]>
Commit | Line | Data |
---|---|---|
1e59de90 TL |
1 | ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; |
2 | ; Copyright(c) 2011-2019 Intel Corporation All rights reserved. | |
3 | ; | |
4 | ; Redistribution and use in source and binary forms, with or without | |
5 | ; modification, are permitted provided that the following conditions | |
6 | ; are met: | |
7 | ; * Redistributions of source code must retain the above copyright | |
8 | ; notice, this list of conditions and the following disclaimer. | |
9 | ; * Redistributions in binary form must reproduce the above copyright | |
10 | ; notice, this list of conditions and the following disclaimer in | |
11 | ; the documentation and/or other materials provided with the | |
12 | ; distribution. | |
13 | ; * Neither the name of Intel Corporation nor the names of its | |
14 | ; contributors may be used to endorse or promote products derived | |
15 | ; from this software without specific prior written permission. | |
16 | ; | |
17 | ; THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS | |
18 | ; "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT | |
19 | ; LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR | |
20 | ; A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT | |
21 | ; OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, | |
22 | ; SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT | |
23 | ; LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, | |
24 | ; DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY | |
25 | ; THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT | |
26 | ; (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE | |
27 | ; OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. | |
28 | ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; | |
29 | ||
30 | %ifndef _AES_COMMON_ASM_ | |
31 | %define _AES_COMMON_ASM_ | |
32 | ||
33 | %include "reg_sizes.asm" | |
34 | ||
35 | ;; ============================================================================= | |
36 | ;; Generic macro to produce code that executes %%OPCODE instruction | |
37 | ;; on selected number of AES blocks (16 bytes long ) between 0 and 16. | |
38 | ;; All three operands of the instruction come from registers. | |
39 | ;; Note: if 3 blocks are left at the end instruction is produced to operate all | |
40 | ;; 4 blocks (full width of ZMM) | |
41 | ||
42 | %macro ZMM_OPCODE3_DSTR_SRC1R_SRC2R_BLOCKS_0_16 14 | |
43 | %define %%NUM_BLOCKS %1 ; [in] numerical value, number of AES blocks (0 to 16) | |
44 | %define %%OPCODE %2 ; [in] instruction name | |
45 | %define %%DST0 %3 ; [out] destination ZMM register | |
46 | %define %%DST1 %4 ; [out] destination ZMM register | |
47 | %define %%DST2 %5 ; [out] destination ZMM register | |
48 | %define %%DST3 %6 ; [out] destination ZMM register | |
49 | %define %%SRC1_0 %7 ; [in] source 1 ZMM register | |
50 | %define %%SRC1_1 %8 ; [in] source 1 ZMM register | |
51 | %define %%SRC1_2 %9 ; [in] source 1 ZMM register | |
