mk-rootfs.sh

hc

2023-12-11 072de836f53be56a70cecf70b43ae43b7ce17376

 kernel/arch/arm64/crypto/sha3-ce-core.S
..
..
@@ -37,20 +37,13 @@
37
37
 	.endm
38
38
 
39
39
 	/*
40
-	 * sha3_ce_transform(u64 *st, const u8 *data, int blocks, int dg_size)
40
+	 * int sha3_ce_transform(u64 *st, const u8 *data, int blocks, int dg_size)
41
41
 	 */
42
42
 	.text
43
-ENTRY(sha3_ce_transform)
44
-	frame_push	4
45
-
46
-	mov	x19, x0
47
-	mov	x20, x1
48
-	mov	x21, x2
49
-	mov	x22, x3
50
-
51
-0:	/* load state */
52
-	add	x8, x19, #32
53
-	ld1	{ v0.1d- v3.1d}, [x19]
43
+SYM_FUNC_START(sha3_ce_transform)
44
+	/* load state */
45
+	add	x8, x0, #32
46
+	ld1	{ v0.1d- v3.1d}, [x0]
54
47
 	ld1	{ v4.1d- v7.1d}, [x8], #32
55
48
 	ld1	{ v8.1d-v11.1d}, [x8], #32
56
49
 	ld1	{v12.1d-v15.1d}, [x8], #32
..
..
@@ -58,13 +51,13 @@
58
51
 	ld1	{v20.1d-v23.1d}, [x8], #32
59
52
 	ld1	{v24.1d}, [x8]
60
53
 
61
-1:	sub	w21, w21, #1
54
+0:	sub	w2, w2, #1
62
55
 	mov	w8, #24
63
56
 	adr_l	x9, .Lsha3_rcon
64
57
 
65
58
 	/* load input */
66
-	ld1	{v25.8b-v28.8b}, [x20], #32
67
-	ld1	{v29.8b-v31.8b}, [x20], #24
59
+	ld1	{v25.8b-v28.8b}, [x1], #32
60
+	ld1	{v29.8b-v31.8b}, [x1], #24
68
61
 	eor	v0.8b, v0.8b, v25.8b
69
62
 	eor	v1.8b, v1.8b, v26.8b
70
63
 	eor	v2.8b, v2.8b, v27.8b
..
..
@@ -73,10 +66,10 @@
73
66
 	eor	v5.8b, v5.8b, v30.8b
74
67
 	eor	v6.8b, v6.8b, v31.8b
75
68
 
76
-	tbnz	x22, #6, 3f		// SHA3-512
69
+	tbnz	x3, #6, 2f		// SHA3-512
77
70
 
78
-	ld1	{v25.8b-v28.8b}, [x20], #32
79
-	ld1	{v29.8b-v30.8b}, [x20], #16
71
+	ld1	{v25.8b-v28.8b}, [x1], #32
72
+	ld1	{v29.8b-v30.8b}, [x1], #16
80
73
 	eor	 v7.8b,  v7.8b, v25.8b
81
74
 	eor	 v8.8b,  v8.8b, v26.8b
82
75
 	eor	 v9.8b,  v9.8b, v27.8b
..
..
@@ -84,34 +77,34 @@
84
77
 	eor	v11.8b, v11.8b, v29.8b
85
78
 	eor	v12.8b, v12.8b, v30.8b
86
79
 
87
-	tbnz	x22, #4, 2f		// SHA3-384 or SHA3-224
80
+	tbnz	x3, #4, 1f		// SHA3-384 or SHA3-224
88
81
 
89
82
 	// SHA3-256
90
-	ld1	{v25.8b-v28.8b}, [x20], #32
83
+	ld1	{v25.8b-v28.8b}, [x1], #32
91
84
 	eor	v13.8b, v13.8b, v25.8b
92
85
 	eor	v14.8b, v14.8b, v26.8b
93
86
 	eor	v15.8b, v15.8b, v27.8b
94
87
 	eor	v16.8b, v16.8b, v28.8b
95
-	b	4f
88
+	b	3f
96
89
 
