Home | History | Annotate | Download | only in asm
      1 
      2 @ ====================================================================
      3 @ Written by Andy Polyakov <appro@openssl.org> for the OpenSSL
      4 @ project. The module is, however, dual licensed under OpenSSL and
      5 @ CRYPTOGAMS licenses depending on where you obtain it. For further
      6 @ details see http://www.openssl.org/~appro/cryptogams/.
      7 @ ====================================================================
      8 
      9 @ AES for ARMv4
     10 
     11 @ January 2007.
     12 @
     13 @ Code uses single 1K S-box and is >2 times faster than code generated
     14 @ by gcc-3.4.1. This is thanks to unique feature of ARMv4 ISA, which
     15 @ allows to merge logical or arithmetic operation with shift or rotate
     16 @ in one instruction and emit combined result every cycle. The module
     17 @ is endian-neutral. The performance is ~42 cycles/byte for 128-bit
     18 @ key [on single-issue Xscale PXA250 core].
     19 
     20 @ May 2007.
     21 @
     22 @ AES_set_[en|de]crypt_key is added.
     23 
     24 @ July 2010.
     25 @
     26 @ Rescheduling for dual-issue pipeline resulted in 12% improvement on
     27 @ Cortex A8 core and ~25 cycles per byte processed with 128-bit key.
     28 
     29 @ February 2011.
     30 @
     31 @ Profiler-assisted and platform-specific optimization resulted in 16%
     32 @ improvement on Cortex A8 core and ~21.5 cycles per byte.
     33 
     34 #ifndef __KERNEL__
     35 # include "arm_arch.h"
     36 #else
     37 # define __ARM_ARCH__ __LINUX_ARM_ARCH__
     38 #endif
     39 
     40 .text
     41 #if __ARM_ARCH__<7
     42 .code	32
     43 #else
     44 .syntax	unified
     45 # ifdef __thumb2__
     46 .thumb
     47 # else
     48 .code	32
     49 # endif
     50 #endif
     51 
     52 .type	AES_Te,%object
     53 .align	5
     54 AES_Te:
     55 .word	0xc66363a5, 0xf87c7c84, 0xee777799, 0xf67b7b8d
     56 .word	0xfff2f20d, 0xd66b6bbd, 0xde6f6fb1, 0x91c5c554
     57 .word	0x60303050, 0x02010103, 0xce6767a9, 0x562b2b7d
     58 .word	0xe7fefe19, 0xb5d7d762, 0x4dababe6, 0xec76769a
     59 .word	0x8fcaca45, 0x1f82829d, 0x89c9c940, 0xfa7d7d87
     60 .word	0xeffafa15, 0xb25959eb, 0x8e4747c9, 0xfbf0f00b
     61 .word	0x41adadec, 0xb3d4d467, 0x5fa2a2fd, 0x45afafea
     62 .word	0x239c9cbf, 0x53a4a4f7, 0xe4727296, 0x9bc0c05b
     63 .word	0x75b7b7c2, 0xe1fdfd1c, 0x3d9393ae, 0x4c26266a
     64 .word	0x6c36365a, 0x7e3f3f41, 0xf5f7f702, 0x83cccc4f
     65 .word	0x6834345c, 0x51a5a5f4, 0xd1e5e534, 0xf9f1f108
     66 .word	0xe2717193, 0xabd8d873, 0x62313153, 0x2a15153f
     67 .word	0x0804040c, 0x95c7c752, 0x46232365, 0x9dc3c35e
     68 .word	0x30181828, 0x379696a1, 0x0a05050f, 0x2f9a9ab5
     69 .word	0x0e070709, 0x24121236, 0x1b80809b, 0xdfe2e23d
     70 .word	0xcdebeb26, 0x4e272769, 0x7fb2b2cd, 0xea75759f
     71 .word	0x1209091b, 0x1d83839e, 0x582c2c74, 0x341a1a2e
     72 .word	0x361b1b2d, 0xdc6e6eb2, 0xb45a5aee, 0x5ba0a0fb
     73 .word	0xa45252f6, 0x763b3b4d, 0xb7d6d661, 0x7db3b3ce
     74 .word	0x5229297b, 0xdde3e33e, 0x5e2f2f71, 0x13848497
     75 .word	0xa65353f5, 0xb9d1d168, 0x00000000, 0xc1eded2c
     76 .word	0x40202060, 0xe3fcfc1f, 0x79b1b1c8, 0xb65b5bed
     77 .word	0xd46a6abe, 0x8dcbcb46, 0x67bebed9, 0x7239394b
     78 .word	0x944a4ade, 0x984c4cd4, 0xb05858e8, 0x85cfcf4a
     79 .word	0xbbd0d06b, 0xc5efef2a, 0x4faaaae5, 0xedfbfb16
     80 .word	0x864343c5, 0x9a4d4dd7, 0x66333355, 0x11858594
     81 .word	0x8a4545cf, 0xe9f9f910, 0x04020206, 0xfe7f7f81
     82 .word	0xa05050f0, 0x783c3c44, 0x259f9fba, 0x4ba8a8e3
     83 .word	0xa25151f3, 0x5da3a3fe, 0x804040c0, 0x058f8f8a
     84 .word	0x3f9292ad, 0x219d9dbc, 0x70383848, 0xf1f5f504
     85 .word	0x63bcbcdf, 0x77b6b6c1, 0xafdada75, 0x42212163
     86 .word	0x20101030, 0xe5ffff1a, 0xfdf3f30e, 0xbfd2d26d
     87 .word	0x81cdcd4c, 0x180c0c14, 0x26131335, 0xc3ecec2f
     88 .word	0xbe5f5fe1, 0x359797a2, 0x884444cc, 0x2e171739
     89 .word	0x93c4c457, 0x55a7a7f2, 0xfc7e7e82, 0x7a3d3d47
     90 .word	0xc86464ac, 0xba5d5de7, 0x3219192b, 0xe6737395
     91 .word	0xc06060a0, 0x19818198, 0x9e4f4fd1, 0xa3dcdc7f
     92 .word	0x44222266, 0x542a2a7e, 0x3b9090ab, 0x0b888883
     93 .word	0x8c4646ca, 0xc7eeee29, 0x6bb8b8d3, 0x2814143c
     94 .word	0xa7dede79, 0xbc5e5ee2, 0x160b0b1d, 0xaddbdb76
     95 .word	0xdbe0e03b, 0x64323256, 0x743a3a4e, 0x140a0a1e
     96 .word	0x924949db, 0x0c06060a, 0x4824246c, 0xb85c5ce4
     97 .word	0x9fc2c25d, 0xbdd3d36e, 0x43acacef, 0xc46262a6
     98 .word	0x399191a8, 0x319595a4, 0xd3e4e437, 0xf279798b
     99 .word	0xd5e7e732, 0x8bc8c843, 0x6e373759, 0xda6d6db7
    100 .word	0x018d8d8c, 0xb1d5d564, 0x9c4e4ed2, 0x49a9a9e0
    101 .word	0xd86c6cb4, 0xac5656fa, 0xf3f4f407, 0xcfeaea25
    102 .word	0xca6565af, 0xf47a7a8e, 0x47aeaee9, 0x10080818
    103 .word	0x6fbabad5, 0xf0787888, 0x4a25256f, 0x5c2e2e72
    104 .word	0x381c1c24, 0x57a6a6f1, 0x73b4b4c7, 0x97c6c651
    105 .word	0xcbe8e823, 0xa1dddd7c, 0xe874749c, 0x3e1f1f21