52 | %define %%SRC1_3 %10 ; [in] source 1 ZMM register | |
53 | %define %%SRC2_0 %11 ; [in] source 2 ZMM register | |
54 | %define %%SRC2_1 %12 ; [in] source 2 ZMM register | |
55 | %define %%SRC2_2 %13 ; [in] source 2 ZMM register | |
56 | %define %%SRC2_3 %14 ; [in] source 2 ZMM register | |
57 | ||
58 | %assign reg_idx 0 | |
59 | %assign blocks_left %%NUM_BLOCKS | |
60 | ||
61 | %rep (%%NUM_BLOCKS / 4) | |
62 | %xdefine %%DSTREG %%DST %+ reg_idx | |
63 | %xdefine %%SRC1REG %%SRC1_ %+ reg_idx | |
64 | %xdefine %%SRC2REG %%SRC2_ %+ reg_idx | |
65 | %%OPCODE %%DSTREG, %%SRC1REG, %%SRC2REG | |
66 | %undef %%DSTREG | |
67 | %undef %%SRC1REG | |
68 | %undef %%SRC2REG | |
69 | %assign reg_idx (reg_idx + 1) | |
70 | %assign blocks_left (blocks_left - 4) | |
71 | %endrep | |
72 | ||
73 | %xdefine %%DSTREG %%DST %+ reg_idx | |
74 | %xdefine %%SRC1REG %%SRC1_ %+ reg_idx | |
75 | %xdefine %%SRC2REG %%SRC2_ %+ reg_idx | |
76 | ||
77 | %if blocks_left == 1 | |
78 | %%OPCODE XWORD(%%DSTREG), XWORD(%%SRC1REG), XWORD(%%SRC2REG) | |
79 | %elif blocks_left == 2 | |
80 | %%OPCODE YWORD(%%DSTREG), YWORD(%%SRC1REG), YWORD(%%SRC2REG) | |
81 | %elif blocks_left == 3 | |
82 | %%OPCODE %%DSTREG, %%SRC1REG, %%SRC2REG | |
83 | %endif | |
84 | ||
85 | %endmacro | |
86 | ||
87 | ;; ============================================================================= | |
88 | ;; Loads specified number of AES blocks into ZMM registers | |
89 | ;; %%FLAGS are optional and only affect behavior when 3 trailing blocks are left | |
90 | ;; - if %%FlAGS not provided then exactly 3 blocks are loaded (move and insert) | |
91 | ;; - if "load_4_instead_of_3" option is passed then 4 blocks are loaded | |
92 | %macro ZMM_LOAD_BLOCKS_0_16 7-8 | |
93 | %define %%NUM_BLOCKS %1 ; [in] numerical value, number of AES blocks (0 to 16) | |
94 | %define %%INP %2 ; [in] input data pointer to read from | |
95 | %define %%DATA_OFFSET %3 ; [in] offset to the output pointer (GP or numerical) | |
96 | %define %%DST0 %4 ; [out] ZMM register with loaded data | |
97 | %define %%DST1 %5 ; [out] ZMM register with loaded data | |
98 | %define %%DST2 %6 ; [out] ZMM register with loaded data | |
99 | %define %%DST3 %7 ; [out] ZMM register with loaded data | |
100 | %define %%FLAGS %8 ; [in] optional "load_4_instead_of_3" | |
101 | ||
102 | %assign src_offset 0 | |
103 | %assign dst_idx 0 | |
104 | ||
105 | %rep (%%NUM_BLOCKS / 4) | |
106 | %xdefine %%DSTREG %%DST %+ dst_idx | |
107 | vmovdqu8 %%DSTREG, [%%INP + %%DATA_OFFSET + src_offset] | |
108 | %undef %%DSTREG | |
109 | %assign src_offset (src_offset + 64) | |
110 | %assign dst_idx (dst_idx + 1) | |
111 | %endrep | |
112 | ||
113 | %assign blocks_left (%%NUM_BLOCKS % 4) | |
114 | %xdefine %%DSTREG %%DST %+ dst_idx | |
115 | ||
116 | %if blocks_left == 1 | |
117 | vmovdqu8 XWORD(%%DSTREG), [%%INP + %%DATA_OFFSET + src_offset] | |
118 | %elif blocks_left == 2 | |
119 | vmovdqu8 YWORD(%%DSTREG), [%%INP + %%DATA_OFFSET + src_offset] | |
