1#!/usr/bin/env perl
2
3# ====================================================================
4# Written by Andy Polyakov <appro@fy.chalmers.se> for the OpenSSL
5# project. The module is, however, dual licensed under OpenSSL and
6# CRYPTOGAMS licenses depending on where you obtain it. For further
7# details see http://www.openssl.org/~appro/cryptogams/.
8# ====================================================================
9
10# AES for PA-RISC.
11#
12# June 2009.
13#
14# The module is mechanical transliteration of aes-sparcv9.pl, but with
15# a twist: S-boxes are compressed even further down to 1K+256B. On
16# PA-7100LC performance is ~40% better than gcc 3.2 generated code and
17# is about 33 cycles per byte processed with 128-bit key. Newer CPUs
18# perform at 16 cycles per byte. It's not faster than code generated
19# by vendor compiler, but recall that it has compressed S-boxes, which
20# requires extra processing.
21#
22# Special thanks to polarhome.com for providing HP-UX account.
23
24$flavour = shift;
25$output = shift;
26open STDOUT,">$output";
27
28if ($flavour =~ /64/) {
29	$LEVEL		="2.0W";
30	$SIZE_T		=8;
31	$FRAME_MARKER	=80;
32	$SAVED_RP	=16;
33	$PUSH		="std";
34	$PUSHMA		="std,ma";
35	$POP		="ldd";
36	$POPMB		="ldd,mb";
37} else {
38	$LEVEL		="1.0";
39	$SIZE_T		=4;
40	$FRAME_MARKER	=48;
41	$SAVED_RP	=20;
42	$PUSH		="stw";
43	$PUSHMA		="stwm";
44	$POP		="ldw";
45	$POPMB		="ldwm";
46}
47
48$FRAME=16*$SIZE_T+$FRAME_MARKER;# 16 saved regs + frame marker
49				#                 [+ argument transfer]
50$inp="%r26";	# arg0
51$out="%r25";	# arg1
52$key="%r24";	# arg2
53
54($s0,$s1,$s2,$s3) = ("%r1","%r2","%r3","%r4");
55($t0,$t1,$t2,$t3) = ("%r5","%r6","%r7","%r8");
56
57($acc0, $acc1, $acc2, $acc3, $acc4, $acc5, $acc6, $acc7,
58 $acc8, $acc9,$acc10,$acc11,$acc12,$acc13,$acc14,$acc15) =
59("%r9","%r10","%r11","%r12","%r13","%r14","%r15","%r16",
60"%r17","%r18","%r19","%r20","%r21","%r22","%r23","%r26");
61
62$tbl="%r28";
63$rounds="%r29";
64
65$code=<<___;
66	.LEVEL	$LEVEL
67	.text
68
69	.EXPORT	aes_encrypt_internal,ENTRY,ARGW0=GR,ARGW1=GR,ARGW2=GR
70	.ALIGN	64
71aes_encrypt_internal
72	.PROC
73	.CALLINFO	FRAME=`$FRAME-16*$SIZE_T`,NO_CALLS,SAVE_RP,ENTRY_GR=18
74	.ENTRY
75	$PUSH	%r2,-$SAVED_RP(%sp)	; standard prologue
76	$PUSHMA	%r3,$FRAME(%sp)
77	$PUSH	%r4,`-$FRAME+1*$SIZE_T`(%sp)
78	$PUSH	%r5,`-$FRAME+2*$SIZE_T`(%sp)
79	$PUSH	%r6,`-$FRAME+3*$SIZE_T`(%sp)
80	$PUSH	%r7,`-$FRAME+4*$SIZE_T`(%sp)
81	$PUSH	%r8,`-$FRAME+5*$SIZE_T`(%sp)
82	$PUSH	%r9,`-$FRAME+6*$SIZE_T`(%sp)
83	$PUSH	%r10,`-$FRAME+7*$SIZE_T`(%sp)
84	$PUSH	%r11,`-$FRAME+8*$SIZE_T`(%sp)
85	$PUSH	%r12,`-$FRAME+9*$SIZE_T`(%sp)
86	$PUSH	%r13,`-$FRAME+10*$SIZE_T`(%sp)
87	$PUSH	%r14,`-$FRAME+11*$SIZE_T`(%sp)
88	$PUSH	%r15,`-$FRAME+12*$SIZE_T`(%sp)
89	$PUSH	%r16,`-$FRAME+13*$SIZE_T`(%sp)
90	$PUSH	%r17,`-$FRAME+14*$SIZE_T`(%sp)
91	$PUSH	%r18,`-$FRAME+15*$SIZE_T`(%sp)
92
93	ldi	3,$t0
94#ifdef __PIC__
95	addil	LT'L\$AES_Te, %r19
96	ldw	RT'L\$AES_Te(%r1), $tbl
97#else
98	ldil	L'L\$AES_Te, %t1
99	ldo	R'L\$AES_Te(%t1), $tbl
100#endif
101
102	and	$inp,$t0,$t0
103	sub	$inp,$t0,$inp
104	ldw	0($inp),$s0
105	ldw	4($inp),$s1
106	ldw	8($inp),$s2
107	comib,=	0,$t0,L\$enc_inp_aligned
108	ldw	12($inp),$s3
109
110	sh3addl	$t0,%r0,$t0
111	subi	32,$t0,$t0
112	mtctl	$t0,%cr11
113	ldw	16($inp),$t1
114	vshd	$s0,$s1,$s0
115	vshd	$s1,$s2,$s1
116	vshd	$s2,$s3,$s2
117	vshd	$s3,$t1,$s3
118
119L\$enc_inp_aligned
120	bl	_parisc_AES_encrypt,%r31
121	nop
122
123	extru,<> $out,31,2,%r0
124	b	L\$enc_out_aligned
125	nop
126
127	_srm	$s0,24,$acc0
128	_srm	$s0,16,$acc1
129	stb	$acc0,0($out)
130	_srm	$s0,8,$acc2
131	stb	$acc1,1($out)
132	_srm	$s1,24,$acc4
133	stb	$acc2,2($out)
134	_srm	$s1,16,$acc5
135	stb	$s0,3($out)
136	_srm	$s1,8,$acc6
137	stb	$acc4,4($out)
138	_srm	$s2,24,$acc0
139	stb	$acc5,5($out)
140	_srm	$s2,16,$acc1
141	stb	$acc6,6($out)
142	_srm	$s2,8,$acc2
143	stb	$s1,7($out)
144	_srm	$s3,24,$acc4
145	stb	$acc0,8($out)
146	_srm	$s3,16,$acc5
147	stb	$acc1,9($out)
148	_srm	$s3,8,$acc6