    106 .word	0x964b4bdd, 0x61bdbddc, 0x0d8b8b86, 0x0f8a8a85
    107 .word	0xe0707090, 0x7c3e3e42, 0x71b5b5c4, 0xcc6666aa
    108 .word	0x904848d8, 0x06030305, 0xf7f6f601, 0x1c0e0e12
    109 .word	0xc26161a3, 0x6a35355f, 0xae5757f9, 0x69b9b9d0
    110 .word	0x17868691, 0x99c1c158, 0x3a1d1d27, 0x279e9eb9
    111 .word	0xd9e1e138, 0xebf8f813, 0x2b9898b3, 0x22111133
    112 .word	0xd26969bb, 0xa9d9d970, 0x078e8e89, 0x339494a7
    113 .word	0x2d9b9bb6, 0x3c1e1e22, 0x15878792, 0xc9e9e920
    114 .word	0x87cece49, 0xaa5555ff, 0x50282878, 0xa5dfdf7a
    115 .word	0x038c8c8f, 0x59a1a1f8, 0x09898980, 0x1a0d0d17
    116 .word	0x65bfbfda, 0xd7e6e631, 0x844242c6, 0xd06868b8
    117 .word	0x824141c3, 0x299999b0, 0x5a2d2d77, 0x1e0f0f11
    118 .word	0x7bb0b0cb, 0xa85454fc, 0x6dbbbbd6, 0x2c16163a
    119 @ Te4[256]
    120 .byte	0x63, 0x7c, 0x77, 0x7b, 0xf2, 0x6b, 0x6f, 0xc5
    121 .byte	0x30, 0x01, 0x67, 0x2b, 0xfe, 0xd7, 0xab, 0x76
    122 .byte	0xca, 0x82, 0xc9, 0x7d, 0xfa, 0x59, 0x47, 0xf0
    123 .byte	0xad, 0xd4, 0xa2, 0xaf, 0x9c, 0xa4, 0x72, 0xc0
    124 .byte	0xb7, 0xfd, 0x93, 0x26, 0x36, 0x3f, 0xf7, 0xcc
    125 .byte	0x34, 0xa5, 0xe5, 0xf1, 0x71, 0xd8, 0x31, 0x15
    126 .byte	0x04, 0xc7, 0x23, 0xc3, 0x18, 0x96, 0x05, 0x9a
    127 .byte	0x07, 0x12, 0x80, 0xe2, 0xeb, 0x27, 0xb2, 0x75
    128 .byte	0x09, 0x83, 0x2c, 0x1a, 0x1b, 0x6e, 0x5a, 0xa0
    129 .byte	0x52, 0x3b, 0xd6, 0xb3, 0x29, 0xe3, 0x2f, 0x84
    130 .byte	0x53, 0xd1, 0x00, 0xed, 0x20, 0xfc, 0xb1, 0x5b
    131 .byte	0x6a, 0xcb, 0xbe, 0x39, 0x4a, 0x4c, 0x58, 0xcf
    132 .byte	0xd0, 0xef, 0xaa, 0xfb, 0x43, 0x4d, 0x33, 0x85
    133 .byte	0x45, 0xf9, 0x02, 0x7f, 0x50, 0x3c, 0x9f, 0xa8
    134 .byte	0x51, 0xa3, 0x40, 0x8f, 0x92, 0x9d, 0x38, 0xf5
    135 .byte	0xbc, 0xb6, 0xda, 0x21, 0x10, 0xff, 0xf3, 0xd2
    136 .byte	0xcd, 0x0c, 0x13, 0xec, 0x5f, 0x97, 0x44, 0x17
    137 .byte	0xc4, 0xa7, 0x7e, 0x3d, 0x64, 0x5d, 0x19, 0x73
    138 .byte	0x60, 0x81, 0x4f, 0xdc, 0x22, 0x2a, 0x90, 0x88
    139 .byte	0x46, 0xee, 0xb8, 0x14, 0xde, 0x5e, 0x0b, 0xdb
    140 .byte	0xe0, 0x32, 0x3a, 0x0a, 0x49, 0x06, 0x24, 0x5c
    141 .byte	0xc2, 0xd3, 0xac, 0x62, 0x91, 0x95, 0xe4, 0x79
    142 .byte	0xe7, 0xc8, 0x37, 0x6d, 0x8d, 0xd5, 0x4e, 0xa9
    143 .byte	0x6c, 0x56, 0xf4, 0xea, 0x65, 0x7a, 0xae, 0x08
    144 .byte	0xba, 0x78, 0x25, 0x2e, 0x1c, 0xa6, 0xb4, 0xc6
    145 .byte	0xe8, 0xdd, 0x74, 0x1f, 0x4b, 0xbd, 0x8b, 0x8a
    146 .byte	0x70, 0x3e, 0xb5, 0x66, 0x48, 0x03, 0xf6, 0x0e
    147 .byte	0x61, 0x35, 0x57, 0xb9, 0x86, 0xc1, 0x1d, 0x9e
    148 .byte	0xe1, 0xf8, 0x98, 0x11, 0x69, 0xd9, 0x8e, 0x94
    149 .byte	0x9b, 0x1e, 0x87, 0xe9, 0xce, 0x55, 0x28, 0xdf
    150 .byte	0x8c, 0xa1, 0x89, 0x0d, 0xbf, 0xe6, 0x42, 0x68
    151 .byte	0x41, 0x99, 0x2d, 0x0f, 0xb0, 0x54, 0xbb, 0x16
    152 @ rcon[]
    153 .word	0x01000000, 0x02000000, 0x04000000, 0x08000000
    154 .word	0x10000000, 0x20000000, 0x40000000, 0x80000000
    155 .word	0x1B000000, 0x36000000, 0, 0, 0, 0, 0, 0
    156 .size	AES_Te,.-AES_Te
    157 
    158 @ void AES_encrypt(const unsigned char *in, unsigned char *out,
    159 @ 		 const AES_KEY *key) {
    160 .global AES_encrypt
    161 .type   AES_encrypt,%function
    162 .align	5
    163 AES_encrypt:
    164 #if __ARM_ARCH__<7
    165 	sub	r3,pc,#8		@ AES_encrypt
    166 #else
    167 	adr	r3,AES_encrypt
    168 #endif
    169 	stmdb   sp!,{r1,r4-r12,lr}
    170 	mov	r12,r0		@ inp
    171 	mov	r11,r2
    172 	sub	r10,r3,#AES_encrypt-AES_Te	@ Te
    173 #if __ARM_ARCH__<7
    174 	ldrb	r0,[r12,#3]	@ load input data in endian-neutral
    175 	ldrb	r4,[r12,#2]	@ manner...
    176 	ldrb	r5,[r12,#1]
    177 	ldrb	r6,[r12,#0]
    178 	orr	r0,r0,r4,lsl#8
    179 	ldrb	r1,[r12,#7]
    180 	orr	r0,r0,r5,lsl#16
    181 	ldrb	r4,[r12,#6]
    182 	orr	r0,r0,r6,lsl#24
    183 	ldrb	r5,[r12,#5]
    184 	ldrb	r6,[r12,#4]
    185 	orr	r1,r1,r4,lsl#8
    186 	ldrb	r2,[r12,#11]
    187 	orr	r1,r1,r5,lsl#16
    188 	ldrb	r4,[r12,#10]
    189 	orr	r1,r1,r6,lsl#24
    190 	ldrb	r5,[r12,#9]
    191 	ldrb	r6,[r12,#8]
    192 	orr	r2,r2,r4,lsl#8
    193 	ldrb	r3,[r12,#15]
    194 	orr	r2,r2,r5,lsl#16
    195 	ldrb	r4,[r12,#14]
    196 	orr	r2,r2,r6,lsl#24
    197 	ldrb	r5,[r12,#13]
    198 	ldrb	r6,[r12,#12]
    199 	orr	r3,r3,r4,lsl#8
    200 	orr	r3,r3,r5,lsl#16
    201 	orr	r3,r3,r6,lsl#24
    202 #else
    203 	ldr	r0,[r12,#0]
    204 	ldr	r1,[r12,#4]
    205 	ldr	r2,[r12,#8]
    206 	ldr	r3,[r12,#12]
    207 #ifdef __ARMEL__
    208 	rev	r0,r0
    209 	rev	r1,r1
    210 	rev	r2,r2
    211 	rev	r3,r3
    212 #endif
    213 #endif
    214 	bl	_armv4_AES_encrypt
    215 
    216 	ldr	r12,[sp],#4		@ pop out
    217 #if __ARM_ARCH__>=7
    218 #ifdef __ARMEL__
    219 	rev	r0,r0
    220 	rev	r1,r1
    221 	rev	r2,r2
    222 	rev	r3,r3
    223 #endif
    224 	str	r0,[r12,#0]
    225 	str	r1,[r12,#4]
    226 	str	r2,[r12,#8]
    227 	str	r3,[r12,#12]
    228 #else
    229 	mov	r4,r0,lsr#24		@ write output in endian-neutral
    230 	mov	r5,r0,lsr#16		@ manner...
