1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
|
#include "global.h"
#include "main.h"
#include "fx.h"
void MI_Copy48B(void *src, void *dst);
ARM_FUNC void MTX_ScaleApply43(struct Mtx43 *mtx, struct Mtx43 *dst, fx32 x, fx32 y, fx32 z){
//this works because matrices are indexed columns first
MTX_ScaleApply33((struct Mtx33 *)mtx, (struct Mtx33 *)dst, x, y, z);
dst->_[9] = mtx->_[9];
dst->_[10] = mtx->_[10];
dst->_[11] = mtx->_[11];
}
ARM_FUNC fx32 MTX_Inverse43(struct Mtx43 *mtx, struct Mtx43 *inv){
struct Mtx43 tempmat;
struct Mtx43 *dst;
fx32 det0, det1, det2, det;
fx32 var0, var1, var2, var3;
if (mtx == inv)
dst = &tempmat;
else
dst = inv;
//subdeterminants
det0 = (fx32)(((fx64)mtx->_[4] * mtx->_[8] - (fx64)mtx->_[5] * mtx->_[7] + (fx64)(1 << (FX32_INT_SHIFT - 1))) >> FX32_INT_SHIFT);
det1 = (fx32)(((fx64)mtx->_[3] * mtx->_[8] - (fx64)mtx->_[5] * mtx->_[6] + (fx64)(1 << (FX32_INT_SHIFT - 1))) >> FX32_INT_SHIFT);
det2 = (fx32)(((fx64)mtx->_[3] * mtx->_[7] - (fx64)mtx->_[4] * mtx->_[6] + (fx64)(1 << (FX32_INT_SHIFT - 1))) >> FX32_INT_SHIFT);
//matrix determinant
det = (fx32)(((fx64)mtx->_[0] * det0 - (fx64)mtx->_[1] * det1 + (fx64)mtx->_[2] * det2 + (fx64)(1 << (FX32_INT_SHIFT - 1))) >> FX32_INT_SHIFT);
if (det == 0)
return -1; //not invertible
FX_InvAsync(det);
var0 = (fx32)(((fx64)mtx->_[1] * mtx->_[8] - (fx64)mtx->_[7] * mtx->_[2]) >> FX32_INT_SHIFT);
var1 = (fx32)(((fx64)mtx->_[1] * mtx->_[5] - (fx64)mtx->_[4] * mtx->_[2]) >> FX32_INT_SHIFT);
var2 = (fx32)(((fx64)mtx->_[0] * mtx->_[8] - (fx64)mtx->_[6] * mtx->_[2]) >> FX32_INT_SHIFT);
var3 = (fx32)(((fx64)mtx->_[0] * mtx->_[5] - (fx64)mtx->_[3] * mtx->_[2]) >> FX32_INT_SHIFT);
fx32 ret = FX_GetDivResult();
dst->_[0] = (fx32)(((fx64)ret * det0) >> FX32_INT_SHIFT);
dst->_[1] = -(fx32)(((fx64)ret * var0) >> FX32_INT_SHIFT);
dst->_[2] = (fx32)(((fx64)ret * var1) >> FX32_INT_SHIFT);
dst->_[3] = -(fx32)(((fx64)ret * det1) >> FX32_INT_SHIFT);
dst->_[4] = (fx32)(((fx64)ret * var2) >> FX32_INT_SHIFT);
dst->_[5] = -(fx32)(((fx64)ret * var3) >> FX32_INT_SHIFT);
dst->_[6] = (fx32)(((fx64)ret * det2) >> FX32_INT_SHIFT);
fx32 temp = (fx32)(((fx64)mtx->_[0] * mtx->_[7] - (fx64)mtx->_[6] * mtx->_[1]) >> FX32_INT_SHIFT);