149	stb	$acc2,10($out)
150	stb	$s2,11($out)
151	stb	$acc4,12($out)
152	stb	$acc5,13($out)
153	stb	$acc6,14($out)
154	b	L\$enc_done
155	stb	$s3,15($out)
156
157L\$enc_out_aligned
158	stw	$s0,0($out)
159	stw	$s1,4($out)
160	stw	$s2,8($out)
161	stw	$s3,12($out)
162
163L\$enc_done
164	$POP	`-$FRAME-$SAVED_RP`(%sp),%r2	; standard epilogue
165	$POP	`-$FRAME+1*$SIZE_T`(%sp),%r4
166	$POP	`-$FRAME+2*$SIZE_T`(%sp),%r5
167	$POP	`-$FRAME+3*$SIZE_T`(%sp),%r6
168	$POP	`-$FRAME+4*$SIZE_T`(%sp),%r7
169	$POP	`-$FRAME+5*$SIZE_T`(%sp),%r8
170	$POP	`-$FRAME+6*$SIZE_T`(%sp),%r9
171	$POP	`-$FRAME+7*$SIZE_T`(%sp),%r10
172	$POP	`-$FRAME+8*$SIZE_T`(%sp),%r11
173	$POP	`-$FRAME+9*$SIZE_T`(%sp),%r12
174	$POP	`-$FRAME+10*$SIZE_T`(%sp),%r13
175	$POP	`-$FRAME+11*$SIZE_T`(%sp),%r14
176	$POP	`-$FRAME+12*$SIZE_T`(%sp),%r15
177	$POP	`-$FRAME+13*$SIZE_T`(%sp),%r16
178	$POP	`-$FRAME+14*$SIZE_T`(%sp),%r17
179	$POP	`-$FRAME+15*$SIZE_T`(%sp),%r18
180	bv	(%r2)
181	.EXIT
182	$POPMB	-$FRAME(%sp),%r3
183	.PROCEND
184
185	.ALIGN	16
186_parisc_AES_encrypt
187	.PROC
188	.CALLINFO	MILLICODE
189	.ENTRY
190	ldw	240($key),$rounds
191	ldw	0($key),$t0
192	ldw	4($key),$t1
193	ldw	8($key),$t2
194	_srm	$rounds,1,$rounds
195	xor	$t0,$s0,$s0
196	ldw	12($key),$t3
197	_srm	$s0,24,$acc0
198	xor	$t1,$s1,$s1
199	ldw	16($key),$t0
200	_srm	$s1,16,$acc1
201	xor	$t2,$s2,$s2
202	ldw	20($key),$t1
203	xor	$t3,$s3,$s3
204	ldw	24($key),$t2
205	ldw	28($key),$t3
206L\$enc_loop
207	_srm	$s2,8,$acc2
208	ldwx,s	$acc0($tbl),$acc0
209	_srm	$s3,0,$acc3
210	ldwx,s	$acc1($tbl),$acc1
211	_srm	$s1,24,$acc4
212	ldwx,s	$acc2($tbl),$acc2
213	_srm	$s2,16,$acc5
214	ldwx,s	$acc3($tbl),$acc3
215	_srm	$s3,8,$acc6
216	ldwx,s	$acc4($tbl),$acc4
217	_srm	$s0,0,$acc7
218	ldwx,s	$acc5($tbl),$acc5
219	_srm	$s2,24,$acc8
220	ldwx,s	$acc6($tbl),$acc6
221	_srm	$s3,16,$acc9
222	ldwx,s	$acc7($tbl),$acc7
223	_srm	$s0,8,$acc10
224	ldwx,s	$acc8($tbl),$acc8
225	_srm	$s1,0,$acc11
226	ldwx,s	$acc9($tbl),$acc9
227	_srm	$s3,24,$acc12
228	ldwx,s	$acc10($tbl),$acc10
229	_srm	$s0,16,$acc13
230	ldwx,s	$acc11($tbl),$acc11
231	_srm	$s1,8,$acc14
232	ldwx,s	$acc12($tbl),$acc12
233	_srm	$s2,0,$acc15
234	ldwx,s	$acc13($tbl),$acc13
235	ldwx,s	$acc14($tbl),$acc14
236	ldwx,s	$acc15($tbl),$acc15
237	addib,= -1,$rounds,L\$enc_last
238	ldo	32($key),$key
239
240		_ror	$acc1,8,$acc1
241		xor	$acc0,$t0,$t0
242	ldw	0($key),$s0
243		_ror	$acc2,16,$acc2
244		xor	$acc1,$t0,$t0
245	ldw	4($key),$s1
246		_ror	$acc3,24,$acc3
247		xor	$acc2,$t0,$t0
248	ldw	8($key),$s2
249		_ror	$acc5,8,$acc5
250		xor	$acc3,$t0,$t0
251	ldw	12($key),$s3
252		_ror	$acc6,16,$acc6
253		xor	$acc4,$t1,$t1
254		_ror	$acc7,24,$acc7
255		xor	$acc5,$t1,$t1
256		_ror	$acc9,8,$acc9
257		xor	$acc6,$t1,$t1
258		_ror	$acc10,16,$acc10
259		xor	$acc7,$t1,$t1
260		_ror	$acc11,24,$acc11
261		xor	$acc8,$t2,$t2
262		_ror	$acc13,8,$acc13
263		xor	$acc9,$t2,$t2
264		_ror	$acc14,16,$acc14
265		xor	$acc10,$t2,$t2
266		_ror	$acc15,24,$acc15
267		xor	$acc11,$t2,$t2
268		xor	$acc12,$acc14,$acc14
269		xor	$acc13,$t3,$t3
270	_srm	$t0,24,$acc0
271		xor	$acc14,$t3,$t3
272	_srm	$t1,16,$acc1
273		xor	$acc15,$t3,$t3
274
275	_srm	$t2,8,$acc2
276	ldwx,s	$acc0($tbl),$acc0
277	_srm	$t3,0,$acc3
278	ldwx,s	$acc1($tbl),$acc1
279	_srm	$t1,24,$acc4
280	ldwx,s	$acc2($tbl),$acc2
281	_srm	$t2,16,$acc5
282	ldwx,s	$acc3($tbl),$acc3
283	_srm	$t3,8,$acc6
284	ldwx,s	$acc4($tbl),$acc4
285	_srm	$t0,0,$acc7
286	ldwx,s	$acc5($tbl),$acc5
287	_srm	$t2,24,$acc8
288	ldwx,s	$acc6($tbl),$acc6
289	_srm	$t3,16,$acc9
290	ldwx,s	$acc7($tbl),$acc7
291	_srm	$t0,8,$acc10
292	ldwx,s	$acc8($tbl),$acc8
293	_srm	$t1,0,$acc11
294	ldwx,s	$acc9($tbl),$acc9
295	_srm	$t3,24,$acc12
296	ldwx,s	$acc10($tbl),$acc10
297	_srm	$t0,16,$acc13
298	ldwx,s	$acc11($tbl),$acc11
299	_srm	$t1,8,$acc14
300	ldwx,s	$acc12($tbl),$acc12
301	_srm	$t2,0,$acc15
302	ldwx,s	$acc13($tbl),$acc13
303		_ror	$acc1,8,$acc1
304	ldwx,s	$acc14($tbl),$acc14
305
306		_ror	$acc2,16,$acc2
307		xor	$acc0,$s0,$s0
308	ldwx,s	$acc15($tbl),$acc15
309		_ror	$acc3,24,$acc3
310		xor	$acc1,$s0,$s0