    231 	mov	r6,r0,lsr#8
    232 	strb	r4,[r12,#0]
    233 	strb	r5,[r12,#1]
    234 	mov	r4,r1,lsr#24
    235 	strb	r6,[r12,#2]
    236 	mov	r5,r1,lsr#16
    237 	strb	r0,[r12,#3]
    238 	mov	r6,r1,lsr#8
    239 	strb	r4,[r12,#4]
    240 	strb	r5,[r12,#5]
    241 	mov	r4,r2,lsr#24
    242 	strb	r6,[r12,#6]
    243 	mov	r5,r2,lsr#16
    244 	strb	r1,[r12,#7]
    245 	mov	r6,r2,lsr#8
    246 	strb	r4,[r12,#8]
    247 	strb	r5,[r12,#9]
    248 	mov	r4,r3,lsr#24
    249 	strb	r6,[r12,#10]
    250 	mov	r5,r3,lsr#16
    251 	strb	r2,[r12,#11]
    252 	mov	r6,r3,lsr#8
    253 	strb	r4,[r12,#12]
    254 	strb	r5,[r12,#13]
    255 	strb	r6,[r12,#14]
    256 	strb	r3,[r12,#15]
    257 #endif
    258 #if __ARM_ARCH__>=5
    259 	ldmia	sp!,{r4-r12,pc}
    260 #else
    261 	ldmia   sp!,{r4-r12,lr}
    262 	tst	lr,#1
    263 	moveq	pc,lr			@ be binary compatible with V4, yet
    264 	.word	0xe12fff1e			@ interoperable with Thumb ISA:-)
    265 #endif
    266 .size	AES_encrypt,.-AES_encrypt
    267 
    268 .type   _armv4_AES_encrypt,%function
    269 .align	2
    270 _armv4_AES_encrypt:
    271 	str	lr,[sp,#-4]!		@ push lr
    272 	ldmia	r11!,{r4-r7}
    273 	eor	r0,r0,r4
    274 	ldr	r12,[r11,#240-16]
    275 	eor	r1,r1,r5
    276 	eor	r2,r2,r6
    277 	eor	r3,r3,r7
    278 	sub	r12,r12,#1
    279 	mov	lr,#255
    280 
    281 	and	r7,lr,r0
    282 	and	r8,lr,r0,lsr#8
    283 	and	r9,lr,r0,lsr#16
    284 	mov	r0,r0,lsr#24
    285 .Lenc_loop:
    286 	ldr	r4,[r10,r7,lsl#2]	@ Te3[s0>>0]
    287 	and	r7,lr,r1,lsr#16	@ i0
    288 	ldr	r5,[r10,r8,lsl#2]	@ Te2[s0>>8]
    289 	and	r8,lr,r1
    290 	ldr	r6,[r10,r9,lsl#2]	@ Te1[s0>>16]
    291 	and	r9,lr,r1,lsr#8
    292 	ldr	r0,[r10,r0,lsl#2]	@ Te0[s0>>24]
    293 	mov	r1,r1,lsr#24
    294 
    295 	ldr	r7,[r10,r7,lsl#2]	@ Te1[s1>>16]
    296 	ldr	r8,[r10,r8,lsl#2]	@ Te3[s1>>0]
    297 	ldr	r9,[r10,r9,lsl#2]	@ Te2[s1>>8]
    298 	eor	r0,r0,r7,ror#8
    299 	ldr	r1,[r10,r1,lsl#2]	@ Te0[s1>>24]
    300 	and	r7,lr,r2,lsr#8	@ i0
    301 	eor	r5,r5,r8,ror#8
    302 	and	r8,lr,r2,lsr#16	@ i1
    303 	eor	r6,r6,r9,ror#8
    304 	and	r9,lr,r2
    305 	ldr	r7,[r10,r7,lsl#2]	@ Te2[s2>>8]
    306 	eor	r1,r1,r4,ror#24
    307 	ldr	r8,[r10,r8,lsl#2]	@ Te1[s2>>16]
    308 	mov	r2,r2,lsr#24
    309 
    310 	ldr	r9,[r10,r9,lsl#2]	@ Te3[s2>>0]
    311 	eor	r0,r0,r7,ror#16
    312 	ldr	r2,[r10,r2,lsl#2]	@ Te0[s2>>24]
    313 	and	r7,lr,r3		@ i0
    314 	eor	r1,r1,r8,ror#8
    315 	and	r8,lr,r3,lsr#8	@ i1
    316 	eor	r6,r6,r9,ror#16
    317 	and	r9,lr,r3,lsr#16	@ i2
    318 	ldr	r7,[r10,r7,lsl#2]	@ Te3[s3>>0]
    319 	eor	r2,r2,r5,ror#16
    320 	ldr	r8,[r10,r8,lsl#2]	@ Te2[s3>>8]
    321 	mov	r3,r3,lsr#24
    322 
    323 	ldr	r9,[r10,r9,lsl#2]	@ Te1[s3>>16]
    324 	eor	r0,r0,r7,ror#24
    325 	ldr	r7,[r11],#16
    326 	eor	r1,r1,r8,ror#16
    327 	ldr	r3,[r10,r3,lsl#2]	@ Te0[s3>>24]
    328 	eor	r2,r2,r9,ror#8
    329 	ldr	r4,[r11,#-12]
    330 	eor	r3,r3,r6,ror#8
    331 
    332 	ldr	r5,[r11,#-8]
    333 	eor	r0,r0,r7
    334 	ldr	r6,[r11,#-4]
    335 	and	r7,lr,r0
    336 	eor	r1,r1,r4
    337 	and	r8,lr,r0,lsr#8
    338 	eor	r2,r2,r5
    339 	and	r9,lr,r0,lsr#16
    340 	eor	r3,r3,r6
    341 	mov	r0,r0,lsr#24
    342 
    343 	subs	r12,r12,#1
    344 	bne	.Lenc_loop
    345 
    346 	add	r10,r10,#2
    347 
    348 	ldrb	r4,[r10,r7,lsl#2]	@ Te4[s0>>0]
    349 	and	r7,lr,r1,lsr#16	@ i0
    350 	ldrb	r5,[r10,r8,lsl#2]	@ Te4[s0>>8]
    351 	and	r8,lr,r1
    352 	ldrb	r6,[r10,r9,lsl#2]	@ Te4[s0>>16]
    353 	and	r9,lr,r1,lsr#8
    354 	ldrb	r0,[r10,r0,lsl#2]	@ Te4[s0>>24]
    355 	mov	r1,r1,lsr#24
    356 
    357 	ldrb	r7,[r10,r7,lsl#2]	@ Te4[s1>>16]
    358 	ldrb	r8,[r10,r8,lsl#2]	@ Te4[s1>>0]
    359 	ldrb	r9,[r10,r9,lsl#2]	@ Te4[s1>>8]
    360 	eor	r0,r7,r0,lsl#8
    361 	ldrb	r1,[r10,r1,lsl#2]	@ Te4[s1>>24]
    362 	and	r7,lr,r2,lsr#8	@ i0
    363 	eor	r5,r8,r5,lsl#8
    364 	and	r8,lr,r2,lsr#16	@ i1
    365 	eor	r6,r9,r6,lsl#8
    366 	and	r9,lr,r2
    367 	ldrb	r7,[r10,r7,lsl#2]	@ Te4[s2>>8]
    368 	eor	r1,r4,r1,lsl#24
    369 	ldrb	r8,[r10,r8,lsl#2]	@ Te4[s2>>16]
    370 	mov	r2,r2,lsr#24
    371 
    372 	ldrb	r9,[r10,r9,lsl#2]	@ Te4[s2>>0]
    373 	eor	r0,r7,r0,lsl#8
    374 	ldrb	r2,[r10,r2,lsl#2]	@ Te4[s2>>24]
    375 	and	r7,lr,r3		@ i0
    376 	eor	r1,r1,r8,lsl#16
    377 	and	r8,lr,r3,lsr#8	@ i1
    378 	eor	r6,r9,r6,lsl#8
    379 	and	r9,lr,r3,lsr#16	@ i2
    380 	ldrb	r7,[r10,r7,lsl#2]	@ Te4[s3>>0]
    381 	eor	r2,r5,r2,lsl#24
    382 	ldrb	r8,[r10,r8,lsl#2]	@ Te4[s3>>8]
    383 	mov	r3,r3,lsr#24
    384 
    385 	ldrb	r9,[r10,r9,lsl#2]	@ Te4[s3>>16]
    386 	eor	r0,r7,r0,lsl#8
    387 	ldr	r7,[r11,#0]
    388 	ldrb	r3,[r10,r3,lsl#2]	@ Te4[s3>>24]
    389 	eor	r1,r1,r8,lsl#8
    390 	ldr	r4,[r11,#4]
    391 	eor	r2,r2,r9,lsl#16
    392 	ldr	r5,[r11,#8]
    393 	eor	r3,r6,r3,lsl#24
    394 	ldr	r6,[r11,#12]
    395 
    396 	eor	r0,r0,r7
    397 	eor	r1,r1,r4
    398 	eor	r2,r2,r5
    399 	eor	r3,r3,r6
    400 
    401 	sub	r10,r10,#2
    402 	ldr	pc,[sp],#4		@ pop and return
    403 .size	_armv4_AES_encrypt,.-_armv4_AES_encrypt
    404 
    405 .global private_AES_set_encrypt_key
    406 .type   private_AES_set_encrypt_key,%function
    407 .align	5
    408 private_AES_set_encrypt_key:
    409 _armv4_AES_set_encrypt_key:
    410 #if __ARM_ARCH__<7
    411 	sub	r3,pc,#8		@ AES_set_encrypt_key
    412 #else
    413 	adr	r3,private_AES_set_encrypt_key
    414 #endif
    415 	teq	r0,#0
    416 #if __ARM_ARCH__>=7
    417 	itt	eq			@ Thumb2 thing, sanity check in ARM
    418 #endif
    419 	moveq	r0,#-1
    420 	beq	.Labrt
    421 	teq	r2,#0
    422 #if __ARM_ARCH__>=7
    423 	itt	eq			@ Thumb2 thing, sanity check in ARM
    424 #endif
    425 	moveq	r0,#-1
    426 	beq	.Labrt
    427 
    428 	teq	r1,#128
    429 	beq	.Lok
    430 	teq	r1,#192
    431 	beq	.Lok
    432 	teq	r1,#256
    433 #if __ARM_ARCH__>=7
    434 	itt	ne			@ Thumb2 thing, sanity check in ARM
    435 #endif
    436 	movne	r0,#-1
    437 	bne	.Labrt
    438 
    439 .Lok:	stmdb   sp!,{r4-r12,lr}
    440 	sub	r10,r3,#_armv4_AES_set_encrypt_key-AES_Te-1024	@ Te4
    441 
    442 	mov	r12,r0		@ inp
    443 	mov	lr,r1			@ bits
    444 	mov	r11,r2			@ key
    445 
    446 #if __ARM_ARCH__<7
    447 	ldrb	r0,[r12,#3]	@ load input data in endian-neutral
    448 	ldrb	r4,[r12,#2]	@ manner...
