summaryrefslogtreecommitdiff
path: root/src/exa_sf_mask.g4a
blob: 5078d014fce859976d5bb573ceed181e2abc58bd (plain)
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
/*
 * Copyright © 2006 Intel Corporation
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Software"),
 * to deal in the Software without restriction, including without limitation
 * the rights to use, copy, modify, merge, publish, distribute, sublicense,
 * and/or sell copies of the Software, and to permit persons to whom the
 * Software is furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice (including the next
 * paragraph) shall be included in all copies or substantial portions of the
 * Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
 * IN THE SOFTWARE.
 *
 * Authors:
 *    Keith Packard <keithp@keithp.com>
 *    Eric Anholt <eric@anholt.net>
 *    Wang Zhenyu <zhenyu.z.wang@intel.com>
 */

/*
 * Inputs (note all sub-register addresses are bytes, not float indices)
 *
 * Note that the vertices will have been reordered:
 *
 * V0 is topmost (leftmost among topmost) (upper left)
 * V1 is next clockwise (lower right)
 * V2 is remaining (lower left)
 *
 *  V0 ...................... XX
 *  |                          .
 *  |                          .
 *  |                          .
 *  V2------------------------V1
 *
 *  G0	    thread state -- just pass along
 *
 *  G1 and G2 are fixed by SF spec
 *
 *  G1.0    reserved
 *  G1.4    Provoking vertex
 *  G1.8    Determinant
 *  G1.12   X1 - X0
 *  G1.16   X2 - X0
 *  G1.20   Y1 - Y0
 *  G1.24   Y2 - Y0
 *  G1.30   reserved
 *
 *  G2.0    Z0
 *  G2.4    1/W0
 *  G2.8    Z1
 *  G2.12   1/W1
 *  G2.16   Z2
 *  G2.20   1/W2
 *  G2.24   reserved
 *  G2.30   reserved
 *
 *  G3 is V0 Vertex Attribute Data from URB (upper left)
 *
 *  G3.0    u0
 *  G3.4    v0
 *
 *  G4 is V1 Vertex Attribute Data from URB (lower right)
 *
 *  G4.0    u1
 *  G4.4    v1
 *
 *  G5 is V2 Vertex Attribute Data from URB (lower left)
 *
 */

/* Compute inverses of the input deltas */
send (4) 0 g6<1>F g1.12<4,4,1>F math inv mlen 1 rlen 1 { align1 };

/* texture location at V0 */
mov (8) m3<1>F g3<8,8,1>F { align1 };

/* compute V1 - V2 (motion in X) for texture coordinates */
add (8) g7<1>F g4<8,8,1>F -g5<8,8,1>F { align1 };

/* multiply by 1/dx */
mul (8) m1<1>F g7<8,8,1>F g6.0<0,1,0>F { align1 };

/* Compute V2 - V0 (motion in Y) for texture coordinates */
add (8) g7<1>F g5<8,8,1>F -g3<8,8,1>F { align1 };

/* multiply by 1/dy */
mul (8) m2<1>F g7<8,8,1>F g6.8<0,1,0>F {align1 };

/* and we're done */
send (8) 0 null g0<8,8,1>F urb 0 transpose used complete mlen 4 rlen 0 { align1 EOT };
nop;
nop;
nop;
nop;
nop;
nop;
nop;
nop;