311	ldw	16($key),$t0
312		_ror	$acc5,8,$acc5
313		xor	$acc2,$s0,$s0
314	ldw	20($key),$t1
315		_ror	$acc6,16,$acc6
316		xor	$acc3,$s0,$s0
317	ldw	24($key),$t2
318		_ror	$acc7,24,$acc7
319		xor	$acc4,$s1,$s1
320	ldw	28($key),$t3
321		_ror	$acc9,8,$acc9
322		xor	$acc5,$s1,$s1
323	ldw	1024+0($tbl),%r0		; prefetch te4
324		_ror	$acc10,16,$acc10
325		xor	$acc6,$s1,$s1
326	ldw	1024+32($tbl),%r0		; prefetch te4
327		_ror	$acc11,24,$acc11
328		xor	$acc7,$s1,$s1
329	ldw	1024+64($tbl),%r0		; prefetch te4
330		_ror	$acc13,8,$acc13
331		xor	$acc8,$s2,$s2
332	ldw	1024+96($tbl),%r0		; prefetch te4
333		_ror	$acc14,16,$acc14
334		xor	$acc9,$s2,$s2
335	ldw	1024+128($tbl),%r0		; prefetch te4
336		_ror	$acc15,24,$acc15
337		xor	$acc10,$s2,$s2
338	ldw	1024+160($tbl),%r0		; prefetch te4
339	_srm	$s0,24,$acc0
340		xor	$acc11,$s2,$s2
341	ldw	1024+192($tbl),%r0		; prefetch te4
342		xor	$acc12,$acc14,$acc14
343		xor	$acc13,$s3,$s3
344	ldw	1024+224($tbl),%r0		; prefetch te4
345	_srm	$s1,16,$acc1
346		xor	$acc14,$s3,$s3
347	b	L\$enc_loop
348		xor	$acc15,$s3,$s3
349
350	.ALIGN	16
351L\$enc_last
352	ldo	1024($tbl),$rounds
353		_ror	$acc1,8,$acc1
354		xor	$acc0,$t0,$t0
355	ldw	0($key),$s0
356		_ror	$acc2,16,$acc2
357		xor	$acc1,$t0,$t0
358	ldw	4($key),$s1
359		_ror	$acc3,24,$acc3
360		xor	$acc2,$t0,$t0
361	ldw	8($key),$s2
362		_ror	$acc5,8,$acc5
363		xor	$acc3,$t0,$t0
364	ldw	12($key),$s3
365		_ror	$acc6,16,$acc6
366		xor	$acc4,$t1,$t1
367		_ror	$acc7,24,$acc7
368		xor	$acc5,$t1,$t1
369		_ror	$acc9,8,$acc9
370		xor	$acc6,$t1,$t1
371		_ror	$acc10,16,$acc10
372		xor	$acc7,$t1,$t1
373		_ror	$acc11,24,$acc11
374		xor	$acc8,$t2,$t2
375		_ror	$acc13,8,$acc13
376		xor	$acc9,$t2,$t2
377		_ror	$acc14,16,$acc14
378		xor	$acc10,$t2,$t2
379		_ror	$acc15,24,$acc15
380		xor	$acc11,$t2,$t2
381		xor	$acc12,$acc14,$acc14
382		xor	$acc13,$t3,$t3
383	_srm	$t0,24,$acc0
384		xor	$acc14,$t3,$t3
385	_srm	$t1,16,$acc1
386		xor	$acc15,$t3,$t3
387
388	_srm	$t2,8,$acc2
389	ldbx	$acc0($rounds),$acc0
390	_srm	$t1,24,$acc4
391	ldbx	$acc1($rounds),$acc1
392	_srm	$t2,16,$acc5
393	_srm	$t3,0,$acc3
394	ldbx	$acc2($rounds),$acc2
395	ldbx	$acc3($rounds),$acc3
396	_srm	$t3,8,$acc6
397	ldbx	$acc4($rounds),$acc4
398	_srm	$t2,24,$acc8
399	ldbx	$acc5($rounds),$acc5
400	_srm	$t3,16,$acc9
401	_srm	$t0,0,$acc7
402	ldbx	$acc6($rounds),$acc6
403	ldbx	$acc7($rounds),$acc7
404	_srm	$t0,8,$acc10
405	ldbx	$acc8($rounds),$acc8
406	_srm	$t3,24,$acc12
407	ldbx	$acc9($rounds),$acc9
408	_srm	$t0,16,$acc13
409	_srm	$t1,0,$acc11
410	ldbx	$acc10($rounds),$acc10
411	_srm	$t1,8,$acc14
412	ldbx	$acc11($rounds),$acc11
413	ldbx	$acc12($rounds),$acc12
414	ldbx	$acc13($rounds),$acc13
415	_srm	$t2,0,$acc15
416	ldbx	$acc14($rounds),$acc14
417
418		dep	$acc0,7,8,$acc3
419	ldbx	$acc15($rounds),$acc15
420		dep	$acc4,7,8,$acc7
421		dep	$acc1,15,8,$acc3
422		dep	$acc5,15,8,$acc7
423		dep	$acc2,23,8,$acc3
424		dep	$acc6,23,8,$acc7
425		xor	$acc3,$s0,$s0
426		xor	$acc7,$s1,$s1
427		dep	$acc8,7,8,$acc11
428		dep	$acc12,7,8,$acc15
429		dep	$acc9,15,8,$acc11
430		dep	$acc13,15,8,$acc15
431		dep	$acc10,23,8,$acc11
432		dep	$acc14,23,8,$acc15
433		xor	$acc11,$s2,$s2
434
435	bv	(%r31)
436	.EXIT
437		xor	$acc15,$s3,$s3
438	.PROCEND
439
440	.section .rodata
441	.ALIGN	64
442L\$AES_Te
443	.WORD	0xc66363a5, 0xf87c7c84, 0xee777799, 0xf67b7b8d
444	.WORD	0xfff2f20d, 0xd66b6bbd, 0xde6f6fb1, 0x91c5c554
445	.WORD	0x60303050, 0x02010103, 0xce6767a9, 0x562b2b7d
446	.WORD	0xe7fefe19, 0xb5d7d762, 0x4dababe6, 0xec76769a
447	.WORD	0x8fcaca45, 0x1f82829d, 0x89c9c940, 0xfa7d7d87
448	.WORD	0xeffafa15, 0xb25959eb, 0x8e4747c9, 0xfbf0f00b
449	.WORD	0x41adadec, 0xb3d4d467, 0x5fa2a2fd, 0x45afafea
450	.WORD	0x239c9cbf, 0x53a4a4f7, 0xe4727296, 0x9bc0c05b
451	.WORD	0x75b7b7c2, 0xe1fdfd1c, 0x3d9393ae, 0x4c26266a
452	.WORD	0x6c36365a, 0x7e3f3f41, 0xf5f7f702, 0x83cccc4f
453	.WORD	0x6834345c, 0x51a5a5f4, 0xd1e5e534, 0xf9f1f108
454	.WORD	0xe2717193, 0xabd8d873, 0x62313153, 0x2a15153f
455	.WORD	0x0804040c, 0x95c7c752, 0x46232365, 0x9dc3c35e
