armv4cpuid.pl 4.6 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14
#!/usr/bin/env perl

$flavour = shift;
$output  = shift;

$0 =~ m/(.*[\/\\])[^\/\\]+$/; $dir=$1;
( $xlate="${dir}arm-xlate.pl" and -f $xlate ) or
( $xlate="${dir}perlasm/arm-xlate.pl" and -f $xlate) or
die "can't locate arm-xlate.pl";

open OUT,"| \"$^X\" $xlate $flavour $output";
*STDOUT=*OUT;

$code.=<<___;
15 16 17
#include "arm_arch.h"

.text
18 19 20 21
#if defined(__thumb2__) && !defined(__APPLE__)
.syntax	unified
.thumb
#else
22
.code	32
23
#endif
24

25
.align	5
26 27 28 29 30 31 32 33 34 35
.global	OPENSSL_atomic_add
.type	OPENSSL_atomic_add,%function
OPENSSL_atomic_add:
#if __ARM_ARCH__>=6
.Ladd:	ldrex	r2,[r0]
	add	r3,r2,r1
	strex	r2,r3,[r0]
	cmp	r2,#0
	bne	.Ladd
	mov	r0,r3
36
	bx	lr
37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
#else
	stmdb	sp!,{r4-r6,lr}
	ldr	r2,.Lspinlock
	adr	r3,.Lspinlock
	mov	r4,r0
	mov	r5,r1
	add	r6,r3,r2	@ &spinlock
	b	.+8
.Lspin:	bl	sched_yield
	mov	r0,#-1
	swp	r0,r0,[r6]
	cmp	r0,#0
	bne	.Lspin

	ldr	r2,[r4]
52
	add	r2,r2,r5
53 54 55 56 57 58 59 60 61 62 63 64 65 66
	str	r2,[r4]
	str	r0,[r6]		@ release spinlock
	ldmia	sp!,{r4-r6,lr}
	tst	lr,#1
	moveq	pc,lr
	.word	0xe12fff1e	@ bx	lr
#endif
.size	OPENSSL_atomic_add,.-OPENSSL_atomic_add

.global	OPENSSL_cleanse
.type	OPENSSL_cleanse,%function
OPENSSL_cleanse:
	eor	ip,ip,ip
	cmp	r1,#7
67 68 69
#ifdef	__thumb2__
	itt	hs
#endif
70
	subhs	r1,r1,#4
71 72 73 74 75
	bhs	.Lot
	cmp	r1,#0
	beq	.Lcleanse_done
.Little:
	strb	ip,[r0],#1
76
	subs	r1,r1,#1
77 78 79 80 81 82
	bhi	.Little
	b	.Lcleanse_done

.Lot:	tst	r0,#3
	beq	.Laligned
	strb	ip,[r0],#1
83
	sub	r1,r1,#1
84 85 86
	b	.Lot
.Laligned:
	str	ip,[r0],#4
87
	subs	r1,r1,#4
88
	bhs	.Laligned
89
	adds	r1,r1,#4
90 91
	bne	.Little
.Lcleanse_done:
92 93 94
#if __ARM_ARCH__>=5
	bx	lr
#else
95 96 97
	tst	lr,#1
	moveq	pc,lr
	.word	0xe12fff1e	@ bx	lr
98
#endif
99 100
.size	OPENSSL_cleanse,.-OPENSSL_cleanse

101 102 103 104 105 106 107 108 109 110 111 112 113 114 115
#if __ARM_MAX_ARCH__>=7
.arch	armv7-a
.fpu	neon

.align	5
.global	_armv7_neon_probe
.type	_armv7_neon_probe,%function
_armv7_neon_probe:
	vorr	q0,q0,q0
	bx	lr
.size	_armv7_neon_probe,.-_armv7_neon_probe

.global	_armv7_tick
.type	_armv7_tick,%function
_armv7_tick:
116 117 118
#ifdef	__APPLE__
	mrrc	p15,0,r0,r1,c14		@ CNTPCT
#else
119
	mrrc	p15,1,r0,r1,c14		@ CNTVCT
120
#endif
121 122 123 124 125 126
	bx	lr
.size	_armv7_tick,.-_armv7_tick

.global	_armv8_aes_probe
.type	_armv8_aes_probe,%function
_armv8_aes_probe:
127 128 129
#if defined(__thumb2__) && !defined(__APPLE__)
	.byte	0xb0,0xff,0x00,0x03	@ aese.8	q0,q0
#else
130
	.byte	0x00,0x03,0xb0,0xf3	@ aese.8	q0,q0
131
#endif
132 133 134 135 136 137
	bx	lr
.size	_armv8_aes_probe,.-_armv8_aes_probe