120 | %elif blocks_left == 3 | |
121 | %ifidn %%FLAGS, load_4_instead_of_3 | |
122 | vmovdqu8 %%DSTREG, [%%INP + %%DATA_OFFSET + src_offset] | |
123 | %else | |
124 | vmovdqu8 YWORD(%%DSTREG), [%%INP + %%DATA_OFFSET + src_offset] | |
125 | vinserti64x2 %%DSTREG, [%%INP + %%DATA_OFFSET + src_offset + 32], 2 | |
126 | %endif | |
127 | %endif | |
128 | ||
129 | %endmacro | |
130 | ||
131 | ;; ============================================================================= | |
132 | ;; Loads specified number of AES blocks into ZMM registers using mask register | |
133 | ;; for the last loaded register (xmm, ymm or zmm). | |
134 | ;; Loads take place at 1 byte granularity. | |
135 | %macro ZMM_LOAD_MASKED_BLOCKS_0_16 8 | |
136 | %define %%NUM_BLOCKS %1 ; [in] numerical value, number of AES blocks (0 to 16) | |
137 | %define %%INP %2 ; [in] input data pointer to read from | |
138 | %define %%DATA_OFFSET %3 ; [in] offset to the output pointer (GP or numerical) | |
139 | %define %%DST0 %4 ; [out] ZMM register with loaded data | |
140 | %define %%DST1 %5 ; [out] ZMM register with loaded data | |
141 | %define %%DST2 %6 ; [out] ZMM register with loaded data | |
142 | %define %%DST3 %7 ; [out] ZMM register with loaded data | |
143 | %define %%MASK %8 ; [in] mask register | |
144 | ||
145 | %assign src_offset 0 | |
146 | %assign dst_idx 0 | |
147 | %assign blocks_left %%NUM_BLOCKS | |
148 | ||
149 | %if %%NUM_BLOCKS > 0 | |
150 | %rep (((%%NUM_BLOCKS + 3) / 4) - 1) | |
151 | %xdefine %%DSTREG %%DST %+ dst_idx | |
152 | vmovdqu8 %%DSTREG, [%%INP + %%DATA_OFFSET + src_offset] | |
153 | %undef %%DSTREG | |
154 | %assign src_offset (src_offset + 64) | |
155 | %assign dst_idx (dst_idx + 1) | |
156 | %assign blocks_left (blocks_left - 4) | |
157 | %endrep | |
158 | %endif ; %if %%NUM_BLOCKS > 0 | |
159 | ||
160 | %xdefine %%DSTREG %%DST %+ dst_idx | |
161 | ||
162 | %if blocks_left == 1 | |
163 | vmovdqu8 XWORD(%%DSTREG){%%MASK}{z}, [%%INP + %%DATA_OFFSET + src_offset] | |
164 | %elif blocks_left == 2 | |
165 | vmovdqu8 YWORD(%%DSTREG){%%MASK}{z}, [%%INP + %%DATA_OFFSET + src_offset] | |
166 | %elif (blocks_left == 3 || blocks_left == 4) | |
167 | vmovdqu8 %%DSTREG{%%MASK}{z}, [%%INP + %%DATA_OFFSET + src_offset] | |
168 | %endif | |
169 | ||
170 | %endmacro | |
171 | ||
172 | ;; ============================================================================= | |
173 | ;; Stores specified number of AES blocks from ZMM registers | |
174 | %macro ZMM_STORE_BLOCKS_0_16 7 | |
175 | %define %%NUM_BLOCKS %1 ; [in] numerical value, number of AES blocks (0 to 16) | |
176 | %define %%OUTP %2 ; [in] output data pointer to write to | |
177 | %define %%DATA_OFFSET %3 ; [in] offset to the output pointer (GP or numerical) | |
178 | %define %%SRC0 %4 ; [in] ZMM register with data to store | |
179 | %define %%SRC1 %5 ; [in] ZMM register with data to store | |
180 | %define %%SRC2 %6 ; [in] ZMM register with data to store | |
