-
Notifications
You must be signed in to change notification settings - Fork 12
/
ntt_dilithium_123_456_78_symbolic.s
317 lines (274 loc) · 9.55 KB
/
ntt_dilithium_123_456_78_symbolic.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
///
/// Copyright (c) 2022 Arm Limited
/// Copyright (c) 2022 Hanno Becker
/// Copyright (c) 2023 Amin Abdulrahman, Matthias Kannwischer
/// SPDX-License-Identifier: MIT
///
/// Permission is hereby granted, free of charge, to any person obtaining a copy
/// of this software and associated documentation files (the "Software"), to deal
/// in the Software without restriction, including without limitation the rights
/// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
/// copies of the Software, and to permit persons to whom the Software is
/// furnished to do so, subject to the following conditions:
///
/// The above copyright notice and this permission notice shall be included in all
/// copies or substantial portions of the Software.
///
/// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
/// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
/// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
/// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
/// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
/// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
/// SOFTWARE.
///
.data
.p2align 4
.word 0
.word 0
roots:
#include "ntt_dilithium_123_456_78_twiddles.s"
.text
// Barrett multiplication
.macro mulmod dst, src, const, const_twisted
vmul.s32 \dst, \src, \const
vqrdmulh.s32 \src, \src, \const_twisted
vmla.s32 \dst, \src, modulus
.endm
.macro ct_butterfly a, b, root, root_twisted
mulmod tmp, \b, \root, \root_twisted
vsub.u32 \b, \a, tmp
vadd.u32 \a, \a, tmp
.endm
.macro qsave loc, a // @slothy:no-unfold
vstrw.32 \a, [sp, #\loc\()]
.endm
.macro qrestore a, loc // @slothy:no-unfold
vldrw.32 \a, [sp, #\loc\()]
.endm
.macro restored a, b, loc // @slothy:no-unfold
ldrd \a, \b, [sp, #\loc\()]
.endm
.macro saved loc, a, b // @slothy:no-unfold
strd \a, \b, [sp, #\loc\()]
.endm
.macro restore a, loc // @slothy:no-unfold
ldr \a, [sp, #\loc\()]
.endm
.macro save loc, a // @slothy:no-unfold
str \a, [sp, #\loc\()]
.endm
// Aligns stack =0 mod 16
.macro align_stack_do // @slothy:no-unfold
mov r11, sp
and r12, r11, #0xC // 8 of ==8 mod 16, 0 otherwise
sub sp, sp, r12 // Align stack to 16 byte
sub sp, sp, #16
str r12, [sp]
.endm
// Reverts initial stack correction
.macro align_stack_undo // @slothy:no-unfold
ldr r12, [sp]
add sp, sp, #16
add sp, sp, r12
.endm
#define STACK_SIZE (5*16+8) // +8 is for alignment
#define QSTACK4 (0*16)
#define QSTACK5 (1*16)
#define QSTACK6 (2*16)
#define ROOT0_STACK (3*16)
#define ROOT1_STACK (3*16 + 8)
#define ROOT4_STACK (4*16)
#define RPTR_STACK (4*16 + 8)
.align 4
roots_addr: .word roots
.syntax unified
.type ntt_dilithium_123_456_78, %function
.global ntt_dilithium_123_456_78
ntt_dilithium_123_456_78:
push {r4-r11,lr}
// Save MVE vector registers
vpush {d8-d15}
align_stack_do
sub sp, sp, #STACK_SIZE
modulus .req r12
r_ptr .req r11
.equ modulus_const, -8380417
movw modulus, #:lower16:modulus_const
movt modulus, #:upper16:modulus_const
ldr r_ptr, roots_addr
in .req r0
in_low .req in
in_high .req r1
add in_high, in, #(4*128)
root2 .req r2
root2_tw .req r3
root3 .req r4
root3_tw .req r5
root5 .req r6
root5_tw .req r7
root6 .req r8
root6_tw .req r9
data0 .req q0
data1 .req q1
data2 .req q2
data3 .req q3
data4 .req q1
data5 .req q2
data6 .req q3
data7 .req q4
tmp .req q7
// Layers 1-3
rtmp .req root6
rtmp_tw .req root6_tw
ldrd rtmp, rtmp_tw, [r_ptr], #(7*8)
saved ROOT0_STACK, rtmp, rtmp_tw
ldrd rtmp, rtmp_tw, [r_ptr, #(1*8 - 7*8)]
saved ROOT1_STACK, rtmp, rtmp_tw
ldrd root2, root2_tw, [r_ptr, #(2*8 - 7*8)]
ldrd root3, root3_tw, [r_ptr, #(3*8 - 7*8)]
ldrd rtmp, rtmp_tw, [r_ptr, #(4*8 - 7*8)]
saved ROOT4_STACK, rtmp, rtmp_tw
ldrd root5, root5_tw, [r_ptr, #(5*8 - 7*8)]
ldrd root6, root6_tw, [r_ptr, #(6*8 - 7*8)]
save RPTR_STACK, r_ptr
.unreq rtmp
.unreq rtmp_tw
rtmp .req r10
