[picodrive.git] / Pico / Misc_amips.s

# vim:filetype=mips

.set noreorder
.set noat

.text
.align 4

.globl memset32 # int *dest, int c, int count

memset32:
ms32_aloop:
    andi    $t0, $a0, 0x3f
    beqz    $t0, ms32_bloop_prep
    nop
    sw      $a1, 0($a0)
    addiu   $a2, -1
    beqz    $a2, ms32_return
    addiu   $a0, 4
    j       ms32_aloop
    nop

ms32_bloop_prep:
    srl     $t0, $a2, 4    # we will do 64 bytes per iteration (cache line)
    beqz    $t0, ms32_bloop_end

ms32_bloop:
    addiu   $t0, -1
    cache   0x18, ($a0)    # create dirty exclusive
    sw      $a1, 0x00($a0)
    sw      $a1, 0x04($a0)
    sw      $a1, 0x08($a0)
    sw      $a1, 0x0c($a0)
    sw      $a1, 0x10($a0)
    sw      $a1, 0x14($a0)
    sw      $a1, 0x18($a0)
    sw      $a1, 0x1c($a0)
    sw      $a1, 0x20($a0)
    sw      $a1, 0x24($a0)
    sw      $a1, 0x28($a0)
    sw      $a1, 0x2c($a0)
    sw      $a1, 0x30($a0)
    sw      $a1, 0x34($a0)
    sw      $a1, 0x38($a0)
    sw      $a1, 0x3c($a0)
    bnez    $t0, ms32_bloop
    addiu   $a0, 0x40

ms32_bloop_end:
    andi    $a2, $a2, 0x0f
    beqz    $a2, ms32_return

ms32_cloop:
    addiu   $a2, -1
    sw      $a1, 0($a0)
    bnez    $a2, ms32_cloop
    addiu   $a0, 4

ms32_return:
    jr      $ra
    nop


.globl memset32_uncached # int *dest, int c, int count

memset32_uncached:
    srl     $t0, $a2, 3    # we will do 32 bytes per iteration
    beqz    $t0, ms32u_bloop_end

ms32u_bloop:
    addiu   $t0, -1
    sw      $a1, 0x00($a0)
    sw      $a1, 0x04($a0)
    sw      $a1, 0x08($a0)
    sw      $a1, 0x0c($a0)
    sw      $a1, 0x10($a0)
    sw      $a1, 0x14($a0)
    sw      $a1, 0x18($a0)
    sw      $a1, 0x1c($a0)
    bnez    $t0, ms32u_bloop
    addiu   $a0, 0x20

ms32u_bloop_end:
    andi    $a2, $a2, 0x0f
    beqz    $a2, ms32u_return

ms32u_cloop:
    addiu   $a2, -1
    sw      $a1, 0($a0)
    bnez    $a2, ms32u_cloop
    addiu   $a0, 4

ms32u_return:
    jr      $ra
    nop


.globl memcpy32 # int *dest, int *src, int count

memcpy32:
mc32_aloop:
    andi    $t0, $a0, 0x3f
    beqz    $t0, mc32_bloop_prep
    nop
    lw      $t1, 0($a1)
    addiu   $a2, -1
    sw      $t1, 0($a0)
    beqz    $a2, mc32_return
    addiu   $a0, 4
    j       mc32_aloop
    addiu   $a1, 4

mc32_bloop_prep:
    srl     $t0, $a2, 4    # we will do 64 bytes per iteration (cache line)
    beqz    $t0, mc32_bloop_end

mc32_bloop:
    addiu   $t0, -1
    cache   0x18, ($a0)    # create dirty exclusive
    lw      $t2, 0x00($a1)
    lw      $t3, 0x04($a1)
    lw      $t4, 0x08($a1)
    lw      $t5, 0x0c($a1)
    lw      $t6, 0x10($a1)
    lw      $t7, 0x14($a1)
    lw      $t8, 0x18($a1)
    lw      $t9, 0x1c($a1)
    sw      $t2, 0x00($a0)
    sw      $t3, 0x04($a0)
    sw      $t4, 0x08($a0)
    sw      $t5, 0x0c($a0)
    sw      $t6, 0x10($a0)
    sw      $t7, 0x14($a0)
    sw      $t8, 0x18($a0)
    sw      $t9, 0x1c($a0)
    lw      $t2, 0x20($a1)
    lw      $t3, 0x24($a1)
    lw      $t4, 0x28($a1)
    lw      $t5, 0x2c($a1)
    lw      $t6, 0x30($a1)
    lw      $t7, 0x34($a1)
    lw      $t8, 0x38($a1)
    lw      $t9, 0x3c($a1)
    sw      $t2, 0x20($a0)
    sw      $t3, 0x24($a0)
    sw      $t4, 0x28($a0)
    sw      $t5, 0x2c($a0)
    sw      $t6, 0x30($a0)
    sw      $t7, 0x34($a0)
    sw      $t8, 0x38($a0)
    sw      $t9, 0x3c($a0)
    addiu   $a0, 0x40
    bnez    $t0, mc32_bloop
    addiu   $a1, 0x40

mc32_bloop_end:
    andi    $a2, $a2, 0x0f
    beqz    $a2, mc32_return

mc32_cloop:
    lw      $t1, 0($a1)
    addiu   $a2, -1
    addiu   $a1, 4
    sw      $t1, 0($a0)
    bnez    $a2, mc32_cloop
    addiu   $a0, 4