181 | %define %%SRC3 %7 ; [in] ZMM register with data to store | |
182 | ||
183 | %assign dst_offset 0 | |
184 | %assign src_idx 0 | |
185 | ||
186 | %rep (%%NUM_BLOCKS / 4) | |
187 | %xdefine %%SRCREG %%SRC %+ src_idx | |
188 | vmovdqu8 [%%OUTP + %%DATA_OFFSET + dst_offset], %%SRCREG | |
189 | %undef %%SRCREG | |
190 | %assign dst_offset (dst_offset + 64) | |
191 | %assign src_idx (src_idx + 1) | |
192 | %endrep | |
193 | ||
194 | %assign blocks_left (%%NUM_BLOCKS % 4) | |
195 | %xdefine %%SRCREG %%SRC %+ src_idx | |
196 | ||
197 | %if blocks_left == 1 | |
198 | vmovdqu8 [%%OUTP + %%DATA_OFFSET + dst_offset], XWORD(%%SRCREG) | |
199 | %elif blocks_left == 2 | |
200 | vmovdqu8 [%%OUTP + %%DATA_OFFSET + dst_offset], YWORD(%%SRCREG) | |
201 | %elif blocks_left == 3 | |
202 | vmovdqu8 [%%OUTP + %%DATA_OFFSET + dst_offset], YWORD(%%SRCREG) | |
203 | vextracti32x4 [%%OUTP + %%DATA_OFFSET + dst_offset + 32], %%SRCREG, 2 | |
204 | %endif | |
205 | ||
206 | %endmacro | |
207 | ||
208 | ;; ============================================================================= | |
209 | ;; Stores specified number of AES blocks from ZMM registers with mask register | |
210 | ;; for the last loaded register (xmm, ymm or zmm). | |
211 | ;; Stores take place at 1 byte granularity. | |
212 | %macro ZMM_STORE_MASKED_BLOCKS_0_16 8 | |
213 | %define %%NUM_BLOCKS %1 ; [in] numerical value, number of AES blocks (0 to 16) | |
214 | %define %%OUTP %2 ; [in] output data pointer to write to | |
215 | %define %%DATA_OFFSET %3 ; [in] offset to the output pointer (GP or numerical) | |
216 | %define %%SRC0 %4 ; [in] ZMM register with data to store | |
217 | %define %%SRC1 %5 ; [in] ZMM register with data to store | |
218 | %define %%SRC2 %6 ; [in] ZMM register with data to store | |
219 | %define %%SRC3 %7 ; [in] ZMM register with data to store | |
220 | %define %%MASK %8 ; [in] mask register | |
221 | ||
222 | %assign dst_offset 0 | |
223 | %assign src_idx 0 | |
224 | %assign blocks_left %%NUM_BLOCKS | |
225 | ||
226 | %if %%NUM_BLOCKS > 0 | |
227 | %rep (((%%NUM_BLOCKS + 3) / 4) - 1) | |
228 | %xdefine %%SRCREG %%SRC %+ src_idx | |
229 | vmovdqu8 [%%OUTP + %%DATA_OFFSET + dst_offset], %%SRCREG | |
230 | %undef %%SRCREG | |
231 | %assign dst_offset (dst_offset + 64) | |
232 | %assign src_idx (src_idx + 1) | |
233 | %assign blocks_left (blocks_left - 4) | |
234 | %endrep | |
235 | %endif ; %if %%NUM_BLOCKS > 0 | |
236 | ||
237 | %xdefine %%SRCREG %%SRC %+ src_idx | |
238 | ||
239 | %if blocks_left == 1 | |
240 | vmovdqu8 [%%OUTP + %%DATA_OFFSET + dst_offset]{%%MASK}, XWORD(%%SRCREG) | |
241 | %elif blocks_left == 2 | |
242 | vmovdqu8 [%%OUTP + %%DATA_OFFSET + dst_offset]{%%MASK}, YWORD(%%SRCREG) | |
243 | %elif (blocks_left == 3 || blocks_left == 4) | |
244 | vmovdqu8 [%%OUTP + %%DATA_OFFSET + dst_offset]{%%MASK}, %%SRCREG | |
245 | %endif | |
246 | ||
247 | %endmacro | |
248 | ||
249 | ;;; =========================================================================== | |