456	.WORD	0x30181828, 0x379696a1, 0x0a05050f, 0x2f9a9ab5
457	.WORD	0x0e070709, 0x24121236, 0x1b80809b, 0xdfe2e23d
458	.WORD	0xcdebeb26, 0x4e272769, 0x7fb2b2cd, 0xea75759f
459	.WORD	0x1209091b, 0x1d83839e, 0x582c2c74, 0x341a1a2e
460	.WORD	0x361b1b2d, 0xdc6e6eb2, 0xb45a5aee, 0x5ba0a0fb
461	.WORD	0xa45252f6, 0x763b3b4d, 0xb7d6d661, 0x7db3b3ce
462	.WORD	0x5229297b, 0xdde3e33e, 0x5e2f2f71, 0x13848497
463	.WORD	0xa65353f5, 0xb9d1d168, 0x00000000, 0xc1eded2c
464	.WORD	0x40202060, 0xe3fcfc1f, 0x79b1b1c8, 0xb65b5bed
465	.WORD	0xd46a6abe, 0x8dcbcb46, 0x67bebed9, 0x7239394b
466	.WORD	0x944a4ade, 0x984c4cd4, 0xb05858e8, 0x85cfcf4a
467	.WORD	0xbbd0d06b, 0xc5efef2a, 0x4faaaae5, 0xedfbfb16
468	.WORD	0x864343c5, 0x9a4d4dd7, 0x66333355, 0x11858594
469	.WORD	0x8a4545cf, 0xe9f9f910, 0x04020206, 0xfe7f7f81
470	.WORD	0xa05050f0, 0x783c3c44, 0x259f9fba, 0x4ba8a8e3
471	.WORD	0xa25151f3, 0x5da3a3fe, 0x804040c0, 0x058f8f8a
472	.WORD	0x3f9292ad, 0x219d9dbc, 0x70383848, 0xf1f5f504
473	.WORD	0x63bcbcdf, 0x77b6b6c1, 0xafdada75, 0x42212163
474	.WORD	0x20101030, 0xe5ffff1a, 0xfdf3f30e, 0xbfd2d26d
475	.WORD	0x81cdcd4c, 0x180c0c14, 0x26131335, 0xc3ecec2f
476	.WORD	0xbe5f5fe1, 0x359797a2, 0x884444cc, 0x2e171739
477	.WORD	0x93c4c457, 0x55a7a7f2, 0xfc7e7e82, 0x7a3d3d47
478	.WORD	0xc86464ac, 0xba5d5de7, 0x3219192b, 0xe6737395
479	.WORD	0xc06060a0, 0x19818198, 0x9e4f4fd1, 0xa3dcdc7f
480	.WORD	0x44222266, 0x542a2a7e, 0x3b9090ab, 0x0b888883
481	.WORD	0x8c4646ca, 0xc7eeee29, 0x6bb8b8d3, 0x2814143c
482	.WORD	0xa7dede79, 0xbc5e5ee2, 0x160b0b1d, 0xaddbdb76
483	.WORD	0xdbe0e03b, 0x64323256, 0x743a3a4e, 0x140a0a1e
484	.WORD	0x924949db, 0x0c06060a, 0x4824246c, 0xb85c5ce4
485	.WORD	0x9fc2c25d, 0xbdd3d36e, 0x43acacef, 0xc46262a6
486	.WORD	0x399191a8, 0x319595a4, 0xd3e4e437, 0xf279798b
487	.WORD	0xd5e7e732, 0x8bc8c843, 0x6e373759, 0xda6d6db7
488	.WORD	0x018d8d8c, 0xb1d5d564, 0x9c4e4ed2, 0x49a9a9e0
489	.WORD	0xd86c6cb4, 0xac5656fa, 0xf3f4f407, 0xcfeaea25
490	.WORD	0xca6565af, 0xf47a7a8e, 0x47aeaee9, 0x10080818
491	.WORD	0x6fbabad5, 0xf0787888, 0x4a25256f, 0x5c2e2e72
492	.WORD	0x381c1c24, 0x57a6a6f1, 0x73b4b4c7, 0x97c6c651
493	.WORD	0xcbe8e823, 0xa1dddd7c, 0xe874749c, 0x3e1f1f21
494	.WORD	0x964b4bdd, 0x61bdbddc, 0x0d8b8b86, 0x0f8a8a85
495	.WORD	0xe0707090, 0x7c3e3e42, 0x71b5b5c4, 0xcc6666aa
496	.WORD	0x904848d8, 0x06030305, 0xf7f6f601, 0x1c0e0e12
497	.WORD	0xc26161a3, 0x6a35355f, 0xae5757f9, 0x69b9b9d0
498	.WORD	0x17868691, 0x99c1c158, 0x3a1d1d27, 0x279e9eb9
499	.WORD	0xd9e1e138, 0xebf8f813, 0x2b9898b3, 0x22111133
500	.WORD	0xd26969bb, 0xa9d9d970, 0x078e8e89, 0x339494a7
501	.WORD	0x2d9b9bb6, 0x3c1e1e22, 0x15878792, 0xc9e9e920
502	.WORD	0x87cece49, 0xaa5555ff, 0x50282878, 0xa5dfdf7a
503	.WORD	0x038c8c8f, 0x59a1a1f8, 0x09898980, 0x1a0d0d17
504	.WORD	0x65bfbfda, 0xd7e6e631, 0x844242c6, 0xd06868b8
505	.WORD	0x824141c3, 0x299999b0, 0x5a2d2d77, 0x1e0f0f11
506	.WORD	0x7bb0b0cb, 0xa85454fc, 0x6dbbbbd6, 0x2c16163a
507	.BYTE	0x63, 0x7c, 0x77, 0x7b, 0xf2, 0x6b, 0x6f, 0xc5
508	.BYTE	0x30, 0x01, 0x67, 0x2b, 0xfe, 0xd7, 0xab, 0x76
509	.BYTE	0xca, 0x82, 0xc9, 0x7d, 0xfa, 0x59, 0x47, 0xf0
510	.BYTE	0xad, 0xd4, 0xa2, 0xaf, 0x9c, 0xa4, 0x72, 0xc0
511	.BYTE	0xb7, 0xfd, 0x93, 0x26, 0x36, 0x3f, 0xf7, 0xcc
512	.BYTE	0x34, 0xa5, 0xe5, 0xf1, 0x71, 0xd8, 0x31, 0x15
513	.BYTE	0x04, 0xc7, 0x23, 0xc3, 0x18, 0x96, 0x05, 0x9a
514	.BYTE	0x07, 0x12, 0x80, 0xe2, 0xeb, 0x27, 0xb2, 0x75
515	.BYTE	0x09, 0x83, 0x2c, 0x1a, 0x1b, 0x6e, 0x5a, 0xa0
516	.BYTE	0x52, 0x3b, 0xd6, 0xb3, 0x29, 0xe3, 0x2f, 0x84
517	.BYTE	0x53, 0xd1, 0x00, 0xed, 0x20, 0xfc, 0xb1, 0x5b
518	.BYTE	0x6a, 0xcb, 0xbe, 0x39, 0x4a, 0x4c, 0x58, 0xcf
519	.BYTE	0xd0, 0xef, 0xaa, 0xfb, 0x43, 0x4d, 0x33, 0x85
520	.BYTE	0x45, 0xf9, 0x02, 0x7f, 0x50, 0x3c, 0x9f, 0xa8
521	.BYTE	0x51, 0xa3, 0x40, 0x8f, 0x92, 0x9d, 0x38, 0xf5
522	.BYTE	0xbc, 0xb6, 0xda, 0x21, 0x10, 0xff, 0xf3, 0xd2
523	.BYTE	0xcd, 0x0c, 0x13, 0xec, 0x5f, 0x97, 0x44, 0x17
524	.BYTE	0xc4, 0xa7, 0x7e, 0x3d, 0x64, 0x5d, 0x19, 0x73
525	.BYTE	0x60, 0x81, 0x4f, 0xdc, 0x22, 0x2a, 0x90, 0x88
