2: shr $51,<t=%r12
shr $51,%r12
# qhasm: r2 &= two51minus1
# asm 1: and <two51minus1=int64#7,<r2=int64#4
# asm 2: and <two51minus1=%rax,<r2=%rcx
and %rax,%rcx
# qhasm: r3 += t
# asm 1: add <t=int64#10,<r3=int64#5
# asm 2: add <t=%r12,<r3=%r8
add %r12,%r8
# qhasm: t = r3
# asm 1: mov <r3=int64#5,>t=int64#10
# asm 2: mov <r3=%r8,>t=%r12
mov %r8,%r12
# qhasm: (uint64) t >>= 51
# asm 1: shr $51,<t=int64#10
# asm 2: shr $51,<t=%r12
shr $51,%r12
# qhasm: r3 &= two51minus1
# asm 1: and <two51minus1=int64#7,<r3=int64#5
# asm 2: and <two51minus1=%rax,<r3=%r8
and %rax,%r8
# qhasm: r4 += t
# asm 1: add <t=int64#10,<r4=int64#6
# asm 2: add <t=%r12,<r4=%r9
add %r12,%r9
# qhasm: t = r4
# asm 1: mov <r4=int64#6,>t=int64#10
# asm 2: mov <r4=%r9,>t=%r12
mov %r9,%r12
# qhasm: (uint64) t >>= 51
# asm 1: shr $51,<t=int64#10
# asm 2: shr $51,<t=%r12
shr $51,%r12
# qhasm: r4 &= two51minus1
# asm 1: and <two51minus1=int64#7,<r4=int64#6
# asm 2: and <two51minus1=%rax,<r4=%r9
and %rax,%r9
# qhasm: t *= 19
# asm 1: imulq $19,<t=int64#10,>t=int64#10
# asm 2: imulq $19,<t=%r12,>t=%r12
imulq $19,%r12,%r12
# qhasm: r0 += t
# asm 1: add <t=int64#10,<r0=int64#2
# asm 2: add <t=%r12,<r0=%rsi
add %r12,%rsi
# qhasm: unsigned>? loop -= 1
# asm 1: sub $1,<loop=int64#9
# asm 2: sub $1,<loop=%r11
sub $1,%r11
# comment:fp stack unchanged by jump
# qhasm: goto reduceloop if unsigned>
ja ._reduceloop
# qhasm: t = 1
# asm 1: mov $1,>t=int64#10
# asm 2: mov $1,>t=%r12
mov $1,%r12
# qhasm: signed<? r0 - two51minus19
# asm 1: cmp <two51minus19=int64#8,<r0=int64#2
# asm 2: cmp <two51minus19=%r10,<r0=%rsi
cmp %r10,%rsi
# qhasm: t = loop if signed<
# asm 1: cmovl <loop=int64#9,<t=int64#10
# asm 2: cmovl <loop=%r11,<t=%r12
cmovl %r11,%r12
# qhasm: =? r1 - two51minus1
# asm 1: cmp <two51minus1=int64#7,<r1=int64#3
# asm 2: cmp <two51minus1=%rax,<r1=%rdx
cmp %rax,%rdx
# qhasm: t = loop if !=
# asm 1: cmovne <loop=int64#9,<t=int64#10
# asm 2: cmovne <loop=%r11,<t=%r12
cmovne %r11,%r12
# qhasm: =? r2 - two51minus1
# asm 1: cmp <two51minus1=int64#7,<r2=int64#4
# asm 2: cmp <two51minus1=%rax,<r2=%rcx
cmp %rax,%rcx
# qhasm: t = loop if !=
# asm 1: cmovne <loop=int64#9,<t=int64#10
# asm 2: cmovne <loop=%r11,<t=%r12
cmovne %r11,%r12
# qhasm: =? r3 - two51minus1
# asm 1: cmp <two51minus1=int64#7,<r3=int64#5
# asm 2: cmp <two51minus1=%rax,<r3=%r8
cmp %rax,%r8
# qhasm: t = loop if !=
# asm 1: cmovne <loop=int64#9,<t=int64#10
# asm 2: cmovne <loop=%r11,<t=%r12
cmovne %r11,%r12
# qhasm: =? r4 - two51minus1
# asm 1: cmp <two51minus1=int64#7,<r4=int64#6
# asm 2: cmp <two51minus1=%rax,<r4=%r9
cmp %rax,%r9
# qhasm: t = loop if !=
# asm 1: cmovne <loop=int64#9,<t=int64#10
# asm 2: cmovne <loop=%r11,<t=%r12
cmovne %r11,%r12
# qhasm: t = -t
# asm 1: neg <t=int64#10
# asm 2: neg <t=%r12
neg %r12