250 | ;;; Handles AES encryption rounds | |
251 | ;;; It handles special cases: the last and first rounds | |
252 | ;;; Optionally, it performs XOR with data after the last AES round. | |
253 | ;;; Uses NROUNDS parameterto check what needs to be done for the current round. | |
254 | ;;; If 3 blocks are trailing then operation on whole ZMM is performed (4 blocks). | |
255 | %macro ZMM_AESENC_ROUND_BLOCKS_0_16 12 | |
256 | %define %%L0B0_3 %1 ; [in/out] zmm; blocks 0 to 3 | |
257 | %define %%L0B4_7 %2 ; [in/out] zmm; blocks 4 to 7 | |
258 | %define %%L0B8_11 %3 ; [in/out] zmm; blocks 8 to 11 | |
259 | %define %%L0B12_15 %4 ; [in/out] zmm; blocks 12 to 15 | |
260 | %define %%KEY %5 ; [in] zmm containing round key | |
261 | %define %%ROUND %6 ; [in] round number | |
262 | %define %%D0_3 %7 ; [in] zmm or no_data; plain/cipher text blocks 0-3 | |
263 | %define %%D4_7 %8 ; [in] zmm or no_data; plain/cipher text blocks 4-7 | |
264 | %define %%D8_11 %9 ; [in] zmm or no_data; plain/cipher text blocks 8-11 | |
265 | %define %%D12_15 %10 ; [in] zmm or no_data; plain/cipher text blocks 12-15 | |
266 | %define %%NUMBL %11 ; [in] number of blocks; numerical value | |
267 | %define %%NROUNDS %12 ; [in] number of rounds; numerical value | |
268 | ||
269 | ;;; === first AES round | |
270 | %if (%%ROUND < 1) | |
271 | ;; round 0 | |
272 | ZMM_OPCODE3_DSTR_SRC1R_SRC2R_BLOCKS_0_16 %%NUMBL, vpxorq, \ | |
273 | %%L0B0_3, %%L0B4_7, %%L0B8_11, %%L0B12_15, \ | |
274 | %%L0B0_3, %%L0B4_7, %%L0B8_11, %%L0B12_15, \ | |
275 | %%KEY, %%KEY, %%KEY, %%KEY | |
276 | %endif ; ROUND 0 | |
277 | ||
278 | ;;; === middle AES rounds | |
279 | %if (%%ROUND >= 1 && %%ROUND <= %%NROUNDS) | |
280 | ;; rounds 1 to 9/11/13 | |
281 | ZMM_OPCODE3_DSTR_SRC1R_SRC2R_BLOCKS_0_16 %%NUMBL, vaesenc, \ | |
282 | %%L0B0_3, %%L0B4_7, %%L0B8_11, %%L0B12_15, \ | |
283 | %%L0B0_3, %%L0B4_7, %%L0B8_11, %%L0B12_15, \ | |
284 | %%KEY, %%KEY, %%KEY, %%KEY | |
285 | %endif ; rounds 1 to 9/11/13 | |
286 | ||
287 | ;;; === last AES round | |
288 | %if (%%ROUND > %%NROUNDS) | |
289 | ;; the last round - mix enclast with text xor's | |
290 | ZMM_OPCODE3_DSTR_SRC1R_SRC2R_BLOCKS_0_16 %%NUMBL, vaesenclast, \ | |
291 | %%L0B0_3, %%L0B4_7, %%L0B8_11, %%L0B12_15, \ | |
292 | %%L0B0_3, %%L0B4_7, %%L0B8_11, %%L0B12_15, \ | |
293 | %%KEY, %%KEY, %%KEY, %%KEY | |
294 | ||
295 | ;;; === XOR with data | |
296 | %ifnidn %%D0_3, no_data | |
297 | %ifnidn %%D4_7, no_data | |
298 | %ifnidn %%D8_11, no_data | |
299 | %ifnidn %%D12_15, no_data | |
300 | ZMM_OPCODE3_DSTR_SRC1R_SRC2R_BLOCKS_0_16 %%NUMBL, vpxorq, \ | |
301 | %%L0B0_3, %%L0B4_7, %%L0B8_11, %%L0B12_15, \ | |
302 | %%L0B0_3, %%L0B4_7, %%L0B8_11, %%L0B12_15, \ | |
303 | %%D0_3, %%D4_7, %%D8_11, %%D12_15 | |
304 | %endif ; !no_data | |
305 | %endif ; !no_data | |
306 | %endif ; !no_data | |
307 | %endif ; !no_data | |
308 | ||
309 | %endif ; The last round | |
310 | ||
311 | %endmacro | |
312 | ||
313 | ;;; =========================================================================== | |
