armv4cpuid.pl 5.3 KB
Newer Older
R
Rich Salz 已提交
1
#! /usr/bin/env perl
M
Matt Caswell 已提交
2
# Copyright 2015-2020 The OpenSSL Project Authors. All Rights Reserved.
R
Rich Salz 已提交
3 4 5 6 7 8
#
# Licensed under the OpenSSL license (the "License").  You may not use
# this file except in compliance with the License.  You can obtain a copy
# in the file LICENSE in the source distribution or at
# https://www.openssl.org/source/license.html

9 10 11 12 13 14 15 16 17 18 19 20 21

$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.=<<___;
22 23 24
#include "arm_arch.h"

.text
25 26 27 28
#if defined(__thumb2__) && !defined(__APPLE__)
.syntax	unified
.thumb
#else
29
.code	32
30
#undef	__thumb2__
31
#endif
32

33
.align	5
34 35 36 37 38 39 40 41 42 43
.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
44
	bx	lr
45 46 47 48 49 50 51 52 53 54 55 56 57 58 59
#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]
60
	add	r2,r2,r5
61 62 63 64 65 66 67 68 69 70 71 72 73 74
	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
75 76 77
#ifdef	__thumb2__
	itt	hs
#endif
78
	subhs	r1,r1,#4
79 80 81 82 83
	bhs	.Lot
	cmp	r1,#0
	beq	.Lcleanse_done
.Little:
	strb	ip,[r0],#1
84
	subs	r1,r1,#1
85 86 87 88 89 90
	bhi	.Little
	b	.Lcleanse_done

.Lot:	tst	r0,#3
	beq	.Laligned
	strb	ip,[r0],#1
91
	sub	r1,r1,#1
92 93 94
	b	.Lot
.Laligned:
	str	ip,[r0],#4
95
	subs	r1,r1,#4
96
	bhs	.Laligned
97
	adds	r1,r1,#4
98 99
	bne	.Little
.Lcleanse_done:
100 101 102
#if __ARM_ARCH__>=5
	bx	lr
#else
103 104 105
	tst	lr,#1
	moveq	pc,lr
	.word	0xe12fff1e	@ bx	lr
106
#endif
107 108
.size	OPENSSL_cleanse,.-OPENSSL_cleanse

A
Andy Polyakov 已提交
109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127
.global	CRYPTO_memcmp
.type	CRYPTO_memcmp,%function
.align	4
CRYPTO_memcmp:
	eor	ip,ip,ip
	cmp	r2,#0
	beq	.Lno_data
	stmdb	sp!,{r4,r5}

.Loop_cmp:
	ldrb	r4,[r0],#1
	ldrb	r5,[r1],#1
	eor	r4,r4,r5
	orr	ip,ip,r4
	subs	r2,r2,#1
	bne	.Loop_cmp

	ldmia	sp!,{r4,r5}
.Lno_data:
128
	rsb	r0,ip,#0
A
Andy Polyakov 已提交
129 130 131 132 133 134 135 136 137 138
	mov	r0,r0,lsr#31
#if __ARM_ARCH__>=5
	bx	lr
#else
	tst	lr,#1
	moveq	pc,lr
	.word	0xe12fff1e	@ bx	lr
#endif
.size	CRYPTO_memcmp,.-CRYPTO_memcmp

139 140 141 142 143 144 145 146 147 148 149 150 151 152 153
#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:
154 155 156
#ifdef	__APPLE__
	mrrc	p15,0,r0,r1,c14		@ CNTPCT
#else
157
	mrrc	p15,1,r0,r1,c14		@ CNTVCT
158
#endif
159 160 161 162 163 164
	bx	lr
.size	_armv7_tick,.-_armv7_tick

.global	_armv8_aes_probe
.type	_armv8_aes_probe,%function
_armv8_aes_probe:
165 166 167
#if defined(__thumb2__) && !defined(__APPLE__)
	.byte	0xb0,0xff,0x00,0x03	@ aese.8	q0,q0
#else
168
	.byte	0x00,0x03,0xb0,0xf3	@ aese.8	q0,q0
169
#endif
170 171 172 173 174 175
	bx	lr
.size	_armv8_aes_probe,.-_armv8_aes_probe