    449 	ldrb	r5,[r12,#1]
    450 	ldrb	r6,[r12,#0]
    451 	orr	r0,r0,r4,lsl#8
    452 	ldrb	r1,[r12,#7]
    453 	orr	r0,r0,r5,lsl#16
    454 	ldrb	r4,[r12,#6]
    455 	orr	r0,r0,r6,lsl#24
    456 	ldrb	r5,[r12,#5]
    457 	ldrb	r6,[r12,#4]
    458 	orr	r1,r1,r4,lsl#8
    459 	ldrb	r2,[r12,#11]
    460 	orr	r1,r1,r5,lsl#16
    461 	ldrb	r4,[r12,#10]
    462 	orr	r1,r1,r6,lsl#24
    463 	ldrb	r5,[r12,#9]
    464 	ldrb	r6,[r12,#8]
    465 	orr	r2,r2,r4,lsl#8
    466 	ldrb	r3,[r12,#15]
    467 	orr	r2,r2,r5,lsl#16
    468 	ldrb	r4,[r12,#14]
    469 	orr	r2,r2,r6,lsl#24
    470 	ldrb	r5,[r12,#13]
    471 	ldrb	r6,[r12,#12]
    472 	orr	r3,r3,r4,lsl#8
    473 	str	r0,[r11],#16
    474 	orr	r3,r3,r5,lsl#16
    475 	str	r1,[r11,#-12]
    476 	orr	r3,r3,r6,lsl#24
    477 	str	r2,[r11,#-8]
    478 	str	r3,[r11,#-4]
    479 #else
    480 	ldr	r0,[r12,#0]
    481 	ldr	r1,[r12,#4]
    482 	ldr	r2,[r12,#8]
    483 	ldr	r3,[r12,#12]
    484 #ifdef __ARMEL__
    485 	rev	r0,r0
    486 	rev	r1,r1
    487 	rev	r2,r2
    488 	rev	r3,r3
    489 #endif
    490 	str	r0,[r11],#16
    491 	str	r1,[r11,#-12]
    492 	str	r2,[r11,#-8]
    493 	str	r3,[r11,#-4]
    494 #endif
    495 
    496 	teq	lr,#128
    497 	bne	.Lnot128
    498 	mov	r12,#10
    499 	str	r12,[r11,#240-16]
    500 	add	r6,r10,#256			@ rcon
    501 	mov	lr,#255
    502 
    503 .L128_loop:
    504 	and	r5,lr,r3,lsr#24
    505 	and	r7,lr,r3,lsr#16
    506 	ldrb	r5,[r10,r5]
    507 	and	r8,lr,r3,lsr#8
    508 	ldrb	r7,[r10,r7]
    509 	and	r9,lr,r3
    510 	ldrb	r8,[r10,r8]
    511 	orr	r5,r5,r7,lsl#24
    512 	ldrb	r9,[r10,r9]
    513 	orr	r5,r5,r8,lsl#16
    514 	ldr	r4,[r6],#4			@ rcon[i++]
    515 	orr	r5,r5,r9,lsl#8
    516 	eor	r5,r5,r4
    517 	eor	r0,r0,r5			@ rk[4]=rk[0]^...
    518 	eor	r1,r1,r0			@ rk[5]=rk[1]^rk[4]
    519 	str	r0,[r11],#16
    520 	eor	r2,r2,r1			@ rk[6]=rk[2]^rk[5]
    521 	str	r1,[r11,#-12]
    522 	eor	r3,r3,r2			@ rk[7]=rk[3]^rk[6]
    523 	str	r2,[r11,#-8]
    524 	subs	r12,r12,#1
    525 	str	r3,[r11,#-4]
    526 	bne	.L128_loop
    527 	sub	r2,r11,#176
    528 	b	.Ldone
    529 
    530 .Lnot128:
    531 #if __ARM_ARCH__<7
    532 	ldrb	r8,[r12,#19]
    533 	ldrb	r4,[r12,#18]
    534 	ldrb	r5,[r12,#17]
    535 	ldrb	r6,[r12,#16]
    536 	orr	r8,r8,r4,lsl#8
    537 	ldrb	r9,[r12,#23]
    538 	orr	r8,r8,r5,lsl#16
    539 	ldrb	r4,[r12,#22]
    540 	orr	r8,r8,r6,lsl#24
    541 	ldrb	r5,[r12,#21]
    542 	ldrb	r6,[r12,#20]
    543 	orr	r9,r9,r4,lsl#8
    544 	orr	r9,r9,r5,lsl#16
    545 	str	r8,[r11],#8
    546 	orr	r9,r9,r6,lsl#24
    547 	str	r9,[r11,#-4]
    548 #else
    549 	ldr	r8,[r12,#16]
    550 	ldr	r9,[r12,#20]
    551 #ifdef __ARMEL__
    552 	rev	r8,r8
    553 	rev	r9,r9
    554 #endif
    555 	str	r8,[r11],#8
    556 	str	r9,[r11,#-4]
    557 #endif
    558 
    559 	teq	lr,#192
    560 	bne	.Lnot192
    561 	mov	r12,#12
    562 	str	r12,[r11,#240-24]
    563 	add	r6,r10,#256			@ rcon
    564 	mov	lr,#255
    565 	mov	r12,#8
    566 
    567 .L192_loop:
    568 	and	r5,lr,r9,lsr#24
    569 	and	r7,lr,r9,lsr#16
    570 	ldrb	r5,[r10,r5]
    571 	and	r8,lr,r9,lsr#8
    572 	ldrb	r7,[r10,r7]
    573 	and	r9,lr,r9
    574 	ldrb	r8,[r10,r8]
    575 	orr	r5,r5,r7,lsl#24
    576 	ldrb	r9,[r10,r9]
    577 	orr	r5,r5,r8,lsl#16
    578 	ldr	r4,[r6],#4			@ rcon[i++]
    579 	orr	r5,r5,r9,lsl#8
    580 	eor	r9,r5,r4
    581 	eor	r0,r0,r9			@ rk[6]=rk[0]^...