.global	_armv8_sha1_probe
.type	_armv8_sha1_probe,%function
_armv8_sha1_probe:
138 139 140
#if defined(__thumb2__) && !defined(__APPLE__)
	.byte	0x00,0xef,0x40,0x0c	@ sha1c.32	q0,q0,q0
#else
141
	.byte	0x40,0x0c,0x00,0xf2	@ sha1c.32	q0,q0,q0
142
#endif
143 144 145 146 147 148
	bx	lr
.size	_armv8_sha1_probe,.-_armv8_sha1_probe

.global	_armv8_sha256_probe
.type	_armv8_sha256_probe,%function
_armv8_sha256_probe:
149 150 151
#if defined(__thumb2__) && !defined(__APPLE__)
	.byte	0x00,0xff,0x40,0x0c	@ sha256h.32	q0,q0,q0
#else
152
	.byte	0x40,0x0c,0x00,0xf3	@ sha256h.32	q0,q0,q0
153
#endif
154 155 156 157 158
	bx	lr
.size	_armv8_sha256_probe,.-_armv8_sha256_probe
.global	_armv8_pmull_probe
.type	_armv8_pmull_probe,%function
_armv8_pmull_probe:
159 160 161
#if defined(__thumb2__) && !defined(__APPLE__)
	.byte	0xa0,0xef,0x00,0x0e	@ vmull.p64	q0,d0,d0
#else
162
	.byte	0x00,0x0e,0xa0,0xf2	@ vmull.p64	q0,d0,d0
163
#endif
164 165 166 167
	bx	lr
.size	_armv8_pmull_probe,.-_armv8_pmull_probe
#endif

168 169 170
.global	OPENSSL_wipe_cpu
.type	OPENSSL_wipe_cpu,%function
OPENSSL_wipe_cpu:
171
#if __ARM_MAX_ARCH__>=7
172 173 174
	ldr	r0,.LOPENSSL_armcap
	adr	r1,.LOPENSSL_armcap
	ldr	r0,[r1,r0]
175 176 177
#ifdef	__APPLE__
	ldr	r0,[r0]
#endif
178
#endif
179 180 181
	eor	r2,r2,r2
	eor	r3,r3,r3
	eor	ip,ip,ip
182
#if __ARM_MAX_ARCH__>=7
183 184
	tst	r0,#1
	beq	.Lwipe_done
185 186 187 188 189 190 191 192 193 194 195 196
	veor	q0, q0, q0
	veor	q1, q1, q1
	veor	q2, q2, q2
	veor	q3, q3, q3
	veor	q8, q8, q8
	veor	q9, q9, q9
	veor	q10, q10, q10
	veor	q11, q11, q11
	veor	q12, q12, q12
	veor	q13, q13, q13
	veor	q14, q14, q14
	veor	q15, q15, q15
197
.Lwipe_done:
198
#endif
199
	mov	r0,sp
200 201 202
#if __ARM_ARCH__>=5
	bx	lr
#else
203 204 205
	tst	lr,#1
	moveq	pc,lr
	.word	0xe12fff1e	@ bx	lr
206
#endif
207 208 209 210 211 212
.size	OPENSSL_wipe_cpu,.-OPENSSL_wipe_cpu

.global	OPENSSL_instrument_bus
.type	OPENSSL_instrument_bus,%function
OPENSSL_instrument_bus:
	eor	r0,r0,r0
213 214 215
#if __ARM_ARCH__>=5
	bx	lr
#else
216 217 218
	tst	lr,#1
	moveq	pc,lr
	.word	0xe12fff1e	@ bx	lr
219
#endif
220 221 222 223 224 225
.size	OPENSSL_instrument_bus,.-OPENSSL_instrument_bus

.global	OPENSSL_instrument_bus2
.type	OPENSSL_instrument_bus2,%function
OPENSSL_instrument_bus2:
	eor	r0,r0,r0
226 227 228
#if __ARM_ARCH__>=5
	bx	lr
#else
229 230 231
	tst	lr,#1
	moveq	pc,lr
	.word	0xe12fff1e	@ bx	lr
232
#endif
233 234 235
.size	OPENSSL_instrument_bus2,.-OPENSSL_instrument_bus2

.align	5
236
#if __ARM_MAX_ARCH__>=7
237
.LOPENSSL_armcap:
238
.word	OPENSSL_armcap_P-.
239
#endif
240 241 242 243 244 245 246 247 248 249 250 251 252 253 254
#if __ARM_ARCH__>=6
.align	5
#else
.Lspinlock:
.word	atomic_add_spinlock-.Lspinlock
.align	5

.data
.align	2
atomic_add_spinlock:
.word	0
#endif

.comm	OPENSSL_armcap_P,4,4
.hidden	OPENSSL_armcap_P
255 256 257 258
___

print $code;
close STDOUT;