mc32_return:
    jr      $ra
    nop
Commit	Line	Data
b542be46	1	# vim:filetype=mips
	2
	3	.set noreorder
	4	.set noat
	5
	6	.text
	7	.align 4
	8
	9	.globl memset32 # int *dest, int c, int count
	10
	11	memset32:
	12	ms32_aloop:
	13	andi $t0, $a0, 0x3f
	14	beqz $t0, ms32_bloop_prep
	15	nop
	16	sw $a1, 0($a0)
	17	addiu $a2, -1
	18	beqz $a2, ms32_return
	19	addiu $a0, 4
	20	j ms32_aloop
	21	nop
	22
	23	ms32_bloop_prep:
	24	srl $t0, $a2, 4 # we will do 64 bytes per iteration (cache line)
	25	beqz $t0, ms32_bloop_end
	26
	27	ms32_bloop:
	28	addiu $t0, -1
	29	cache 0x18, ($a0) # create dirty exclusive
	30	sw $a1, 0x00($a0)
	31	sw $a1, 0x04($a0)
	32	sw $a1, 0x08($a0)
	33	sw $a1, 0x0c($a0)
	34	sw $a1, 0x10($a0)
	35	sw $a1, 0x14($a0)
	36	sw $a1, 0x18($a0)
	37	sw $a1, 0x1c($a0)
	38	sw $a1, 0x20($a0)
	39	sw $a1, 0x24($a0)
	40	sw $a1, 0x28($a0)
	41	sw $a1, 0x2c($a0)
	42	sw $a1, 0x30($a0)
	43	sw $a1, 0x34($a0)
	44	sw $a1, 0x38($a0)
	45	sw $a1, 0x3c($a0)
	46	bnez $t0, ms32_bloop
	47	addiu $a0, 0x40
	48
	49	ms32_bloop_end:
	50	andi $a2, $a2, 0x0f
	51	beqz $a2, ms32_return
	52
	53	ms32_cloop:
	54	addiu $a2, -1
	55	sw $a1, 0($a0)
	56	bnez $a2, ms32_cloop
	57	addiu $a0, 4
	58
	59	ms32_return:
	60	jr $ra
	61	nop
	62
	63
	64	.globl memset32_uncached # int *dest, int c, int count
65
66	memset32_uncached:
67	srl $t0, $a2, 3 # we will do 32 bytes per iteration
68	beqz $t0, ms32u_bloop_end
69
70	ms32u_bloop:
71	addiu $t0, -1
72	sw $a1, 0x00($a0)
73	sw $a1, 0x04($a0)
74	sw $a1, 0x08($a0)
75	sw $a1, 0x0c($a0)
76	sw $a1, 0x10($a0)
77	sw $a1, 0x14($a0)
78	sw $a1, 0x18($a0)
79	sw $a1, 0x1c($a0)
80	bnez $t0, ms32u_bloop
81	addiu $a0, 0x20
82
83	ms32u_bloop_end:
84	andi $a2, $a2, 0x0f
85	beqz $a2, ms32u_return
86
87	ms32u_cloop:
88	addiu $a2, -1
89	sw $a1, 0($a0)
90	bnez $a2, ms32u_cloop
91	addiu $a0, 4
92
93	ms32u_return:
94	jr $ra
95	nop
96
97
98	.globl memcpy32 # int dest, int src, int count
99
100	memcpy32:
101	mc32_aloop:
102	andi $t0, $a0, 0x3f
103	beqz $t0, mc32_bloop_prep
104	nop
105	lw $t1, 0($a1)
106	addiu $a2, -1
107	sw $t1, 0($a0)
108	beqz $a2, mc32_return
109	addiu $a0, 4
110	j mc32_aloop
111	addiu $a1, 4
112
113	mc32_bloop_prep:
114	srl $t0, $a2, 4 # we will do 64 bytes per iteration (cache line)
115	beqz $t0, mc32_bloop_end
116
117	mc32_bloop:
118	addiu $t0, -1
119	cache 0x18, ($a0) # create dirty exclusive
120	lw $t2, 0x00($a1)
121	lw $t3, 0x04($a1)
122	lw $t4, 0x08($a1)
123	lw $t5, 0x0c($a1)
124	lw $t6, 0x10($a1)
125	lw $t7, 0x14($a1)
126	lw $t8, 0x18($a1)
127	lw $t9, 0x1c($a1)
128	sw $t2, 0x00($a0)
129	sw $t3, 0x04($a0)
130	sw $t4, 0x08($a0)
131	sw $t5, 0x0c($a0)
132	sw $t6, 0x10($a0)
133	sw $t7, 0x14($a0)
134	sw $t8, 0x18($a0)
135	sw $t9, 0x1c($a0)
136	lw $t2, 0x20($a1)
137	lw $t3, 0x24($a1)
138	lw $t4, 0x28($a1)
139	lw $t5, 0x2c($a1)
140	lw $t6, 0x30($a1)
141	lw $t7, 0x34($a1)
142	lw $t8, 0x38($a1)
143	lw $t9, 0x3c($a1)
144	sw $t2, 0x20($a0)
145	sw $t3, 0x24($a0)
146	sw $t4, 0x28($a0)
147	sw $t5, 0x2c($a0)
148	sw $t6, 0x30($a0)
149	sw $t7, 0x34($a0)
150	sw $t8, 0x38($a0)
151	sw $t9, 0x3c($a0)
152	addiu $a0, 0x40
153	bnez $t0, mc32_bloop
154	addiu $a1, 0x40
155
156	mc32_bloop_end:
157	andi $a2, $a2, 0x0f
158	beqz $a2, mc32_return
159
160	mc32_cloop:
161	lw $t1, 0($a1)
162	addiu $a2, -1
163	addiu $a1, 4
164	sw $t1, 0($a0)
165	bnez $a2, mc32_cloop
166	addiu $a0, 4
167
168	mc32_return:
169	jr $ra
170	nop
171