dst->_[7] = -(fx32)(((fx64)ret * temp) >> FX32_INT_SHIFT);
fx32 temp1 = (fx32)(((fx64)mtx->_[0] * mtx->_[4] - (fx64)mtx->_[3] * mtx->_[1]) >> FX32_INT_SHIFT);
dst->_[8] = (fx32)(((fx64)ret * temp1) >> FX32_INT_SHIFT);
dst->_[9] = -(fx32)(((fx64)dst->_[0] * mtx->_[9] + (fx64)dst->_[3] * mtx->_[10] + (fx64)dst->_[6] * mtx->_[11]) >> FX32_INT_SHIFT);
dst->_[10] = -(fx32)(((fx64)dst->_[1] * mtx->_[9] + (fx64)dst->_[4] * mtx->_[10] + (fx64)dst->_[7] * mtx->_[11]) >> FX32_INT_SHIFT);
dst->_[11] = -(fx32)(((fx64)dst->_[2] * mtx->_[9] + (fx64)dst->_[5] * mtx->_[10] + (fx64)dst->_[8] * mtx->_[11]) >> FX32_INT_SHIFT);
if (dst == &tempmat)
MI_Copy48B(&tempmat, inv);
return 0;
}
ARM_FUNC void MTX_Concat43(struct Mtx43 *a, struct Mtx43 *b, struct Mtx43 *c){
struct Mtx43 temp;
struct Mtx43 *dst;
fx32 a0, a1, a2;
fx32 b0, b1, b2;
if (c == b)
dst = &temp;
else
dst = c;
a0 = a->_[0];
a1 = a->_[1];
a2 = a->_[2];
dst->_[0] = (fx32)(((fx64)a0 * b->_[0] + (fx64)a1 * b->_[3] + (fx64)a2 * b->_[6] ) >> FX32_INT_SHIFT);
dst->_[1] = (fx32)(((fx64)a0 * b->_[1] + (fx64)a1 * b->_[4] + (fx64)a2 * b->_[7] ) >> FX32_INT_SHIFT);
b0 = b->_[2];
b1 = b->_[5];
b2 = b->_[8];
dst->_[2] = (fx32)(((fx64)a0 * b0 + (fx64)a1 * b1 + (fx64)a2 * b2) >> FX32_INT_SHIFT);
a0 = a->_[3];
a1 = a->_[4];
a2 = a->_[5];
dst->_[5] = (fx32)(((fx64)a0 * b0 + (fx64)a1 * b1 + (fx64)a2 * b2) >> FX32_INT_SHIFT);
dst->_[4] = (fx32)(((fx64)a0 * b->_[1] + (fx64)a1 * b->_[4] + (fx64)a2 * b->_[7] ) >> FX32_INT_SHIFT);
b0 = b->_[0];
b1 = b->_[3];
b2 = b->_[6];
dst->_[3] = (fx32)(((fx64)a0 * b0 + (fx64)a1 * b1 + (fx64)a2 * b2) >> FX32_INT_SHIFT);
a0 = a->_[6];
a1 = a->_[7];
a2 = a->_[8];
dst->_[6] = (fx32)(((fx64)a0 * b0 + (fx64)a1 * b1 + (fx64)a2 * b2) >> FX32_INT_SHIFT);
dst->_[7] = (fx32)(((fx64)a0 * b->_[1] + (fx64)a1 * b->_[4] + (fx64)a2 * b->_[7] ) >> FX32_INT_SHIFT);
b0 = b->_[2];
b1 = b->_[5];
b2 = b->_[8];
dst->_[8] = (fx32)(((fx64)a0 * b0 + (fx64)a1 * b1 + (fx64)a2 * b2) >> FX32_INT_SHIFT);
a0 = a->_[9];
a1 = a->_[10];
a2 = a->_[11];
dst->_[11] = (fx32)((((fx64)a0 * b0 + (fx64)a1 * b1 + (fx64)a2 * b2) >> FX32_INT_SHIFT) + b->_[11]);