.global	_armv8_sha1_probe
.type	_armv8_sha1_probe,%function
_armv8_sha1_probe:
176 177 178
#if defined(__thumb2__) && !defined(__APPLE__)
	.byte	0x00,0xef,0x40,0x0c	@ sha1c.32	q0,q0,q0
#else
179
	.byte	0x40,0x0c,0x00,0xf2	@ sha1c.32	q0,q0,q0
180
#endif
181 182 183 184 185 186
	bx	lr
.size	_armv8_sha1_probe,.-_armv8_sha1_probe

.global	_armv8_sha256_probe
.type	_armv8_sha256_probe,%function
_armv8_sha256_probe:
187 188 189
#if defined(__thumb2__) && !defined(__APPLE__)
	.byte	0x00,0xff,0x40,0x0c	@ sha256h.32	q0,q0,q0
#else
190
	.byte	0x40,0x0c,0x00,0xf3	@ sha256h.32	q0,q0,q0
191
#endif
192 193 194 195 196
	bx	lr
.size	_armv8_sha256_probe,.-_armv8_sha256_probe
.global	_armv8_pmull_probe
.type	_armv8_pmull_probe,%function
_armv8_pmull_probe:
197 198 199
#if defined(__thumb2__) && !defined(__APPLE__)
	.byte	0xa0,0xef,0x00,0x0e	@ vmull.p64	q0,d0,d0
#else
200
	.byte	0x00,0x0e,0xa0,0xf2	@ vmull.p64	q0,d0,d0
201
#endif
202 203 204 205
	bx	lr
.size	_armv8_pmull_probe,.-_armv8_pmull_probe
#endif

206 207 208
.global	OPENSSL_wipe_cpu
.type	OPENSSL_wipe_cpu,%function
OPENSSL_wipe_cpu:
209
#if __ARM_MAX_ARCH__>=7
210 211 212
	ldr	r0,.LOPENSSL_armcap
	adr	r1,.LOPENSSL_armcap
	ldr	r0,[r1,r0]
213 214 215
#ifdef	__APPLE__
	ldr	r0,[r0]
#endif
216
#endif
217 218 219
	eor	r2,r2,r2
	eor	r3,r3,r3
	eor	ip,ip,ip
220
#if __ARM_MAX_ARCH__>=7
221 222
	tst	r0,#1
	beq	.Lwipe_done
223 224 225 226 227 228 229 230 231 232 233 234
	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
235
.Lwipe_done:
236
#endif
237
	mov	r0,sp
238 239 240
#if __ARM_ARCH__>=5
	bx	lr
#else
241 242 243
	tst	lr,#1
	moveq	pc,lr
	.word	0xe12fff1e	@ bx	lr
244
#endif
245 246 247 248 249 250
.size	OPENSSL_wipe_cpu,.-OPENSSL_wipe_cpu

.global	OPENSSL_instrument_bus
.type	OPENSSL_instrument_bus,%function
OPENSSL_instrument_bus:
	eor	r0,r0,r0
251 252 253
#if __ARM_ARCH__>=5
	bx	lr
#else
254 255 256
	tst	lr,#1
	moveq	pc,lr
	.word	0xe12fff1e	@ bx	lr
257
#endif
258 259 260 261 262 263
.size	OPENSSL_instrument_bus,.-OPENSSL_instrument_bus

.global	OPENSSL_instrument_bus2
.type	OPENSSL_instrument_bus2,%function
OPENSSL_instrument_bus2:
	eor	r0,r0,r0
264 265 266
#if __ARM_ARCH__>=5
	bx	lr
#else
267 268 269
	tst	lr,#1
	moveq	pc,lr
	.word	0xe12fff1e	@ bx	lr
270
#endif
271 272 273
.size	OPENSSL_instrument_bus2,.-OPENSSL_instrument_bus2

.align	5
274
#if __ARM_MAX_ARCH__>=7
275
.LOPENSSL_armcap:
276
.word	OPENSSL_armcap_P-.
277
#endif
278 279 280 281 282 283 284 285 286 287 288 289 290 291 292
#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
293 294 295
___

print $code;
296
close STDOUT or die "error closing STDOUT: $!";