    582 	eor	r1,r1,r0			@ rk[7]=rk[1]^rk[6]
    583 	str	r0,[r11],#24
    584 	eor	r2,r2,r1			@ rk[8]=rk[2]^rk[7]
    585 	str	r1,[r11,#-20]
    586 	eor	r3,r3,r2			@ rk[9]=rk[3]^rk[8]
    587 	str	r2,[r11,#-16]
    588 	subs	r12,r12,#1
    589 	str	r3,[r11,#-12]
    590 #if __ARM_ARCH__>=7
    591 	itt	eq				@ Thumb2 thing, sanity check in ARM
    592 #endif
    593 	subeq	r2,r11,#216
    594 	beq	.Ldone
    595 
    596 	ldr	r7,[r11,#-32]
    597 	ldr	r8,[r11,#-28]
    598 	eor	r7,r7,r3			@ rk[10]=rk[4]^rk[9]
    599 	eor	r9,r8,r7			@ rk[11]=rk[5]^rk[10]
    600 	str	r7,[r11,#-8]
    601 	str	r9,[r11,#-4]
    602 	b	.L192_loop
    603 
    604 .Lnot192:
    605 #if __ARM_ARCH__<7
    606 	ldrb	r8,[r12,#27]
    607 	ldrb	r4,[r12,#26]
    608 	ldrb	r5,[r12,#25]
    609 	ldrb	r6,[r12,#24]
    610 	orr	r8,r8,r4,lsl#8
    611 	ldrb	r9,[r12,#31]
    612 	orr	r8,r8,r5,lsl#16
    613 	ldrb	r4,[r12,#30]
    614 	orr	r8,r8,r6,lsl#24
    615 	ldrb	r5,[r12,#29]
    616 	ldrb	r6,[r12,#28]
    617 	orr	r9,r9,r4,lsl#8
    618 	orr	r9,r9,r5,lsl#16
    619 	str	r8,[r11],#8
    620 	orr	r9,r9,r6,lsl#24
    621 	str	r9,[r11,#-4]
    622 #else
    623 	ldr	r8,[r12,#24]
    624 	ldr	r9,[r12,#28]
    625 #ifdef __ARMEL__
    626 	rev	r8,r8
    627 	rev	r9,r9
    628 #endif
    629 	str	r8,[r11],#8
    630 	str	r9,[r11,#-4]
    631 #endif
    632 
    633 	mov	r12,#14
    634 	str	r12,[r11,#240-32]
    635 	add	r6,r10,#256			@ rcon
    636 	mov	lr,#255
    637 	mov	r12,#7
    638 
    639 .L256_loop:
    640 	and	r5,lr,r9,lsr#24
    641 	and	r7,lr,r9,lsr#16
    642 	ldrb	r5,[r10,r5]
    643 	and	r8,lr,r9,lsr#8
    644 	ldrb	r7,[r10,r7]
    645 	and	r9,lr,r9
    646 	ldrb	r8,[r10,r8]
    647 	orr	r5,r5,r7,lsl#24
    648 	ldrb	r9,[r10,r9]
    649 	orr	r5,r5,r8,lsl#16
    650 	ldr	r4,[r6],#4			@ rcon[i++]
    651 	orr	r5,r5,r9,lsl#8
    652 	eor	r9,r5,r4
    653 	eor	r0,r0,r9			@ rk[8]=rk[0]^...
    654 	eor	r1,r1,r0			@ rk[9]=rk[1]^rk[8]
    655 	str	r0,[r11],#32
    656 	eor	r2,r2,r1			@ rk[10]=rk[2]^rk[9]
    657 	str	r1,[r11,#-28]
    658 	eor	r3,r3,r2			@ rk[11]=rk[3]^rk[10]
    659 	str	r2,[r11,#-24]
    660 	subs	r12,r12,#1
    661 	str	r3,[r11,#-20]
    662 #if __ARM_ARCH__>=7
    663 	itt	eq				@ Thumb2 thing, sanity check in ARM
    664 #endif
    665 	subeq	r2,r11,#256
    666 	beq	.Ldone
    667 
    668 	and	r5,lr,r3
    669 	and	r7,lr,r3,lsr#8
    670 	ldrb	r5,[r10,r5]
    671 	and	r8,lr,r3,lsr#16
    672 	ldrb	r7,[r10,r7]
    673 	and	r9,lr,r3,lsr#24
    674 	ldrb	r8,[r10,r8]
    675 	orr	r5,r5,r7,lsl#8
    676 	ldrb	r9,[r10,r9]
    677 	orr	r5,r5,r8,lsl#16
    678 	ldr	r4,[r11,#-48]
    679 	orr	r5,r5,r9,lsl#24
    680 
    681 	ldr	r7,[r11,#-44]
    682 	ldr	r8,[r11,#-40]
    683 	eor	r4,r4,r5			@ rk[12]=rk[4]^...
    684 	ldr	r9,[r11,#-36]
    685 	eor	r7,r7,r4			@ rk[13]=rk[5]^rk[12]
    686 	str	r4,[r11,#-16]
    687 	eor	r8,r8,r7			@ rk[14]=rk[6]^rk[13]
    688 	str	r7,[r11,#-12]
    689 	eor	r9,r9,r8			@ rk[15]=rk[7]^rk[14]
    690 	str	r8,[r11,#-8]
    691 	str	r9,[r11,#-4]
    692 	b	.L256_loop
    693 
    694 .align	2
    695 .Ldone:	mov	r0,#0
    696 	ldmia   sp!,{r4-r12,lr}
    697 .Labrt:
    698 #if __ARM_ARCH__>=5
    699 	bx	lr				@ .word	0xe12fff1e
    700 #else
    701 	tst	lr,#1
    702 	moveq	pc,lr			@ be binary compatible with V4, yet
    703 	.word	0xe12fff1e			@ interoperable with Thumb ISA:-)
    704 #endif
    705 .size	private_AES_set_encrypt_key,.-private_AES_set_encrypt_key
    706 
    707 .global private_AES_set_decrypt_key
    708 .type   private_AES_set_decrypt_key,%function
    709 .align	5
    710 private_AES_set_decrypt_key:
    711 	str	lr,[sp,#-4]!            @ push lr
    712 	bl	_armv4_AES_set_encrypt_key
    713 	teq	r0,#0
    714 	ldr	lr,[sp],#4              @ pop lr
    715 	bne	.Labrt
    716 
    717 	mov	r0,r2			@ AES_set_encrypt_key preserves r2,
    718 	mov	r1,r2			@ which is AES_KEY *key
    719 	b	_armv4_AES_set_enc2dec_key
    720 .size	private_AES_set_decrypt_key,.-private_AES_set_decrypt_key
    721 
    722 @ void AES_set_enc2dec_key(const AES_KEY *inp,AES_KEY *out)
    723 .global	AES_set_enc2dec_key
    724 .type	AES_set_enc2dec_key,%function
    725 .align	5
    726 AES_set_enc2dec_key:
    727 _armv4_AES_set_enc2dec_key:
    728 	stmdb   sp!,{r4-r12,lr}
    729 
    730 	ldr	r12,[r0,#240]
    731 	mov	r7,r0			@ input
    732 	add	r8,r0,r12,lsl#4
    733 	mov	r11,r1			@ ouput
    734 	add	r10,r1,r12,lsl#4
    735 	str	r12,[r1,#240]
    736 
    737 .Linv:	ldr	r0,[r7],#16
    738 	ldr	r1,[r7,#-12]
    739 	ldr	r2,[r7,#-8]
    740 	ldr	r3,[r7,#-4]
    741 	ldr	r4,[r8],#-16
    742 	ldr	r5,[r8,#16+4]
    743 	ldr	r6,[r8,#16+8]
    744 	ldr	r9,[r8,#16+12]
    745 	str	r0,[r10],#-16
    746 	str	r1,[r10,#16+4]
    747 	str	r2,[r10,#16+8]
    748 	str	r3,[r10,#16+12]
    749 	str	r4,[r11],#16
    750 	str	r5,[r11,#-12]
    751 	str	r6,[r11,#-8]