97
-2:	tbz	x22, #2, 4f		// bit 2 cleared? SHA-384
90
+1:	tbz	x3, #2, 3f		// bit 2 cleared? SHA-384
98
91
 
99
92
 	// SHA3-224
100
-	ld1	{v25.8b-v28.8b}, [x20], #32
101
-	ld1	{v29.8b}, [x20], #8
93
+	ld1	{v25.8b-v28.8b}, [x1], #32
94
+	ld1	{v29.8b}, [x1], #8
102
95
 	eor	v13.8b, v13.8b, v25.8b
103
96
 	eor	v14.8b, v14.8b, v26.8b
104
97
 	eor	v15.8b, v15.8b, v27.8b
105
98
 	eor	v16.8b, v16.8b, v28.8b
106
99
 	eor	v17.8b, v17.8b, v29.8b
107
-	b	4f
100
+	b	3f
108
101
 
109
102
 	// SHA3-512
110
-3:	ld1	{v25.8b-v26.8b}, [x20], #16
103
+2:	ld1	{v25.8b-v26.8b}, [x1], #16
111
104
 	eor	 v7.8b,  v7.8b, v25.8b
112
105
 	eor	 v8.8b,  v8.8b, v26.8b
113
106
 
114
-4:	sub	w8, w8, #1
107
+3:	sub	w8, w8, #1
115
108
 
116
109
 	eor3	v29.16b,  v4.16b,  v9.16b, v14.16b
117
110
 	eor3	v26.16b,  v1.16b,  v6.16b, v11.16b
..
..
@@ -190,35 +183,21 @@
190
183
 
191
184
 	eor	 v0.16b,  v0.16b, v31.16b
192
185
 
193
-	cbnz	w8, 4b
194
-	cbz	w21, 5f
195
-
196
-	if_will_cond_yield_neon
197
-	add	x8, x19, #32
198
-	st1	{ v0.1d- v3.1d}, [x19]
199
-	st1	{ v4.1d- v7.1d}, [x8], #32
200
-	st1	{ v8.1d-v11.1d}, [x8], #32
201
-	st1	{v12.1d-v15.1d}, [x8], #32
202
-	st1	{v16.1d-v19.1d}, [x8], #32
203
-	st1	{v20.1d-v23.1d}, [x8], #32
204
-	st1	{v24.1d}, [x8]
205
-	do_cond_yield_neon
206
-	b		0b
207
-	endif_yield_neon
208
-
209
-	b	1b
186
+	cbnz	w8, 3b
187
+	cond_yield 4f, x8, x9
188
+	cbnz	w2, 0b
210
189
 
211
190
 	/* save state */
212
-5:	st1	{ v0.1d- v3.1d}, [x19], #32
213
-	st1	{ v4.1d- v7.1d}, [x19], #32
214
-	st1	{ v8.1d-v11.1d}, [x19], #32
215
-	st1	{v12.1d-v15.1d}, [x19], #32
216
-	st1	{v16.1d-v19.1d}, [x19], #32
217
-	st1	{v20.1d-v23.1d}, [x19], #32
218
-	st1	{v24.1d}, [x19]
219
-	frame_pop
191
+4:	st1	{ v0.1d- v3.1d}, [x0], #32
192
+	st1	{ v4.1d- v7.1d}, [x0], #32
193
+	st1	{ v8.1d-v11.1d}, [x0], #32
194
+	st1	{v12.1d-v15.1d}, [x0], #32
195
+	st1	{v16.1d-v19.1d}, [x0], #32
196
+	st1	{v20.1d-v23.1d}, [x0], #32
197
+	st1	{v24.1d}, [x0]
198
+	mov	w0, w2
220
199
 	ret
221
-ENDPROC(sha3_ce_transform)
200
+SYM_FUNC_END(sha3_ce_transform)
222
201
 
223
202
 	.section	".rodata", "a"
224
203
 	.align		8