526	.BYTE	0x46, 0xee, 0xb8, 0x14, 0xde, 0x5e, 0x0b, 0xdb
527	.BYTE	0xe0, 0x32, 0x3a, 0x0a, 0x49, 0x06, 0x24, 0x5c
528	.BYTE	0xc2, 0xd3, 0xac, 0x62, 0x91, 0x95, 0xe4, 0x79
529	.BYTE	0xe7, 0xc8, 0x37, 0x6d, 0x8d, 0xd5, 0x4e, 0xa9
530	.BYTE	0x6c, 0x56, 0xf4, 0xea, 0x65, 0x7a, 0xae, 0x08
531	.BYTE	0xba, 0x78, 0x25, 0x2e, 0x1c, 0xa6, 0xb4, 0xc6
532	.BYTE	0xe8, 0xdd, 0x74, 0x1f, 0x4b, 0xbd, 0x8b, 0x8a
533	.BYTE	0x70, 0x3e, 0xb5, 0x66, 0x48, 0x03, 0xf6, 0x0e
534	.BYTE	0x61, 0x35, 0x57, 0xb9, 0x86, 0xc1, 0x1d, 0x9e
535	.BYTE	0xe1, 0xf8, 0x98, 0x11, 0x69, 0xd9, 0x8e, 0x94
536	.BYTE	0x9b, 0x1e, 0x87, 0xe9, 0xce, 0x55, 0x28, 0xdf
537	.BYTE	0x8c, 0xa1, 0x89, 0x0d, 0xbf, 0xe6, 0x42, 0x68
538	.BYTE	0x41, 0x99, 0x2d, 0x0f, 0xb0, 0x54, 0xbb, 0x16
539	.previous
540___
541
542$code.=<<___;
543	.EXPORT	aes_decrypt_internal,ENTRY,ARGW0=GR,ARGW1=GR,ARGW2=GR
544	.ALIGN	16
545aes_decrypt_internal
546	.PROC
547	.CALLINFO	FRAME=`$FRAME-16*$SIZE_T`,NO_CALLS,SAVE_RP,ENTRY_GR=18
548	.ENTRY
549	$PUSH	%r2,-$SAVED_RP(%sp)	; standard prologue
550	$PUSHMA	%r3,$FRAME(%sp)
551	$PUSH	%r4,`-$FRAME+1*$SIZE_T`(%sp)
552	$PUSH	%r5,`-$FRAME+2*$SIZE_T`(%sp)
553	$PUSH	%r6,`-$FRAME+3*$SIZE_T`(%sp)
554	$PUSH	%r7,`-$FRAME+4*$SIZE_T`(%sp)
555	$PUSH	%r8,`-$FRAME+5*$SIZE_T`(%sp)
556	$PUSH	%r9,`-$FRAME+6*$SIZE_T`(%sp)
557	$PUSH	%r10,`-$FRAME+7*$SIZE_T`(%sp)
558	$PUSH	%r11,`-$FRAME+8*$SIZE_T`(%sp)
559	$PUSH	%r12,`-$FRAME+9*$SIZE_T`(%sp)
560	$PUSH	%r13,`-$FRAME+10*$SIZE_T`(%sp)
561	$PUSH	%r14,`-$FRAME+11*$SIZE_T`(%sp)
562	$PUSH	%r15,`-$FRAME+12*$SIZE_T`(%sp)
563	$PUSH	%r16,`-$FRAME+13*$SIZE_T`(%sp)
564	$PUSH	%r17,`-$FRAME+14*$SIZE_T`(%sp)
565	$PUSH	%r18,`-$FRAME+15*$SIZE_T`(%sp)
566
567	ldi	3,$t0
568#ifdef __PIC__
569	addil	LT'L\$AES_Td, %r19
570	ldw	RT'L\$AES_Td(%r1), $tbl
571#else
572	ldil	L'L\$AES_Td, %t1
573	ldo	R'L\$AES_Td(%t1), $tbl
574#endif
575
576	and	$inp,$t0,$t0
577	sub	$inp,$t0,$inp
578	ldw	0($inp),$s0
579	ldw	4($inp),$s1
580	ldw	8($inp),$s2
581	comib,=	0,$t0,L\$dec_inp_aligned
582	ldw	12($inp),$s3
583
584	sh3addl	$t0,%r0,$t0
585	subi	32,$t0,$t0
586	mtctl	$t0,%cr11
587	ldw	16($inp),$t1
588	vshd	$s0,$s1,$s0
589	vshd	$s1,$s2,$s1
590	vshd	$s2,$s3,$s2
591	vshd	$s3,$t1,$s3
592
593L\$dec_inp_aligned
594	bl	_parisc_AES_decrypt,%r31
595	nop
596
597	extru,<> $out,31,2,%r0
598	b	L\$dec_out_aligned
599	nop
600
601	_srm	$s0,24,$acc0
602	_srm	$s0,16,$acc1
603	stb	$acc0,0($out)
604	_srm	$s0,8,$acc2
605	stb	$acc1,1($out)
606	_srm	$s1,24,$acc4
607	stb	$acc2,2($out)
608	_srm	$s1,16,$acc5
609	stb	$s0,3($out)
610	_srm	$s1,8,$acc6
611	stb	$acc4,4($out)
612	_srm	$s2,24,$acc0
613	stb	$acc5,5($out)
614	_srm	$s2,16,$acc1
615	stb	$acc6,6($out)
616	_srm	$s2,8,$acc2
617	stb	$s1,7($out)
618	_srm	$s3,24,$acc4
619	stb	$acc0,8($out)
620	_srm	$s3,16,$acc5
621	stb	$acc1,9($out)
622	_srm	$s3,8,$acc6
623	stb	$acc2,10($out)
624	stb	$s2,11($out)
625	stb	$acc4,12($out)
626	stb	$acc5,13($out)
627	stb	$acc6,14($out)
628	b	L\$dec_done
629	stb	$s3,15($out)
630
631L\$dec_out_aligned
632	stw	$s0,0($out)
633	stw	$s1,4($out)
634	stw	$s2,8($out)
635	stw	$s3,12($out)
636
637L\$dec_done
638	$POP	`-$FRAME-$SAVED_RP`(%sp),%r2	; standard epilogue
639	$POP	`-$FRAME+1*$SIZE_T`(%sp),%r4
640	$POP	`-$FRAME+2*$SIZE_T`(%sp),%r5
641	$POP	`-$FRAME+3*$SIZE_T`(%sp),%r6
642	$POP	`-$FRAME+4*$SIZE_T`(%sp),%r7
643	$POP	`-$FRAME+5*$SIZE_T`(%sp),%r8
644	$POP	`-$FRAME+6*$SIZE_T`(%sp),%r9
645	$POP	`-$FRAME+7*$SIZE_T`(%sp),%r10
646	$POP	`-$FRAME+8*$SIZE_T`(%sp),%r11
647	$POP	`-$FRAME+9*$SIZE_T`(%sp),%r12
648	$POP	`-$FRAME+10*$SIZE_T`(%sp),%r13
649	$POP	`-$FRAME+11*$SIZE_T`(%sp),%r14
650	$POP	`-$FRAME+12*$SIZE_T`(%sp),%r15
651	$POP	`-$FRAME+13*$SIZE_T`(%sp),%r16
652	$POP	`-$FRAME+14*$SIZE_T`(%sp),%r17
653	$POP	`-$FRAME+15*$SIZE_T`(%sp),%r18
654	bv	(%r2)
655	.EXIT
656	$POPMB	-$FRAME(%sp),%r3
657	.PROCEND
658
659	.ALIGN	16
660_parisc_AES_decrypt
661	.PROC
662	.CALLINFO	MILLICODE
663	.ENTRY
664	ldw	240($key),$rounds
665	ldw	0($key),$t0
666	ldw	4($key),$t1
667	ldw	8($key),$t2
668	ldw	12($key),$t3
669	_srm	$rounds,1,$rounds
670	xor	$t0,$s0,$s0