# qhasm: two51minus1 &= t
# asm 1: and <t=int64#10,<two51minus1=int64#7
# asm 2: and <t=%r12,<two51minus1=%rax
and %r12,%rax
# qhasm: two51minus19 &= t
# asm 1: and <t=int64#10,<two51minus19=int64#8
# asm 2: and <t=%r12,<two51minus19=%r10
and %r12,%r10
# qhasm: r0 -= two51minus19
# asm 1: sub <two51minus19=int64#8,<r0=int64#2
# asm 2: sub <two51minus19=%r10,<r0=%rsi
sub %r10,%rsi
# qhasm: r1 -= two51minus1
# asm 1: sub <two51minus1=int64#7,<r1=int64#3
# asm 2: sub <two51minus1=%rax,<r1=%rdx
sub %rax,%rdx
# qhasm: r2 -= two51minus1
# asm 1: sub <two51minus1=int64#7,<r2=int64#4
# asm 2: sub <two51minus1=%rax,<r2=%rcx
sub %rax,%rcx
# qhasm: r3 -= two51minus1
# asm 1: sub <two51minus1=int64#7,<r3=int64#5
# asm 2: sub <two51minus1=%rax,<r3=%r8
sub %rax,%r8
# qhasm: r4 -= two51minus1
# asm 1: sub <two51minus1=int64#7,<r4=int64#6
# asm 2: sub <two51minus1=%rax,<r4=%r9
sub %rax,%r9
# qhasm: *(uint64 *)(rp + 0) = r0
# asm 1: movq <r0=int64#2,0(<rp=int64#1)
# asm 2: movq <r0=%rsi,0(<rp=%rdi)
movq %rsi,0(%rdi)
# qhasm: *(uint64 *)(rp + 8) = r1
# asm 1: movq <r1=int64#3,8(<rp=int64#1)
# asm 2: movq <r1=%rdx,8(<rp=%rdi)
movq %rdx,8(%rdi)
# qhasm: *(uint64 *)(rp + 16) = r2
# asm 1: movq <r2=int64#4,16(<rp=int64#1)
# asm 2: movq <r2=%rcx,16(<rp=%rdi)
movq %rcx,16(%rdi)
# qhasm: *(uint64 *)(rp + 24) = r3
# asm 1: movq <r3=int64#5,24(<rp=int64#1)
# asm 2: movq <r3=%r8,24(<rp=%rdi)
movq %r8,24(%rdi)
# qhasm: *(uint64 *)(rp + 32) = r4
# asm 1: movq <r4=int64#6,32(<rp=int64#1)
# asm 2: movq <r4=%r9,32(<rp=%rdi)
movq %r9,32(%rdi)
# qhasm: caller1 = caller1_stack
# asm 1: movq <caller1_stack=stack64#1,>caller1=int64#9
# asm 2: movq <caller1_stack=0(%rsp),>caller1=%r11
movq 0(%rsp),%r11
# qhasm: caller2 = caller2_stack
# asm 1: movq <caller2_stack=stack64#2,>caller2=int64#10
# asm 2: movq <caller2_stack=8(%rsp),>caller2=%r12
movq 8(%rsp),%r12
# qhasm: caller3 = caller3_stack
# asm 1: movq <caller3_stack=stack64#3,>caller3=int64#11
# asm 2: movq <caller3_stack=16(%rsp),>caller3=%r13
movq 16(%rsp),%r13
# qhasm: caller4 = caller4_stack
# asm 1: movq <caller4_stack=stack64#4,>caller4=int64#12
# asm 2: movq <caller4_stack=24(%rsp),>caller4=%r14
movq 24(%rsp),%r14
# qhasm: caller5 = caller5_stack
# asm 1: movq <caller5_stack=stack64#5,>caller5=int64#13
# asm 2: movq <caller5_stack=32(%rsp),>caller5=%r15
movq 32(%rsp),%r15
# qhasm: caller6 = caller6_stack
# asm 1: movq <caller6_stack=stack64#6,>caller6=int64#14
# asm 2: movq <caller6_stack=40(%rsp),>caller6=%rbx
movq 40(%rsp),%rbx
# qhasm: caller7 = caller7_stack
# asm 1: movq <caller7_stack=stack64#7,>caller7=int64#15
# asm 2: movq <caller7_stack=48(%rsp),>caller7=%rbp
movq 48(%rsp),%rbp
# qhasm: leave
add %r11,%rsp
mov %rdi,%rax
mov %rsi,%rdx
ret