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
|
; AesOpt.asm -- Intel's AES.
; 2009-12-12 : Igor Pavlov : Public domain
%include "7zAsm.asm"
MY_ASM_START
%ifndef x64
; FIXME .xmm
%endif
%define rD r2
%define rN r0
%macro MY_PROLOG 1 ; MY_PROLOG macro reg:req
%ifdef x64
%ifdef CYGWIN64
; ivAes : %rcx
; data : %rdx
; numBlocks : %r8
%else
mov RCX,RDI
mov R8 ,RDX
mov RDX,RSI
%endif
; movdqa [r4 + 8], xmm6
; movdqa [r4 + 8 + 16], xmm7
%endif
push r3
push r5
%ifdef x64
%ifdef CYGWIN64
push r6
%endif
mov rN, r8
%else
push r6
mov ecx, [r4 + REG_SIZE * 4]
mov edx, [r4 + REG_SIZE * 5]
mov rN, [r4 + REG_SIZE * 6]
%endif
mov x6, [r1 + 16]
shl x6, 5
movdqa %1, [r1] ; reg
add r1, 32
%endmacro
%macro MY_EPILOG 0
%ifdef x64
%ifdef CYGWIN64
pop r6
%endif
%else
pop r6
%endif
pop r5
pop r3
%ifdef x64
; movdqa xmm6, [r4 + 8]
; movdqa xmm7, [r4 + 8 + 16]
%endif
MY_ENDP
%endmacro
ways equ 4
ways16 equ (ways * 16)
%macro OP_W 2 ; op, op2
%define i 0
%1 xmm0,%2
%define i 1
%1 xmm1,%2
%define i 2
%1 xmm2,%2
%define i 3
%1 xmm3,%2
%endmacro
%macro LOAD_OP 2 ; LOAD_OP macro op:req, offs:req
%1 xmm0, [r1 + r3 %2]
%endmacro
%macro LOAD_OP_W 2 ; LOAD_OP_W macro op:req, offs:req
movdqa xmm7, [r1 + r3 %2]
; OP_W %1, xmm7
%1 xmm0,xmm7
%1 xmm1,xmm7
%1 xmm2,xmm7
%1 xmm3,xmm7
%endmacro
; ---------- AES-CBC Decode ----------
%macro CBC_DEC_UPDATE 2 ; CBC_DEC_UPDATE macro reg, offs
pxor %1, xmm6
movdqa xmm6, [rD + %2]
movdqa [rD + %2], %1
%endmacro
%macro DECODE 1 ; macro op:req
%1 aesdec, +16
%%B:
%1 aesdec, +0
%1 aesdec, -16
sub x3, 32
jnz %%B
%1 aesdeclast, +0
%endmacro
; void AesCbc_Decode_Intel(UInt32 *ivAes, Byte *data, size_t numBlocks)
MY_PROC AesCbc_Decode_Intel, 3
MY_PROLOG xmm6
sub x6, 32
jmp check2
align 16
nextBlocks2:
mov x3, x6
OP_W movdqa, [rD + i * 16]
LOAD_OP_W pxor, +32
DECODE LOAD_OP_W
;OP_W CBC_DEC_UPDATE, i * 16
CBC_DEC_UPDATE xmm0, 0 * 16
CBC_DEC_UPDATE xmm1, 1 * 16
CBC_DEC_UPDATE xmm2, 2 * 16
CBC_DEC_UPDATE xmm3, 3 * 16
add rD, ways16
check2:
sub rN, ways
jnc nextBlocks2
add rN, ways
jmp check
nextBlock:
mov x3, x6
movdqa xmm1, [rD]
LOAD_OP movdqa, +32
pxor xmm0, xmm1
DECODE LOAD_OP
pxor xmm0, xmm6
movdqa [rD], xmm0
movdqa xmm6, xmm1
add rD, 16
check:
sub rN, 1
jnc nextBlock
movdqa [r1 - 32], xmm6
MY_EPILOG
; ---------- AES-CBC Encode ----------
%macro ENCODE 1 ; macro op:req
%1 aesenc, -16
%%B:
%1 aesenc, +0
%1 aesenc, +16
add r3, 32
jnz %%B
%1 aesenclast, +0
%endmacro
MY_PROC AesCbc_Encode_Intel, 3
MY_PROLOG xmm0
add r1, r6
neg r6
add r6, 32
jmp check_e
align 16
nextBlock_e:
mov r3, r6
pxor xmm0, [rD]
pxor xmm0, [r1 + r3 - 32]
ENCODE LOAD_OP
movdqa [rD], xmm0
add rD, 16
check_e:
sub rN, 1
jnc nextBlock_e
movdqa [r1 + r6 - 64], xmm0
MY_EPILOG
; ---------- AES-CTR ----------
%macro XOR_UPD_1 2 ; reg, offs
pxor %1, [rD + %2]
%endmacro
%macro XOR_UPD_2 2 ; reg, offs
movdqa [rD + %2], %1
%endmacro
MY_PROC AesCtr_Code_Intel, 3
MY_PROLOG xmm6
mov r5, r4
shr r5, 4
dec r5
shl r5, 4
mov DWORD [r5], 1
mov DWORD [r5 + 4], 0
mov DWORD [r5 + 8], 0
mov DWORD [r5 + 12], 0
add r1, r6
neg r6
add r6, 32
jmp check2_c
align 16
nextBlocks2_c:
movdqa xmm7, [r5]
; i = 0
; rept ways
; paddq xmm6, xmm7
; movdqa @CatStr(xmm,%i), xmm6
; i = i + 1
; endm
paddq xmm6, xmm7
movdqa xmm0, xmm6
paddq xmm6, xmm7
movdqa xmm1, xmm6
paddq xmm6, xmm7
movdqa xmm2, xmm6
paddq xmm6, xmm7
movdqa xmm3, xmm6
mov r3, r6
LOAD_OP_W pxor, -32
ENCODE LOAD_OP_W
;OP_W XOR_UPD_1, i * 16
XOR_UPD_1 xmm0, 0 * 16
XOR_UPD_1 xmm1, 1 * 16
XOR_UPD_1 xmm2, 2 * 16
XOR_UPD_1 xmm3, 3 * 16
;OP_W XOR_UPD_2, i * 16
XOR_UPD_2 xmm0, 0 * 16
XOR_UPD_2 xmm1, 1 * 16
XOR_UPD_2 xmm2, 2 * 16
XOR_UPD_2 xmm3, 3 * 16
add rD, ways16
check2_c:
sub rN, ways
jnc nextBlocks2_c
add rN, ways
jmp check_c
nextBlock_c:
paddq xmm6, [r5]
mov r3, r6
movdqa xmm0, [r1 + r3 - 32]
pxor xmm0, xmm6
ENCODE LOAD_OP
XOR_UPD_1 xmm0, 0
XOR_UPD_2 xmm0, 0
add rD, 16
check_c:
sub rN, 1
jnc nextBlock_c
movdqa [r1 + r6 - 64], xmm6
MY_EPILOG
; end
%ifidn __OUTPUT_FORMAT__,elf
section .note.GNU-stack noalloc noexec nowrite progbits
%endif
|