    752 	str	r9,[r11,#-4]
    753 	teq	r7,r8
    754 	bne	.Linv
    755 
    756 	ldr	r0,[r7]
    757 	ldr	r1,[r7,#4]
    758 	ldr	r2,[r7,#8]
    759 	ldr	r3,[r7,#12]
    760 	str	r0,[r11]
    761 	str	r1,[r11,#4]
    762 	str	r2,[r11,#8]
    763 	str	r3,[r11,#12]
    764 	sub	r11,r11,r12,lsl#3
    765 	ldr	r0,[r11,#16]!		@ prefetch tp1
    766 	mov	r7,#0x80
    767 	mov	r8,#0x1b
    768 	orr	r7,r7,#0x8000
    769 	orr	r8,r8,#0x1b00
    770 	orr	r7,r7,r7,lsl#16
    771 	orr	r8,r8,r8,lsl#16
    772 	sub	r12,r12,#1
    773 	mvn	r9,r7
    774 	mov	r12,r12,lsl#2	@ (rounds-1)*4
    775 
    776 .Lmix:	and	r4,r0,r7
    777 	and	r1,r0,r9
    778 	sub	r4,r4,r4,lsr#7
    779 	and	r4,r4,r8
    780 	eor	r1,r4,r1,lsl#1	@ tp2
    781 
    782 	and	r4,r1,r7
    783 	and	r2,r1,r9
    784 	sub	r4,r4,r4,lsr#7
    785 	and	r4,r4,r8
    786 	eor	r2,r4,r2,lsl#1	@ tp4
    787 
    788 	and	r4,r2,r7
    789 	and	r3,r2,r9
    790 	sub	r4,r4,r4,lsr#7
    791 	and	r4,r4,r8
    792 	eor	r3,r4,r3,lsl#1	@ tp8
    793 
    794 	eor	r4,r1,r2
    795 	eor	r5,r0,r3		@ tp9
    796 	eor	r4,r4,r3		@ tpe
    797 	eor	r4,r4,r1,ror#24
    798 	eor	r4,r4,r5,ror#24	@ ^= ROTATE(tpb=tp9^tp2,8)
    799 	eor	r4,r4,r2,ror#16
    800 	eor	r4,r4,r5,ror#16	@ ^= ROTATE(tpd=tp9^tp4,16)
    801 	eor	r4,r4,r5,ror#8	@ ^= ROTATE(tp9,24)
    802 
    803 	ldr	r0,[r11,#4]		@ prefetch tp1
    804 	str	r4,[r11],#4
    805 	subs	r12,r12,#1
    806 	bne	.Lmix
    807 
    808 	mov	r0,#0
    809 #if __ARM_ARCH__>=5
    810 	ldmia	sp!,{r4-r12,pc}
    811 #else
    812 	ldmia   sp!,{r4-r12,lr}
    813 	tst	lr,#1
    814 	moveq	pc,lr			@ be binary compatible with V4, yet
    815 	.word	0xe12fff1e			@ interoperable with Thumb ISA:-)
    816 #endif
    817 .size	AES_set_enc2dec_key,.-AES_set_enc2dec_key
    818 
    819 .type	AES_Td,%object
    820 .align	5
    821 AES_Td:
    822 .word	0x51f4a750, 0x7e416553, 0x1a17a4c3, 0x3a275e96
    823 .word	0x3bab6bcb, 0x1f9d45f1, 0xacfa58ab, 0x4be30393
    824 .word	0x2030fa55, 0xad766df6, 0x88cc7691, 0xf5024c25
    825 .word	0x4fe5d7fc, 0xc52acbd7, 0x26354480, 0xb562a38f
    826 .word	0xdeb15a49, 0x25ba1b67, 0x45ea0e98, 0x5dfec0e1
    827 .word	0xc32f7502, 0x814cf012, 0x8d4697a3, 0x6bd3f9c6
    828 .word	0x038f5fe7, 0x15929c95, 0xbf6d7aeb, 0x955259da
    829 .word	0xd4be832d, 0x587421d3, 0x49e06929, 0x8ec9c844
    830 .word	0x75c2896a, 0xf48e7978, 0x99583e6b, 0x27b971dd
    831 .word	0xbee14fb6, 0xf088ad17, 0xc920ac66, 0x7dce3ab4
    832 .word	0x63df4a18, 0xe51a3182, 0x97513360, 0x62537f45
    833 .word	0xb16477e0, 0xbb6bae84, 0xfe81a01c, 0xf9082b94
    834 .word	0x70486858, 0x8f45fd19, 0x94de6c87, 0x527bf8b7
    835 .word	0xab73d323, 0x724b02e2, 0xe31f8f57, 0x6655ab2a
    836 .word	0xb2eb2807, 0x2fb5c203, 0x86c57b9a, 0xd33708a5
    837 .word	0x302887f2, 0x23bfa5b2, 0x02036aba, 0xed16825c
    838 .word	0x8acf1c2b, 0xa779b492, 0xf307f2f0, 0x4e69e2a1
    839 .word	0x65daf4cd, 0x0605bed5, 0xd134621f, 0xc4a6fe8a
    840 .word	0x342e539d, 0xa2f355a0, 0x058ae132, 0xa4f6eb75
    841 .word	0x0b83ec39, 0x4060efaa, 0x5e719f06, 0xbd6e1051
    842 .word	0x3e218af9, 0x96dd063d, 0xdd3e05ae, 0x4de6bd46
    843 .word	0x91548db5, 0x71c45d05, 0x0406d46f, 0x605015ff
    844 .word	0x1998fb24, 0xd6bde997, 0x894043cc, 0x67d99e77
    845 .word	0xb0e842bd, 0x07898b88, 0xe7195b38, 0x79c8eedb
    846 .word	0xa17c0a47, 0x7c420fe9, 0xf8841ec9, 0x00000000
    847 .word	0x09808683, 0x322bed48, 0x1e1170ac, 0x6c5a724e
    848 .word	0xfd0efffb, 0x0f853856, 0x3daed51e, 0x362d3927
    849 .word	0x0a0fd964, 0x685ca621, 0x9b5b54d1, 0x24362e3a
    850 .word	0x0c0a67b1, 0x9357e70f, 0xb4ee96d2, 0x1b9b919e
    851 .word	0x80c0c54f, 0x61dc20a2, 0x5a774b69, 0x1c121a16
    852 .word	0xe293ba0a, 0xc0a02ae5, 0x3c22e043, 0x121b171d
    853 .word	0x0e090d0b, 0xf28bc7ad, 0x2db6a8b9, 0x141ea9c8
    854 .word	0x57f11985, 0xaf75074c, 0xee99ddbb, 0xa37f60fd
    855 .word	0xf701269f, 0x5c72f5bc, 0x44663bc5, 0x5bfb7e34
    856 .word	0x8b432976, 0xcb23c6dc, 0xb6edfc68, 0xb8e4f163
    857 .word	0xd731dcca, 0x42638510, 0x13972240, 0x84c61120
    858 .word	0x854a247d, 0xd2bb3df8, 0xaef93211, 0xc729a16d
    859 .word	0x1d9e2f4b, 0xdcb230f3, 0x0d8652ec, 0x77c1e3d0
    860 .word	0x2bb3166c, 0xa970b999, 0x119448fa, 0x47e96422
    861 .word	0xa8fc8cc4, 0xa0f03f1a, 0x567d2cd8, 0x223390ef
    862 .word	0x87494ec7, 0xd938d1c1, 0x8ccaa2fe, 0x98d40b36
    863 .word	0xa6f581cf, 0xa57ade28, 0xdab78e26, 0x3fadbfa4
    864 .word	0x2c3a9de4, 0x5078920d, 0x6a5fcc9b, 0x547e4662
    865 .word	0xf68d13c2, 0x90d8b8e8, 0x2e39f75e, 0x82c3aff5
    866 .word	0x9f5d80be, 0x69d0937c, 0x6fd52da9, 0xcf2512b3
    867 .word	0xc8ac993b, 0x10187da7, 0xe89c636e, 0xdb3bbb7b
    868 .word	0xcd267809, 0x6e5918f4, 0xec9ab701, 0x834f9aa8
    869 .word	0xe6956e65, 0xaaffe67e, 0x21bccf08, 0xef15e8e6
    870 .word	0xbae79bd9, 0x4a6f36ce, 0xea9f09d4, 0x29b07cd6
    871 .word	0x31a4b2af, 0x2a3f2331, 0xc6a59430, 0x35a266c0