671	ldw	16($key),$t0
672	xor	$t1,$s1,$s1
673	ldw	20($key),$t1
674	_srm	$s0,24,$acc0
675	xor	$t2,$s2,$s2
676	ldw	24($key),$t2
677	xor	$t3,$s3,$s3
678	ldw	28($key),$t3
679	_srm	$s3,16,$acc1
680L\$dec_loop
681	_srm	$s2,8,$acc2
682	ldwx,s	$acc0($tbl),$acc0
683	_srm	$s1,0,$acc3
684	ldwx,s	$acc1($tbl),$acc1
685	_srm	$s1,24,$acc4
686	ldwx,s	$acc2($tbl),$acc2
687	_srm	$s0,16,$acc5
688	ldwx,s	$acc3($tbl),$acc3
689	_srm	$s3,8,$acc6
690	ldwx,s	$acc4($tbl),$acc4
691	_srm	$s2,0,$acc7
692	ldwx,s	$acc5($tbl),$acc5
693	_srm	$s2,24,$acc8
694	ldwx,s	$acc6($tbl),$acc6
695	_srm	$s1,16,$acc9
696	ldwx,s	$acc7($tbl),$acc7
697	_srm	$s0,8,$acc10
698	ldwx,s	$acc8($tbl),$acc8
699	_srm	$s3,0,$acc11
700	ldwx,s	$acc9($tbl),$acc9
701	_srm	$s3,24,$acc12
702	ldwx,s	$acc10($tbl),$acc10
703	_srm	$s2,16,$acc13
704	ldwx,s	$acc11($tbl),$acc11
705	_srm	$s1,8,$acc14
706	ldwx,s	$acc12($tbl),$acc12
707	_srm	$s0,0,$acc15
708	ldwx,s	$acc13($tbl),$acc13
709	ldwx,s	$acc14($tbl),$acc14
710	ldwx,s	$acc15($tbl),$acc15
711	addib,= -1,$rounds,L\$dec_last
712	ldo	32($key),$key
713
714		_ror	$acc1,8,$acc1
715		xor	$acc0,$t0,$t0
716	ldw	0($key),$s0
717		_ror	$acc2,16,$acc2
718		xor	$acc1,$t0,$t0
719	ldw	4($key),$s1
720		_ror	$acc3,24,$acc3
721		xor	$acc2,$t0,$t0
722	ldw	8($key),$s2
723		_ror	$acc5,8,$acc5
724		xor	$acc3,$t0,$t0
725	ldw	12($key),$s3
726		_ror	$acc6,16,$acc6
727		xor	$acc4,$t1,$t1
728		_ror	$acc7,24,$acc7
729		xor	$acc5,$t1,$t1
730		_ror	$acc9,8,$acc9
731		xor	$acc6,$t1,$t1
732		_ror	$acc10,16,$acc10
733		xor	$acc7,$t1,$t1
734		_ror	$acc11,24,$acc11
735		xor	$acc8,$t2,$t2
736		_ror	$acc13,8,$acc13
737		xor	$acc9,$t2,$t2
738		_ror	$acc14,16,$acc14
739		xor	$acc10,$t2,$t2
740		_ror	$acc15,24,$acc15
741		xor	$acc11,$t2,$t2
742		xor	$acc12,$acc14,$acc14
743		xor	$acc13,$t3,$t3
744	_srm	$t0,24,$acc0
745		xor	$acc14,$t3,$t3
746		xor	$acc15,$t3,$t3
747	_srm	$t3,16,$acc1
748
749	_srm	$t2,8,$acc2
750	ldwx,s	$acc0($tbl),$acc0
751	_srm	$t1,0,$acc3
752	ldwx,s	$acc1($tbl),$acc1
753	_srm	$t1,24,$acc4
754	ldwx,s	$acc2($tbl),$acc2
755	_srm	$t0,16,$acc5
756	ldwx,s	$acc3($tbl),$acc3
757	_srm	$t3,8,$acc6
758	ldwx,s	$acc4($tbl),$acc4
759	_srm	$t2,0,$acc7
760	ldwx,s	$acc5($tbl),$acc5
761	_srm	$t2,24,$acc8
762	ldwx,s	$acc6($tbl),$acc6
763	_srm	$t1,16,$acc9
764	ldwx,s	$acc7($tbl),$acc7
765	_srm	$t0,8,$acc10
766	ldwx,s	$acc8($tbl),$acc8
767	_srm	$t3,0,$acc11
768	ldwx,s	$acc9($tbl),$acc9
769	_srm	$t3,24,$acc12
770	ldwx,s	$acc10($tbl),$acc10
771	_srm	$t2,16,$acc13
772	ldwx,s	$acc11($tbl),$acc11
773	_srm	$t1,8,$acc14
774	ldwx,s	$acc12($tbl),$acc12
775	_srm	$t0,0,$acc15
776	ldwx,s	$acc13($tbl),$acc13
777		_ror	$acc1,8,$acc1
778	ldwx,s	$acc14($tbl),$acc14
779
780		_ror	$acc2,16,$acc2
781		xor	$acc0,$s0,$s0
782	ldwx,s	$acc15($tbl),$acc15
783		_ror	$acc3,24,$acc3
784		xor	$acc1,$s0,$s0
785	ldw	16($key),$t0
786		_ror	$acc5,8,$acc5
787		xor	$acc2,$s0,$s0
788	ldw	20($key),$t1
789		_ror	$acc6,16,$acc6
790		xor	$acc3,$s0,$s0
791	ldw	24($key),$t2
792		_ror	$acc7,24,$acc7
793		xor	$acc4,$s1,$s1
794	ldw	28($key),$t3
795		_ror	$acc9,8,$acc9
796		xor	$acc5,$s1,$s1
797	ldw	1024+0($tbl),%r0		; prefetch td4
798		_ror	$acc10,16,$acc10
799		xor	$acc6,$s1,$s1
800	ldw	1024+32($tbl),%r0		; prefetch td4
801		_ror	$acc11,24,$acc11
802		xor	$acc7,$s1,$s1
803	ldw	1024+64($tbl),%r0		; prefetch td4
804		_ror	$acc13,8,$acc13
805		xor	$acc8,$s2,$s2
806	ldw	1024+96($tbl),%r0		; prefetch td4
807		_ror	$acc14,16,$acc14
808		xor	$acc9,$s2,$s2
809	ldw	1024+128($tbl),%r0		; prefetch td4
810		_ror	$acc15,24,$acc15
811		xor	$acc10,$s2,$s2
812	ldw	1024+160($tbl),%r0		; prefetch td4
813	_srm	$s0,24,$acc0
814		xor	$acc11,$s2,$s2
815	ldw	1024+192($tbl),%r0		; prefetch td4
816		xor	$acc12,$acc14,$acc14
817		xor	$acc13,$s3,$s3
818	ldw	1024+224($tbl),%r0		; prefetch td4
819		xor	$acc14,$s3,$s3
820		xor	$acc15,$s3,$s3
821	b	L\$dec_loop
822	_srm	$s3,16,$acc1
823
824	.ALIGN	16
825L\$dec_last
826	ldo	1024($tbl),$rounds
827		_ror	$acc1,8,$acc1
828		xor	$acc0,$t0,$t0
829	ldw	0($key),$s0
830		_ror	$acc2,16,$acc2
831		xor	$acc1,$t0,$t0