dst->_[10] = (fx32)((((fx64)a0 * b->_[1] + (fx64)a1 * b->_[4] + (fx64)a2 * b->_[7]) >> FX32_INT_SHIFT) + b->_[10]);
dst->_[9] = (fx32)((((fx64)a0 * b->_[0] + (fx64)a1 * b->_[3] + (fx64)a2 * b->_[6]) >> FX32_INT_SHIFT) + b->_[9]);
if (dst == &temp)
*c = temp;
}
ARM_FUNC void MTX_MultVec43(struct Vecx32 *vec, struct Mtx43 *mtx, struct Vecx32 *dst){
fx32 x, y, z;
x = vec->x;
y = vec->y;
z = vec->z;
dst->x = (fx32)(((fx64)x * mtx->_[0] + (fx64)y * mtx->_[3] + (fx64)z * mtx->_[6]) >> FX32_INT_SHIFT);
dst->x += mtx->_[9];
dst->y = (fx32)(((fx64)x * mtx->_[1] + (fx64)y * mtx->_[4] + (fx64)z * mtx->_[7]) >> FX32_INT_SHIFT);
dst->y += mtx->_[10];
dst->z = (fx32)(((fx64)x * mtx->_[2] + (fx64)y * mtx->_[5] + (fx64)z * mtx->_[8]) >> FX32_INT_SHIFT);
dst->z += mtx->_[11];
}
ARM_FUNC asm void MTX_Identity43_(struct Mtx43 *mtx){
mov r2, #0x1000
mov r3, #0x0
stmia r0!, {r2-r3}
mov r1, #0x0
stmia r0!, {r1,r3}
stmia r0!, {r2-r3}
stmia r0!, {r1,r3}
stmia r0!, {r2-r3}
stmia r0!, {r1,r3}
bx lr
}
ARM_FUNC asm void MTX_Copy43To44_(struct Mtx43 *src, struct Mtx44 *dst){
stmdb sp!, {r4}
mov r12, #0x0
ldmia r0!, {r2-r4}
stmia r1!, {r2-r4,r12}
ldmia r0!, {r2-r4}
stmia r1!, {r2-r4,r12}
ldmia r0!, {r2-r4}
stmia r1!, {r2-r4,r12}
mov r12, #0x1000
ldmia r0!, {r2-r4}
stmia r1!, {r2-r4,r12}
ldmia sp!, {r4}
bx lr
}
THUMB_FUNC asm void MTX_Scale43_(struct Mtx43 *dst, fx32 x, fx32 y, fx32 z){
stmia r0!, {r1}
mov r1, #0x0
str r3, [r0, #0x1c]
mov r3, #0x0
stmia r0!, {r1,r3}
stmia r0!, {r1-r3}
mov r2, #0x0
stmia r0!, {r1,r3}
add r0, #0x4
stmia r0!, {r1-r3}
bx lr
}
THUMB_FUNC asm void MTX_RotX43_(struct Mtx43 *mtx, fx32 sinphi, fx32 cosphi){
str r1, [r0, #0x14]
neg r1, r1
str r1, [r0, #0x1c]
mov r1, #0x1
lsl r1, r1, #0xc
stmia r0!, {r1}
mov r3, #0x0
mov r1, #0x0
stmia r0!, {r1,r3}
stmia r0!, {r1-r2}
str r1, [r0, #0x4]
add r0, #0xc
stmia r0!, {r2-r3}
stmia r0!, {r1,r3}
bx lr
}
THUMB_FUNC asm void MTX_RotY43_(struct Mtx43 *mtx, fx32 sinphi, fx32 cosphi){
str r1, [r0, #0x18]
mov r3, #0x0
stmia r0!, {r2-r3}
neg r1, r1
stmia r0!, {r1,r3}
mov r1, #0x1
lsl r1, r1, #0xc
stmia r0!, {r1,r3}
add r0, #0x4
mov r1, #0x0
stmia r0!, {r1-r3}
stmia r0!, {r1,r3}
bx lr
}
|
