crypto/aes/asm/aesni-x86[_64].pl update, up to 14% improvement on
[openssl.git] / crypto / armv4cpuid.S
1 #include "arm_arch.h"
2
3 .text
4 .code   32
5
6 .align  5
7 .global _armv7_neon_probe
8 .type   _armv7_neon_probe,%function
9 _armv7_neon_probe:
10         .word   0xf26ee1fe      @ vorr  q15,q15,q15
11         .word   0xe12fff1e      @ bx    lr
12 .size   _armv7_neon_probe,.-_armv7_neon_probe
13
14 .global _armv7_tick
15 .type   _armv7_tick,%function
16 _armv7_tick:
17         mrc     p15,0,r0,c9,c13,0
18         .word   0xe12fff1e      @ bx    lr
19 .size   _armv7_tick,.-_armv7_tick
20
21 .global OPENSSL_atomic_add
22 .type   OPENSSL_atomic_add,%function
23 OPENSSL_atomic_add:
24 #if __ARM_ARCH__>=6
25 .Ladd:  ldrex   r2,[r0]
26         add     r3,r2,r1
27         strex   r2,r3,[r0]
28         cmp     r2,#0
29         bne     .Ladd
30         mov     r0,r3
31         .word   0xe12fff1e      @ bx    lr
32 #else
33         stmdb   sp!,{r4-r6,lr}
34         ldr     r2,.Lspinlock
35         adr     r3,.Lspinlock
36         mov     r4,r0
37         mov     r5,r1
38         add     r6,r3,r2        @ &spinlock
39         b       .+8
40 .Lspin: bl      sched_yield
41         mov     r0,#-1
42         swp     r0,r0,[r6]
43         cmp     r0,#0
44         bne     .Lspin
45
46         ldr     r2,[r4]
47         add     r2,r2,r5
48         str     r2,[r4]
49         str     r0,[r6]         @ release spinlock
50         ldmia   sp!,{r4-r6,lr}
51         tst     lr,#1
52         moveq   pc,lr
53         .word   0xe12fff1e      @ bx    lr
54 #endif
55 .size   OPENSSL_atomic_add,.-OPENSSL_atomic_add
56
57 .global OPENSSL_cleanse
58 .type   OPENSSL_cleanse,%function
59 OPENSSL_cleanse:
60         eor     ip,ip,ip
61         cmp     r1,#7
62         subhs   r1,r1,#4
63         bhs     .Lot
64         cmp     r1,#0
65         beq     .Lcleanse_done
66 .Little:
67         strb    ip,[r0],#1
68         subs    r1,r1,#1
69         bhi     .Little
70         b       .Lcleanse_done
71
72 .Lot:   tst     r0,#3
73         beq     .Laligned
74         strb    ip,[r0],#1
75         sub     r1,r1,#1
76         b       .Lot
77 .Laligned:
78         str     ip,[r0],#4
79         subs    r1,r1,#4
80         bhs     .Laligned
81         adds    r1,r1,#4
82         bne     .Little
83 .Lcleanse_done:
84         tst     lr,#1
85         moveq   pc,lr
86         .word   0xe12fff1e      @ bx    lr
87 .size   OPENSSL_cleanse,.-OPENSSL_cleanse
88
89 .global OPENSSL_wipe_cpu
90 .type   OPENSSL_wipe_cpu,%function
91 OPENSSL_wipe_cpu:
92         ldr     r0,.LOPENSSL_armcap
93         adr     r1,.LOPENSSL_armcap
94         ldr     r0,[r1,r0]
95         eor     r2,r2,r2
96         eor     r3,r3,r3
97         eor     ip,ip,ip
98         tst     r0,#1
99         beq     .Lwipe_done
100         .word   0xf3000150      @ veor    q0, q0, q0
101         .word   0xf3022152      @ veor    q1, q1, q1
102         .word   0xf3044154      @ veor    q2, q2, q2
103         .word   0xf3066156      @ veor    q3, q3, q3
104         .word   0xf34001f0      @ veor    q8, q8, q8
105         .word   0xf34221f2      @ veor    q9, q9, q9
106         .word   0xf34441f4      @ veor    q10, q10, q10
107         .word   0xf34661f6      @ veor    q11, q11, q11
108         .word   0xf34881f8      @ veor    q12, q12, q12
109         .word   0xf34aa1fa      @ veor    q13, q13, q13
110         .word   0xf34cc1fc      @ veor    q14, q14, q14
111         .word   0xf34ee1fe      @ veor    q15, q15, q15
112 .Lwipe_done:
113         mov     r0,sp
114         tst     lr,#1
115         moveq   pc,lr
116         .word   0xe12fff1e      @ bx    lr
117 .size   OPENSSL_wipe_cpu,.-OPENSSL_wipe_cpu
118
119 .global OPENSSL_instrument_bus
120 .type   OPENSSL_instrument_bus,%function
121 OPENSSL_instrument_bus:
122         eor     r0,r0,r0
123         tst     lr,#1
124         moveq   pc,lr
125         .word   0xe12fff1e      @ bx    lr
126 .size   OPENSSL_instrument_bus,.-OPENSSL_instrument_bus
127
128 .global OPENSSL_instrument_bus2
129 .type   OPENSSL_instrument_bus2,%function
130 OPENSSL_instrument_bus2:
131         eor     r0,r0,r0
132         tst     lr,#1
133         moveq   pc,lr
134         .word   0xe12fff1e      @ bx    lr
135 .size   OPENSSL_instrument_bus2,.-OPENSSL_instrument_bus2
136
137 .align  5
138 .LOPENSSL_armcap:
139 .word   OPENSSL_armcap_P-.LOPENSSL_armcap
140 #if __ARM_ARCH__>=6
141 .align  5
142 #else
143 .Lspinlock:
144 .word   atomic_add_spinlock-.Lspinlock
145 .align  5
146
147 .data
148 .align  2
149 atomic_add_spinlock:
150 .word   0
151 #endif
152
153 .comm   OPENSSL_armcap_P,4,4
154 .hidden OPENSSL_armcap_P