    872 .word	0x744ebc37, 0xfc82caa6, 0xe090d0b0, 0x33a7d815
    873 .word	0xf104984a, 0x41ecdaf7, 0x7fcd500e, 0x1791f62f
    874 .word	0x764dd68d, 0x43efb04d, 0xccaa4d54, 0xe49604df
    875 .word	0x9ed1b5e3, 0x4c6a881b, 0xc12c1fb8, 0x4665517f
    876 .word	0x9d5eea04, 0x018c355d, 0xfa877473, 0xfb0b412e
    877 .word	0xb3671d5a, 0x92dbd252, 0xe9105633, 0x6dd64713
    878 .word	0x9ad7618c, 0x37a10c7a, 0x59f8148e, 0xeb133c89
    879 .word	0xcea927ee, 0xb761c935, 0xe11ce5ed, 0x7a47b13c
    880 .word	0x9cd2df59, 0x55f2733f, 0x1814ce79, 0x73c737bf
    881 .word	0x53f7cdea, 0x5ffdaa5b, 0xdf3d6f14, 0x7844db86
    882 .word	0xcaaff381, 0xb968c43e, 0x3824342c, 0xc2a3405f
    883 .word	0x161dc372, 0xbce2250c, 0x283c498b, 0xff0d9541
    884 .word	0x39a80171, 0x080cb3de, 0xd8b4e49c, 0x6456c190
    885 .word	0x7bcb8461, 0xd532b670, 0x486c5c74, 0xd0b85742
    886 @ Td4[256]
    887 .byte	0x52, 0x09, 0x6a, 0xd5, 0x30, 0x36, 0xa5, 0x38
    888 .byte	0xbf, 0x40, 0xa3, 0x9e, 0x81, 0xf3, 0xd7, 0xfb
    889 .byte	0x7c, 0xe3, 0x39, 0x82, 0x9b, 0x2f, 0xff, 0x87
    890 .byte	0x34, 0x8e, 0x43, 0x44, 0xc4, 0xde, 0xe9, 0xcb
    891 .byte	0x54, 0x7b, 0x94, 0x32, 0xa6, 0xc2, 0x23, 0x3d
    892 .byte	0xee, 0x4c, 0x95, 0x0b, 0x42, 0xfa, 0xc3, 0x4e
    893 .byte	0x08, 0x2e, 0xa1, 0x66, 0x28, 0xd9, 0x24, 0xb2
    894 .byte	0x76, 0x5b, 0xa2, 0x49, 0x6d, 0x8b, 0xd1, 0x25
    895 .byte	0x72, 0xf8, 0xf6, 0x64, 0x86, 0x68, 0x98, 0x16
    896 .byte	0xd4, 0xa4, 0x5c, 0xcc, 0x5d, 0x65, 0xb6, 0x92
    897 .byte	0x6c, 0x70, 0x48, 0x50, 0xfd, 0xed, 0xb9, 0xda
    898 .byte	0x5e, 0x15, 0x46, 0x57, 0xa7, 0x8d, 0x9d, 0x84
    899 .byte	0x90, 0xd8, 0xab, 0x00, 0x8c, 0xbc, 0xd3, 0x0a
    900 .byte	0xf7, 0xe4, 0x58, 0x05, 0xb8, 0xb3, 0x45, 0x06
    901 .byte	0xd0, 0x2c, 0x1e, 0x8f, 0xca, 0x3f, 0x0f, 0x02
    902 .byte	0xc1, 0xaf, 0xbd, 0x03, 0x01, 0x13, 0x8a, 0x6b
    903 .byte	0x3a, 0x91, 0x11, 0x41, 0x4f, 0x67, 0xdc, 0xea
    904 .byte	0x97, 0xf2, 0xcf, 0xce, 0xf0, 0xb4, 0xe6, 0x73
    905 .byte	0x96, 0xac, 0x74, 0x22, 0xe7, 0xad, 0x35, 0x85
    906 .byte	0xe2, 0xf9, 0x37, 0xe8, 0x1c, 0x75, 0xdf, 0x6e
    907 .byte	0x47, 0xf1, 0x1a, 0x71, 0x1d, 0x29, 0xc5, 0x89
    908 .byte	0x6f, 0xb7, 0x62, 0x0e, 0xaa, 0x18, 0xbe, 0x1b
    909 .byte	0xfc, 0x56, 0x3e, 0x4b, 0xc6, 0xd2, 0x79, 0x20
    910 .byte	0x9a, 0xdb, 0xc0, 0xfe, 0x78, 0xcd, 0x5a, 0xf4
    911 .byte	0x1f, 0xdd, 0xa8, 0x33, 0x88, 0x07, 0xc7, 0x31
    912 .byte	0xb1, 0x12, 0x10, 0x59, 0x27, 0x80, 0xec, 0x5f
    913 .byte	0x60, 0x51, 0x7f, 0xa9, 0x19, 0xb5, 0x4a, 0x0d
    914 .byte	0x2d, 0xe5, 0x7a, 0x9f, 0x93, 0xc9, 0x9c, 0xef
    915 .byte	0xa0, 0xe0, 0x3b, 0x4d, 0xae, 0x2a, 0xf5, 0xb0
    916 .byte	0xc8, 0xeb, 0xbb, 0x3c, 0x83, 0x53, 0x99, 0x61
    917 .byte	0x17, 0x2b, 0x04, 0x7e, 0xba, 0x77, 0xd6, 0x26
    918 .byte	0xe1, 0x69, 0x14, 0x63, 0x55, 0x21, 0x0c, 0x7d
    919 .size	AES_Td,.-AES_Td
    920 
    921 @ void AES_decrypt(const unsigned char *in, unsigned char *out,
    922 @ 		 const AES_KEY *key) {
    923 .global AES_decrypt
    924 .type   AES_decrypt,%function
    925 .align	5
    926 AES_decrypt:
    927 #if __ARM_ARCH__<7
    928 	sub	r3,pc,#8		@ AES_decrypt
    929 #else
    930 	adr	r3,AES_decrypt
    931 #endif
    932 	stmdb   sp!,{r1,r4-r12,lr}
    933 	mov	r12,r0		@ inp
    934 	mov	r11,r2
    935 	sub	r10,r3,#AES_decrypt-AES_Td		@ Td
    936 #if __ARM_ARCH__<7
    937 	ldrb	r0,[r12,#3]	@ load input data in endian-neutral
    938 	ldrb	r4,[r12,#2]	@ manner...
    939 	ldrb	r5,[r12,#1]
    940 	ldrb	r6,[r12,#0]
    941 	orr	r0,r0,r4,lsl#8
    942 	ldrb	r1,[r12,#7]
    943 	orr	r0,r0,r5,lsl#16
    944 	ldrb	r4,[r12,#6]
    945 	orr	r0,r0,r6,lsl#24
    946 	ldrb	r5,[r12,#5]
    947 	ldrb	r6,[r12,#4]
    948 	orr	r1,r1,r4,lsl#8
    949 	ldrb	r2,[r12,#11]
    950 	orr	r1,r1,r5,lsl#16
    951 	ldrb	r4,[r12,#10]
    952 	orr	r1,r1,r6,lsl#24
    953 	ldrb	r5,[r12,#9]
    954 	ldrb	r6,[r12,#8]
    955 	orr	r2,r2,r4,lsl#8
    956 	ldrb	r3,[r12,#15]
    957 	orr	r2,r2,r5,lsl#16
    958 	ldrb	r4,[r12,#14]
    959 	orr	r2,r2,r6,lsl#24
    960 	ldrb	r5,[r12,#13]
    961 	ldrb	r6,[r12,#12]
    962 	orr	r3,r3,r4,lsl#8
    963 	orr	r3,r3,r5,lsl#16
    964 	orr	r3,r3,r6,lsl#24
    965 #else
    966 	ldr	r0,[r12,#0]
    967 	ldr	r1,[r12,#4]
    968 	ldr	r2,[r12,#8]
    969 	ldr	r3,[r12,#12]
    970 #ifdef __ARMEL__
    971 	rev	r0,r0
    972 	rev	r1,r1
    973 	rev	r2,r2
    974 	rev	r3,r3
    975 #endif
    976 #endif
    977 	bl	_armv4_AES_decrypt
    978 
    979 	ldr	r12,[sp],#4		@ pop out
    980 #if __ARM_ARCH__>=7
    981 #ifdef __ARMEL__
    982 	rev	r0,r0
    983 	rev	r1,r1
    984 	rev	r2,r2
    985 	rev	r3,r3
    986 #endif
    987 	str	r0,[r12,#0]
    988 	str	r1,[r12,#4]
    989 	str	r2,[r12,#8]
    990 	str	r3,[r12,#12]
    991 #else
    992 	mov	r4,r0,lsr#24		@ write output in endian-neutral
    993 	mov	r5,r0,lsr#16		@ manner...