rtmp_tw .req r11
mov lr, #8
.align 2
layer123_loop:
vldrw.32 data0, [in_low]
vldrw.32 data4, [in_high]
restored rtmp, rtmp_tw, ROOT0_STACK
ct_butterfly data0, data4, rtmp, rtmp_tw
qsave QSTACK4, data4
vldrw.32 data1, [in_low, #128]
vldrw.32 data5, [in_high, #128]
ct_butterfly data1, data5, rtmp, rtmp_tw
qsave QSTACK5, data5
vldrw.32 data2, [in_low, #256]
vldrw.32 data6, [in_high, #256]
ct_butterfly data2, data6, rtmp, rtmp_tw
qsave QSTACK6, data6
vldrw.32 data3, [in_low, #384]
vldrw.32 data7, [in_high, #384]
ct_butterfly data3, data7, rtmp, rtmp_tw
restored rtmp, rtmp_tw, ROOT1_STACK
ct_butterfly data0, data2, rtmp, rtmp_tw
ct_butterfly data1, data3, rtmp, rtmp_tw
ct_butterfly data0, data1, root2, root2_tw
ct_butterfly data2, data3, root3, root3_tw
vstrw.32 data0, [in_low], #16
vstrw.32 data1, [in_low, #(128-16)]
vstrw.32 data2, [in_low, #(256-16)]
vstrw.32 data3, [in_low, #(384-16)]
qrestore data4, QSTACK4
qrestore data5, QSTACK5
qrestore data6, QSTACK6
restored rtmp, rtmp_tw, ROOT4_STACK
ct_butterfly data4, data6, rtmp, rtmp_tw
ct_butterfly data5, data7, rtmp, rtmp_tw
ct_butterfly data4, data5, root5, root5_tw
ct_butterfly data6, data7, root6, root6_tw
vstrw.32 data4, [in_high], #16
vstrw.32 data5, [in_high, #(128-16)]
vstrw.32 data6, [in_high, #(256-16)]
vstrw.32 data7, [in_high, #(384-16)]
le lr, layer123_loop
.unreq in_high
.unreq in_low
sub in, in, #(128)
restore r_ptr, RPTR_STACK
// Layers 4,5,6
.unreq rtmp
.unreq rtmp_tw
rtmp .req r3
rtmp_tw .req r4
mov lr, #8
.align 2
layer456_loop:
ldrd root0, root0_tw, [r_ptr], #(7*8)
ldrd root1, root1_tw, [r_ptr, #((-7 + 1)*8)]
ldrd root2, root2_tw, [r_ptr, #((-7 + 2)*8)]
ldrd root3, root3_tw, [r_ptr, #((-7 + 3)*8)]
ldrd root4, root4_tw, [r_ptr, #((-7 + 4)*8)]
ldrd root5, root5_tw, [r_ptr, #((-7 + 5)*8)]
ldrd root6, root6_tw, [r_ptr, #((-7 + 6)*8)]
vldrw.32 data0, [in]
vldrw.32 data1, [in, #16]
vldrw.32 data2, [in, #32]
vldrw.32 data3, [in, #48]
vldrw.32 data4, [in, #64]
vldrw.32 data5, [in, #80]
vldrw.32 data6, [in, #96]
vldrw.32 data7, [in, #112]
ct_butterfly data0, data4, root0, root0_tw
ct_butterfly data1, data5, root0, root0_tw
ct_butterfly data2, data6, root0, root0_tw
ct_butterfly data3, data7, root0, root0_tw
qsave QSTACK4, data4
qsave QSTACK5, data5
qsave QSTACK6, data6
ct_butterfly data0, data2, root1, root1_tw
ct_butterfly data1, data3, root1, root1_tw
ct_butterfly data0, data1, root2, root2_tw
ct_butterfly data2, data3, root3, root3_tw
vstrw.32 data0, [in], #128
vstrw.32 data1, [in, #(-128+16)]
vstrw.32 data2, [in, #(-128+32)]
vstrw.32 data3, [in, #(-128+48)]
qrestore data4, QSTACK4
qrestore data5, QSTACK5
qrestore data6, QSTACK6
ct_butterfly data4, data6, root4, root4_tw
ct_butterfly data5, data7, root4, root4_tw
ct_butterfly data4, data5, root5, root5_tw
ct_butterfly data6, data7, root6, root6_tw
vstrw.32 data4, [in, #(-128+64)]
vstrw.32 data5, [in, #(-128+80)]
vstrw.32 data6, [in, #(-128+96)]
vstrw.32 data7, [in, #(-128+112)]
le lr, layer456_loop
sub in, in, #(4*256)
.unreq rtmp
.unreq rtmp_tw
.unreq root2
.unreq root2_tw
// Layers 7,8
root0 .req q5
root0_tw .req q6
root1 .req q5
root1_tw .req q6
root2 .req q5
root2_tw .req q6
mov lr, #16
.align 2
layer78_loop:
vld40.32 {data0, data1, data2, data3}, [in]
vld41.32 {data0, data1, data2, data3}, [in]
vld42.32 {data0, data1, data2, data3}, [in]
vld43.32 {data0, data1, data2, data3}, [in]!
vldrw.32 root0, [r_ptr], #+96
vldrw.32 root0_tw, [r_ptr, #(+16-96)]
ct_butterfly data0, data2, root0, root0_tw
ct_butterfly data1, data3, root0, root0_tw
vldrw.32 root1, [r_ptr, #(32 - 96)]
vldrw.32 root1_tw, [r_ptr, #(48 - 96)]
ct_butterfly data0, data1, root1, root1_tw
vldrw.32 root2, [r_ptr, #(64-96)]
vldrw.32 root2_tw, [r_ptr, #(80-96)]
ct_butterfly data2, data3, root2, root2_tw
vstrw.32 data0, [in, #( 0 - 64)]
vstrw.32 data1, [in, #(16 - 64)]
vstrw.32 data2, [in, #(32 - 64)]
vstrw.32 data3, [in, #(48 - 64)]
le lr, layer78_loop
add sp, sp, #STACK_SIZE
align_stack_undo
// Restore MVE vector registers
vpop {d8-d15}
// Restore GPRs
pop {r4-r11,lr}
bx lr