832	ldw	4($key),$s1
833		_ror	$acc3,24,$acc3
834		xor	$acc2,$t0,$t0
835	ldw	8($key),$s2
836		_ror	$acc5,8,$acc5
837		xor	$acc3,$t0,$t0
838	ldw	12($key),$s3
839		_ror	$acc6,16,$acc6
840		xor	$acc4,$t1,$t1
841		_ror	$acc7,24,$acc7
842		xor	$acc5,$t1,$t1
843		_ror	$acc9,8,$acc9
844		xor	$acc6,$t1,$t1
845		_ror	$acc10,16,$acc10
846		xor	$acc7,$t1,$t1
847		_ror	$acc11,24,$acc11
848		xor	$acc8,$t2,$t2
849		_ror	$acc13,8,$acc13
850		xor	$acc9,$t2,$t2
851		_ror	$acc14,16,$acc14
852		xor	$acc10,$t2,$t2
853		_ror	$acc15,24,$acc15
854		xor	$acc11,$t2,$t2
855		xor	$acc12,$acc14,$acc14
856		xor	$acc13,$t3,$t3
857	_srm	$t0,24,$acc0
858		xor	$acc14,$t3,$t3
859		xor	$acc15,$t3,$t3
860	_srm	$t3,16,$acc1
861
862	_srm	$t2,8,$acc2
863	ldbx	$acc0($rounds),$acc0
864	_srm	$t1,24,$acc4
865	ldbx	$acc1($rounds),$acc1
866	_srm	$t0,16,$acc5
867	_srm	$t1,0,$acc3
868	ldbx	$acc2($rounds),$acc2
869	ldbx	$acc3($rounds),$acc3
870	_srm	$t3,8,$acc6
871	ldbx	$acc4($rounds),$acc4
872	_srm	$t2,24,$acc8
873	ldbx	$acc5($rounds),$acc5
874	_srm	$t1,16,$acc9
875	_srm	$t2,0,$acc7
876	ldbx	$acc6($rounds),$acc6
877	ldbx	$acc7($rounds),$acc7
878	_srm	$t0,8,$acc10
879	ldbx	$acc8($rounds),$acc8
880	_srm	$t3,24,$acc12
881	ldbx	$acc9($rounds),$acc9
882	_srm	$t2,16,$acc13
883	_srm	$t3,0,$acc11
884	ldbx	$acc10($rounds),$acc10
885	_srm	$t1,8,$acc14
886	ldbx	$acc11($rounds),$acc11
887	ldbx	$acc12($rounds),$acc12
888	ldbx	$acc13($rounds),$acc13
889	_srm	$t0,0,$acc15
890	ldbx	$acc14($rounds),$acc14
891
892		dep	$acc0,7,8,$acc3
893	ldbx	$acc15($rounds),$acc15
894		dep	$acc4,7,8,$acc7
895		dep	$acc1,15,8,$acc3
896		dep	$acc5,15,8,$acc7
897		dep	$acc2,23,8,$acc3
898		dep	$acc6,23,8,$acc7
899		xor	$acc3,$s0,$s0
900		xor	$acc7,$s1,$s1
901		dep	$acc8,7,8,$acc11
902		dep	$acc12,7,8,$acc15
903		dep	$acc9,15,8,$acc11
904		dep	$acc13,15,8,$acc15
905		dep	$acc10,23,8,$acc11
906		dep	$acc14,23,8,$acc15
907		xor	$acc11,$s2,$s2
908
909	bv	(%r31)
910	.EXIT
911		xor	$acc15,$s3,$s3
912	.PROCEND
913
914	.section .rodata
915	.ALIGN	64
916L\$AES_Td
917	.WORD	0x51f4a750, 0x7e416553, 0x1a17a4c3, 0x3a275e96
918	.WORD	0x3bab6bcb, 0x1f9d45f1, 0xacfa58ab, 0x4be30393
919	.WORD	0x2030fa55, 0xad766df6, 0x88cc7691, 0xf5024c25
920	.WORD	0x4fe5d7fc, 0xc52acbd7, 0x26354480, 0xb562a38f
921	.WORD	0xdeb15a49, 0x25ba1b67, 0x45ea0e98, 0x5dfec0e1
922	.WORD	0xc32f7502, 0x814cf012, 0x8d4697a3, 0x6bd3f9c6
923	.WORD	0x038f5fe7, 0x15929c95, 0xbf6d7aeb, 0x955259da
924	.WORD	0xd4be832d, 0x587421d3, 0x49e06929, 0x8ec9c844
925	.WORD	0x75c2896a, 0xf48e7978, 0x99583e6b, 0x27b971dd
926	.WORD	0xbee14fb6, 0xf088ad17, 0xc920ac66, 0x7dce3ab4
927	.WORD	0x63df4a18, 0xe51a3182, 0x97513360, 0x62537f45
928	.WORD	0xb16477e0, 0xbb6bae84, 0xfe81a01c, 0xf9082b94
929	.WORD	0x70486858, 0x8f45fd19, 0x94de6c87, 0x527bf8b7
930	.WORD	0xab73d323, 0x724b02e2, 0xe31f8f57, 0x6655ab2a
931	.WORD	0xb2eb2807, 0x2fb5c203, 0x86c57b9a, 0xd33708a5
932	.WORD	0x302887f2, 0x23bfa5b2, 0x02036aba, 0xed16825c
933	.WORD	0x8acf1c2b, 0xa779b492, 0xf307f2f0, 0x4e69e2a1
934	.WORD	0x65daf4cd, 0x0605bed5, 0xd134621f, 0xc4a6fe8a
935	.WORD	0x342e539d, 0xa2f355a0, 0x058ae132, 0xa4f6eb75
936	.WORD	0x0b83ec39, 0x4060efaa, 0x5e719f06, 0xbd6e1051
937	.WORD	0x3e218af9, 0x96dd063d, 0xdd3e05ae, 0x4de6bd46
938	.WORD	0x91548db5, 0x71c45d05, 0x0406d46f, 0x605015ff
939	.WORD	0x1998fb24, 0xd6bde997, 0x894043cc, 0x67d99e77
940	.WORD	0xb0e842bd, 0x07898b88, 0xe7195b38, 0x79c8eedb
941	.WORD	0xa17c0a47, 0x7c420fe9, 0xf8841ec9, 0x00000000
942	.WORD	0x09808683, 0x322bed48, 0x1e1170ac, 0x6c5a724e
943	.WORD	0xfd0efffb, 0x0f853856, 0x3daed51e, 0x362d3927
944	.WORD	0x0a0fd964, 0x685ca621, 0x9b5b54d1, 0x24362e3a
945	.WORD	0x0c0a67b1, 0x9357e70f, 0xb4ee96d2, 0x1b9b919e
946	.WORD	0x80c0c54f, 0x61dc20a2, 0x5a774b69, 0x1c121a16
947	.WORD	0xe293ba0a, 0xc0a02ae5, 0x3c22e043, 0x121b171d
948	.WORD	0x0e090d0b, 0xf28bc7ad, 0x2db6a8b9, 0x141ea9c8
949	.WORD	0x57f11985, 0xaf75074c, 0xee99ddbb, 0xa37f60fd
950	.WORD	0xf701269f, 0x5c72f5bc, 0x44663bc5, 0x5bfb7e34
951	.WORD	0x8b432976, 0xcb23c6dc, 0xb6edfc68, 0xb8e4f163