314 | ;;; Handles AES decryption rounds | |
315 | ;;; It handles special cases: the last and first rounds | |
316 | ;;; Optionally, it performs XOR with data after the last AES round. | |
317 | ;;; Uses NROUNDS parameter to check what needs to be done for the current round. | |
318 | ;;; If 3 blocks are trailing then operation on whole ZMM is performed (4 blocks). | |
319 | %macro ZMM_AESDEC_ROUND_BLOCKS_0_16 12 | |
320 | %define %%L0B0_3 %1 ; [in/out] zmm; blocks 0 to 3 | |
321 | %define %%L0B4_7 %2 ; [in/out] zmm; blocks 4 to 7 | |
322 | %define %%L0B8_11 %3 ; [in/out] zmm; blocks 8 to 11 | |
323 | %define %%L0B12_15 %4 ; [in/out] zmm; blocks 12 to 15 | |
324 | %define %%KEY %5 ; [in] zmm containing round key | |
325 | %define %%ROUND %6 ; [in] round number | |
326 | %define %%D0_3 %7 ; [in] zmm or no_data; cipher text blocks 0-3 | |
327 | %define %%D4_7 %8 ; [in] zmm or no_data; cipher text blocks 4-7 | |
328 | %define %%D8_11 %9 ; [in] zmm or no_data; cipher text blocks 8-11 | |
329 | %define %%D12_15 %10 ; [in] zmm or no_data; cipher text blocks 12-15 | |
330 | %define %%NUMBL %11 ; [in] number of blocks; numerical value | |
331 | %define %%NROUNDS %12 ; [in] number of rounds; numerical value | |
332 | ||
333 | ;;; === first AES round | |
334 | %if (%%ROUND < 1) | |
335 | ;; round 0 | |
336 | ZMM_OPCODE3_DSTR_SRC1R_SRC2R_BLOCKS_0_16 %%NUMBL, vpxorq, \ | |
337 | %%L0B0_3, %%L0B4_7, %%L0B8_11, %%L0B12_15, \ | |
338 | %%L0B0_3, %%L0B4_7, %%L0B8_11, %%L0B12_15, \ | |
339 | %%KEY, %%KEY, %%KEY, %%KEY | |
340 | %endif ; ROUND 0 | |
341 | ||
342 | ;;; === middle AES rounds | |
343 | %if (%%ROUND >= 1 && %%ROUND <= %%NROUNDS) | |
344 | ;; rounds 1 to 9/11/13 | |
345 | ZMM_OPCODE3_DSTR_SRC1R_SRC2R_BLOCKS_0_16 %%NUMBL, vaesdec, \ | |
346 | %%L0B0_3, %%L0B4_7, %%L0B8_11, %%L0B12_15, \ | |
347 | %%L0B0_3, %%L0B4_7, %%L0B8_11, %%L0B12_15, \ | |
348 | %%KEY, %%KEY, %%KEY, %%KEY | |
349 | %endif ; rounds 1 to 9/11/13 | |
350 | ||
351 | ;;; === last AES round | |
352 | %if (%%ROUND > %%NROUNDS) | |
353 | ;; the last round - mix enclast with text xor's | |
354 | ZMM_OPCODE3_DSTR_SRC1R_SRC2R_BLOCKS_0_16 %%NUMBL, vaesdeclast, \ | |
355 | %%L0B0_3, %%L0B4_7, %%L0B8_11, %%L0B12_15, \ | |
356 | %%L0B0_3, %%L0B4_7, %%L0B8_11, %%L0B12_15, \ | |
357 | %%KEY, %%KEY, %%KEY, %%KEY | |
358 | ||
359 | ;;; === XOR with data | |
360 | %ifnidn %%D0_3, no_data | |
361 | %ifnidn %%D4_7, no_data | |
362 | %ifnidn %%D8_11, no_data | |
363 | %ifnidn %%D12_15, no_data | |
364 | ZMM_OPCODE3_DSTR_SRC1R_SRC2R_BLOCKS_0_16 %%NUMBL, vpxorq, \ | |
365 | %%L0B0_3, %%L0B4_7, %%L0B8_11, %%L0B12_15, \ | |
366 | %%L0B0_3, %%L0B4_7, %%L0B8_11, %%L0B12_15, \ | |
367 | %%D0_3, %%D4_7, %%D8_11, %%D12_15 | |
368 | %endif ; !no_data | |
369 | %endif ; !no_data | |
370 | %endif ; !no_data | |
371 | %endif ; !no_data | |
372 | ||
373 | %endif ; The last round | |
374 | ||
375 | %endmacro | |
376 | ||
377 | %endif ;; _AES_COMMON_ASM |