    994 	mov	r6,r0,lsr#8
    995 	strb	r4,[r12,#0]
    996 	strb	r5,[r12,#1]
    997 	mov	r4,r1,lsr#24
    998 	strb	r6,[r12,#2]
    999 	mov	r5,r1,lsr#16
   1000 	strb	r0,[r12,#3]
   1001 	mov	r6,r1,lsr#8
   1002 	strb	r4,[r12,#4]
   1003 	strb	r5,[r12,#5]
   1004 	mov	r4,r2,lsr#24
   1005 	strb	r6,[r12,#6]
   1006 	mov	r5,r2,lsr#16
   1007 	strb	r1,[r12,#7]
   1008 	mov	r6,r2,lsr#8
   1009 	strb	r4,[r12,#8]
   1010 	strb	r5,[r12,#9]
   1011 	mov	r4,r3,lsr#24
   1012 	strb	r6,[r12,#10]
   1013 	mov	r5,r3,lsr#16
   1014 	strb	r2,[r12,#11]
   1015 	mov	r6,r3,lsr#8
   1016 	strb	r4,[r12,#12]
   1017 	strb	r5,[r12,#13]
   1018 	strb	r6,[r12,#14]
   1019 	strb	r3,[r12,#15]
   1020 #endif
   1021 #if __ARM_ARCH__>=5
   1022 	ldmia	sp!,{r4-r12,pc}
   1023 #else
   1024 	ldmia   sp!,{r4-r12,lr}
   1025 	tst	lr,#1
   1026 	moveq	pc,lr			@ be binary compatible with V4, yet
   1027 	.word	0xe12fff1e			@ interoperable with Thumb ISA:-)
   1028 #endif
   1029 .size	AES_decrypt,.-AES_decrypt
   1030 
   1031 .type   _armv4_AES_decrypt,%function
   1032 .align	2
   1033 _armv4_AES_decrypt:
   1034 	str	lr,[sp,#-4]!		@ push lr
   1035 	ldmia	r11!,{r4-r7}
   1036 	eor	r0,r0,r4
   1037 	ldr	r12,[r11,#240-16]
   1038 	eor	r1,r1,r5
   1039 	eor	r2,r2,r6
   1040 	eor	r3,r3,r7
   1041 	sub	r12,r12,#1
   1042 	mov	lr,#255
   1043 
   1044 	and	r7,lr,r0,lsr#16
   1045 	and	r8,lr,r0,lsr#8
   1046 	and	r9,lr,r0
   1047 	mov	r0,r0,lsr#24
   1048 .Ldec_loop:
   1049 	ldr	r4,[r10,r7,lsl#2]	@ Td1[s0>>16]
   1050 	and	r7,lr,r1		@ i0
   1051 	ldr	r5,[r10,r8,lsl#2]	@ Td2[s0>>8]
   1052 	and	r8,lr,r1,lsr#16
   1053 	ldr	r6,[r10,r9,lsl#2]	@ Td3[s0>>0]
   1054 	and	r9,lr,r1,lsr#8
   1055 	ldr	r0,[r10,r0,lsl#2]	@ Td0[s0>>24]
   1056 	mov	r1,r1,lsr#24
   1057 
   1058 	ldr	r7,[r10,r7,lsl#2]	@ Td3[s1>>0]
   1059 	ldr	r8,[r10,r8,lsl#2]	@ Td1[s1>>16]
   1060 	ldr	r9,[r10,r9,lsl#2]	@ Td2[s1>>8]
   1061 	eor	r0,r0,r7,ror#24
   1062 	ldr	r1,[r10,r1,lsl#2]	@ Td0[s1>>24]
   1063 	and	r7,lr,r2,lsr#8	@ i0
   1064 	eor	r5,r8,r5,ror#8
   1065 	and	r8,lr,r2		@ i1
   1066 	eor	r6,r9,r6,ror#8
   1067 	and	r9,lr,r2,lsr#16
   1068 	ldr	r7,[r10,r7,lsl#2]	@ Td2[s2>>8]
   1069 	eor	r1,r1,r4,ror#8
   1070 	ldr	r8,[r10,r8,lsl#2]	@ Td3[s2>>0]
   1071 	mov	r2,r2,lsr#24
   1072 
   1073 	ldr	r9,[r10,r9,lsl#2]	@ Td1[s2>>16]
   1074 	eor	r0,r0,r7,ror#16
   1075 	ldr	r2,[r10,r2,lsl#2]	@ Td0[s2>>24]
   1076 	and	r7,lr,r3,lsr#16	@ i0
   1077 	eor	r1,r1,r8,ror#24
   1078 	and	r8,lr,r3,lsr#8	@ i1
   1079 	eor	r6,r9,r6,ror#8
   1080 	and	r9,lr,r3		@ i2
   1081 	ldr	r7,[r10,r7,lsl#2]	@ Td1[s3>>16]
   1082 	eor	r2,r2,r5,ror#8
   1083 	ldr	r8,[r10,r8,lsl#2]	@ Td2[s3>>8]
   1084 	mov	r3,r3,lsr#24
   1085 
   1086 	ldr	r9,[r10,r9,lsl#2]	@ Td3[s3>>0]
   1087 	eor	r0,r0,r7,ror#8
   1088 	ldr	r7,[r11],#16
   1089 	eor	r1,r1,r8,ror#16
   1090 	ldr	r3,[r10,r3,lsl#2]	@ Td0[s3>>24]
   1091 	eor	r2,r2,r9,ror#24
   1092 
   1093 	ldr	r4,[r11,#-12]
   1094 	eor	r0,r0,r7
   1095 	ldr	r5,[r11,#-8]
   1096 	eor	r3,r3,r6,ror#8
   1097 	ldr	r6,[r11,#-4]
   1098 	and	r7,lr,r0,lsr#16
   1099 	eor	r1,r1,r4
   1100 	and	r8,lr,r0,lsr#8
   1101 	eor	r2,r2,r5
   1102 	and	r9,lr,r0
   1103 	eor	r3,r3,r6
   1104 	mov	r0,r0,lsr#24
   1105 
   1106 	subs	r12,r12,#1
   1107 	bne	.Ldec_loop
   1108 
   1109 	add	r10,r10,#1024
   1110 
   1111 	ldr	r5,[r10,#0]		@ prefetch Td4
   1112 	ldr	r6,[r10,#32]
   1113 	ldr	r4,[r10,#64]
   1114 	ldr	r5,[r10,#96]
   1115 	ldr	r6,[r10,#128]
   1116 	ldr	r4,[r10,#160]
   1117 	ldr	r5,[r10,#192]
   1118 	ldr	r6,[r10,#224]
   1119 
   1120 	ldrb	r0,[r10,r0]		@ Td4[s0>>24]
   1121 	ldrb	r4,[r10,r7]		@ Td4[s0>>16]
   1122 	and	r7,lr,r1		@ i0
   1123 	ldrb	r5,[r10,r8]		@ Td4[s0>>8]
   1124 	and	r8,lr,r1,lsr#16
   1125 	ldrb	r6,[r10,r9]		@ Td4[s0>>0]
   1126 	and	r9,lr,r1,lsr#8
   1127 
   1128 	add	r1,r10,r1,lsr#24
   1129 	ldrb	r7,[r10,r7]		@ Td4[s1>>0]
   1130 	ldrb	r1,[r1]		@ Td4[s1>>24]
   1131 	ldrb	r8,[r10,r8]		@ Td4[s1>>16]
   1132 	eor	r0,r7,r0,lsl#24
   1133 	ldrb	r9,[r10,r9]		@ Td4[s1>>8]
   1134 	eor	r1,r4,r1,lsl#8
   1135 	and	r7,lr,r2,lsr#8	@ i0
   1136 	eor	r5,r5,r8,lsl#8
   1137 	and	r8,lr,r2		@ i1
   1138 	ldrb	r7,[r10,r7]		@ Td4[s2>>8]
   1139 	eor	r6,r6,r9,lsl#8
   1140 	ldrb	r8,[r10,r8]		@ Td4[s2>>0]
   1141 	and	r9,lr,r2,lsr#16
   1142 
   1143 	add	r2,r10,r2,lsr#24
   1144 	ldrb	r2,[r2]		@ Td4[s2>>24]
   1145 	eor	r0,r0,r7,lsl#8
   1146 	ldrb	r9,[r10,r9]		@ Td4[s2>>16]
   1147 	eor	r1,r8,r1,lsl#16
   1148 	and	r7,lr,r3,lsr#16	@ i0
   1149 	eor	r2,r5,r2,lsl#16
   1150 	and	r8,lr,r3,lsr#8	@ i1
   1151 	ldrb	r7,[r10,r7]		@ Td4[s3>>16]
   1152 	eor	r6,r6,r9,lsl#16
   1153 	ldrb	r8,[r10,r8]		@ Td4[s3>>8]
   1154 	and	r9,lr,r3		@ i2
   1155 
   1156 	add	r3,r10,r3,lsr#24
   1157 	ldrb	r9,[r10,r9]		@ Td4[s3>>0]
   1158 	ldrb	r3,[r3]		@ Td4[s3>>24]
   1159 	eor	r0,r0,r7,lsl#16
   1160 	ldr	r7,[r11,#0]
   1161 	eor	r1,r1,r8,lsl#8
   1162 	ldr	r4,[r11,#4]
   1163 	eor	r2,r9,r2,lsl#8
   1164 	ldr	r5,[r11,#8]
   1165 	eor	r3,r6,r3,lsl#24
   1166 	ldr	r6,[r11,#12]
   1167 
   1168 	eor	r0,r0,r7
   1169 	eor	r1,r1,r4
   1170 	eor	r2,r2,r5
   1171 	eor	r3,r3,r6
   1172 
   1173 	sub	r10,r10,#1024
   1174 	ldr	pc,[sp],#4		@ pop and return
   1175 .size	_armv4_AES_decrypt,.-_armv4_AES_decrypt
   1176 .asciz	"AES for ARMv4, CRYPTOGAMS by <appro (at) openssl.org>"
   1177 .align	2
   1178