952	.WORD	0xd731dcca, 0x42638510, 0x13972240, 0x84c61120
953	.WORD	0x854a247d, 0xd2bb3df8, 0xaef93211, 0xc729a16d
954	.WORD	0x1d9e2f4b, 0xdcb230f3, 0x0d8652ec, 0x77c1e3d0
955	.WORD	0x2bb3166c, 0xa970b999, 0x119448fa, 0x47e96422
956	.WORD	0xa8fc8cc4, 0xa0f03f1a, 0x567d2cd8, 0x223390ef
957	.WORD	0x87494ec7, 0xd938d1c1, 0x8ccaa2fe, 0x98d40b36
958	.WORD	0xa6f581cf, 0xa57ade28, 0xdab78e26, 0x3fadbfa4
959	.WORD	0x2c3a9de4, 0x5078920d, 0x6a5fcc9b, 0x547e4662
960	.WORD	0xf68d13c2, 0x90d8b8e8, 0x2e39f75e, 0x82c3aff5
961	.WORD	0x9f5d80be, 0x69d0937c, 0x6fd52da9, 0xcf2512b3
962	.WORD	0xc8ac993b, 0x10187da7, 0xe89c636e, 0xdb3bbb7b
963	.WORD	0xcd267809, 0x6e5918f4, 0xec9ab701, 0x834f9aa8
964	.WORD	0xe6956e65, 0xaaffe67e, 0x21bccf08, 0xef15e8e6
965	.WORD	0xbae79bd9, 0x4a6f36ce, 0xea9f09d4, 0x29b07cd6
966	.WORD	0x31a4b2af, 0x2a3f2331, 0xc6a59430, 0x35a266c0
967	.WORD	0x744ebc37, 0xfc82caa6, 0xe090d0b0, 0x33a7d815
968	.WORD	0xf104984a, 0x41ecdaf7, 0x7fcd500e, 0x1791f62f
969	.WORD	0x764dd68d, 0x43efb04d, 0xccaa4d54, 0xe49604df
970	.WORD	0x9ed1b5e3, 0x4c6a881b, 0xc12c1fb8, 0x4665517f
971	.WORD	0x9d5eea04, 0x018c355d, 0xfa877473, 0xfb0b412e
972	.WORD	0xb3671d5a, 0x92dbd252, 0xe9105633, 0x6dd64713
973	.WORD	0x9ad7618c, 0x37a10c7a, 0x59f8148e, 0xeb133c89
974	.WORD	0xcea927ee, 0xb761c935, 0xe11ce5ed, 0x7a47b13c
975	.WORD	0x9cd2df59, 0x55f2733f, 0x1814ce79, 0x73c737bf
976	.WORD	0x53f7cdea, 0x5ffdaa5b, 0xdf3d6f14, 0x7844db86
977	.WORD	0xcaaff381, 0xb968c43e, 0x3824342c, 0xc2a3405f
978	.WORD	0x161dc372, 0xbce2250c, 0x283c498b, 0xff0d9541
979	.WORD	0x39a80171, 0x080cb3de, 0xd8b4e49c, 0x6456c190
980	.WORD	0x7bcb8461, 0xd532b670, 0x486c5c74, 0xd0b85742
981	.BYTE	0x52, 0x09, 0x6a, 0xd5, 0x30, 0x36, 0xa5, 0x38
982	.BYTE	0xbf, 0x40, 0xa3, 0x9e, 0x81, 0xf3, 0xd7, 0xfb
983	.BYTE	0x7c, 0xe3, 0x39, 0x82, 0x9b, 0x2f, 0xff, 0x87
984	.BYTE	0x34, 0x8e, 0x43, 0x44, 0xc4, 0xde, 0xe9, 0xcb
985	.BYTE	0x54, 0x7b, 0x94, 0x32, 0xa6, 0xc2, 0x23, 0x3d
986	.BYTE	0xee, 0x4c, 0x95, 0x0b, 0x42, 0xfa, 0xc3, 0x4e
987	.BYTE	0x08, 0x2e, 0xa1, 0x66, 0x28, 0xd9, 0x24, 0xb2
988	.BYTE	0x76, 0x5b, 0xa2, 0x49, 0x6d, 0x8b, 0xd1, 0x25
989	.BYTE	0x72, 0xf8, 0xf6, 0x64, 0x86, 0x68, 0x98, 0x16
990	.BYTE	0xd4, 0xa4, 0x5c, 0xcc, 0x5d, 0x65, 0xb6, 0x92
991	.BYTE	0x6c, 0x70, 0x48, 0x50, 0xfd, 0xed, 0xb9, 0xda
992	.BYTE	0x5e, 0x15, 0x46, 0x57, 0xa7, 0x8d, 0x9d, 0x84
993	.BYTE	0x90, 0xd8, 0xab, 0x00, 0x8c, 0xbc, 0xd3, 0x0a
994	.BYTE	0xf7, 0xe4, 0x58, 0x05, 0xb8, 0xb3, 0x45, 0x06
995	.BYTE	0xd0, 0x2c, 0x1e, 0x8f, 0xca, 0x3f, 0x0f, 0x02
996	.BYTE	0xc1, 0xaf, 0xbd, 0x03, 0x01, 0x13, 0x8a, 0x6b
997	.BYTE	0x3a, 0x91, 0x11, 0x41, 0x4f, 0x67, 0xdc, 0xea
998	.BYTE	0x97, 0xf2, 0xcf, 0xce, 0xf0, 0xb4, 0xe6, 0x73
999	.BYTE	0x96, 0xac, 0x74, 0x22, 0xe7, 0xad, 0x35, 0x85
1000	.BYTE	0xe2, 0xf9, 0x37, 0xe8, 0x1c, 0x75, 0xdf, 0x6e
1001	.BYTE	0x47, 0xf1, 0x1a, 0x71, 0x1d, 0x29, 0xc5, 0x89
1002	.BYTE	0x6f, 0xb7, 0x62, 0x0e, 0xaa, 0x18, 0xbe, 0x1b
1003	.BYTE	0xfc, 0x56, 0x3e, 0x4b, 0xc6, 0xd2, 0x79, 0x20
1004	.BYTE	0x9a, 0xdb, 0xc0, 0xfe, 0x78, 0xcd, 0x5a, 0xf4
1005	.BYTE	0x1f, 0xdd, 0xa8, 0x33, 0x88, 0x07, 0xc7, 0x31
1006	.BYTE	0xb1, 0x12, 0x10, 0x59, 0x27, 0x80, 0xec, 0x5f
1007	.BYTE	0x60, 0x51, 0x7f, 0xa9, 0x19, 0xb5, 0x4a, 0x0d
1008	.BYTE	0x2d, 0xe5, 0x7a, 0x9f, 0x93, 0xc9, 0x9c, 0xef
1009	.BYTE	0xa0, 0xe0, 0x3b, 0x4d, 0xae, 0x2a, 0xf5, 0xb0
1010	.BYTE	0xc8, 0xeb, 0xbb, 0x3c, 0x83, 0x53, 0x99, 0x61
1011	.BYTE	0x17, 0x2b, 0x04, 0x7e, 0xba, 0x77, 0xd6, 0x26
1012	.BYTE	0xe1, 0x69, 0x14, 0x63, 0x55, 0x21, 0x0c, 0x7d
1013	.previous
1014___
1015
1016foreach (split("\n",$code)) {
1017	s/\`([^\`]*)\`/eval $1/ge;
1018
1019	# translate made up instructons: _ror, _srm
1020	s/_ror(\s+)(%r[0-9]+),/shd$1$2,$2,/				or
1021
1022	s/_srm(\s+%r[0-9]+),([0-9]+),/
1023		$SIZE_T==4 ? sprintf("extru%s,%d,8,",$1,31-$2)
1024		:            sprintf("extrd,u%s,%d,8,",$1,63-$2)/e;
1025
1026	s/,\*/,/			if ($SIZE_T==4);
1027	s/\bbv\b(.*\(%r2\))/bve$1/	if ($SIZE_T==8);
1028	print $_,"\n";
1029}
1030close STDOUT;
1031