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
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
|
/**************************************************************************
*
* Copyright 2007-2009 VMware, Inc.
* All Rights Reserved.
*
* 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, sub license, 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 NON-INFRINGEMENT.
* IN NO EVENT SHALL VMWARE AND/OR ITS SUPPLIERS 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.
*
**************************************************************************/
/*
* Rasterization for binned triangles within a tile
*/
#include "util/u_math.h"
#include "lp_debug.h"
#include "lp_rast_priv.h"
#include "lp_tile_soa.h"
/**
* Add a 4x4 block of pixels to the block list.
* All pixels are known to be inside the triangle's bounds.
*/
static void
block_full_4( struct lp_rasterizer *rast, int x, int y )
{
const unsigned i = rast->nr_blocks;
assert(x % 4 == 0);
assert(y % 4 == 0);
rast->blocks[i].x = x;
rast->blocks[i].y = y;
rast->blocks[i].mask = ~0;
rast->nr_blocks++;
}
/**
* Add a 16x16 block of pixels to the block list.
* All pixels are known to be inside the triangle's bounds.
*/
static void
block_full_16( struct lp_rasterizer *rast, int x, int y )
{
unsigned ix, iy;
assert(x % 16 == 0);
assert(y % 16 == 0);
for (iy = 0; iy < 16; iy += 4)
for (ix = 0; ix < 16; ix += 4)
block_full_4(rast, x + ix, y + iy);
}
/**
* Evaluate each pixel in a 4x4 block to determine if it lies within
* the triangle's bounds.
* Generate a mask of in/out flags and add the block to the blocks list.
*/
static void
do_block_4( struct lp_rasterizer *rast,
const struct lp_rast_triangle *tri,
int x, int y,
int c1,
int c2,
int c3 )
{
int i;
unsigned mask = 0;
assert(x % 4 == 0);
assert(y % 4 == 0);
for (i = 0; i < 16; i++) {
int any_negative = ((c1 + tri->step[0][i]) |
(c2 + tri->step[1][i]) |
(c3 + tri->step[2][i])) >> 31;
mask |= (~any_negative) & (1 << i);
}
/* As we do trivial reject already, masks should rarely be all zero:
*/
if (mask) {
const unsigned i = rast->nr_blocks;
rast->blocks[i].x = x;
rast->blocks[i].y = y;
rast->blocks[i].mask = mask;
rast->nr_blocks++;
}
}
/**
* Evaluate a 16x16 block of pixels to determine which 4x4 subblocks are in/out
* of the triangle's bounds.
*/
static void
do_block_16( struct lp_rasterizer *rast,
const struct lp_rast_triangle *tri,
int x, int y,
int c1,
int c2,
int c3 )
{
int ix, iy, i = 0;
int ei1 = tri->ei1 * 4;
int ei2 = tri->ei2 * 4;
int ei3 = tri->ei3 * 4;
int eo1 = tri->eo1 * 4;
int eo2 = tri->eo2 * 4;
int eo3 = tri->eo3 * 4;
assert(x % 16 == 0);
assert(y % 16 == 0);
for (iy = 0; iy < 16; iy+=4) {
for (ix = 0; ix < 16; ix+=4, i++) {
int cx1 = c1 + (tri->step[0][i] * 4);
int cx2 = c2 + (tri->step[1][i] * 4);
int cx3 = c3 + (tri->step[2][i] * 4);
if (cx1 + eo1 < 0 ||
cx2 + eo2 < 0 ||
cx3 + eo3 < 0) {
/* the block is completely outside the triangle - nop */
}
else if (cx1 + ei1 > 0 &&
cx2 + ei2 > 0 &&
cx3 + ei3 > 0) {
/* the block is completely inside the triangle */
block_full_4(rast, x+ix, y+iy);
}
else {
/* the block is partially in/out of the triangle */
do_block_4(rast, tri, x+ix, y+iy, cx1, cx2, cx3);
}
}
}
}
/**
* Scan the tile in chunks and figure out which pixels to rasterize
* for this triangle.
*/
void
lp_rast_triangle( struct lp_rasterizer *rast,
const union lp_rast_cmd_arg arg )
{
const struct lp_rast_triangle *tri = arg.triangle;
int x = rast->x;
int y = rast->y;
int ix, iy;
unsigned i = 0;
int c1 = tri->c1 + tri->dx12 * y - tri->dy12 * x;
int c2 = tri->c2 + tri->dx23 * y - tri->dy23 * x;
int c3 = tri->c3 + tri->dx31 * y - tri->dy31 * x;
int ei1 = tri->ei1 * 16;
int ei2 = tri->ei2 * 16;
int ei3 = tri->ei3 * 16;
int eo1 = tri->eo1 * 16;
int eo2 = tri->eo2 * 16;
int eo3 = tri->eo3 * 16;
assert(Elements(rast->blocks) == (TILE_SIZE * TILE_SIZE) / (4*4));
LP_DBG(DEBUG_RAST, "lp_rast_triangle\n");
rast->nr_blocks = 0;
/* Walk over the tile to build a list of 4x4 pixel blocks which will
* be filled/shaded. We do this at two granularities: 16x16 blocks
* and then 4x4 blocks.
*/
for (iy = 0; iy < TILE_SIZE; iy += 16) {
for (ix = 0; ix < TILE_SIZE; ix += 16, i++) {
int cx1 = c1 + (tri->step[0][i] * 16);
int cx2 = c2 + (tri->step[1][i] * 16);
int cx3 = c3 + (tri->step[2][i] * 16);
if (cx1 + eo1 < 0 ||
cx2 + eo2 < 0 ||
cx3 + eo3 < 0) {
/* the block is completely outside the triangle - nop */
}
else if (cx1 + ei1 > 0 &&
cx2 + ei2 > 0 &&
cx3 + ei3 > 0) {
/* the block is completely inside the triangle */
block_full_16(rast, x+ix, y+iy);
}
else {
/* the block is partially in/out of the triangle */
do_block_16(rast, tri, x+ix, y+iy, cx1, cx2, cx3);
}
}
}
assert(rast->nr_blocks <= Elements(rast->blocks));
/* Shade the 4x4 pixel blocks */
for (i = 0; i < rast->nr_blocks; i++)
lp_rast_shade_quads(rast, &tri->inputs,
rast->blocks[i].x,
rast->blocks[i].y,
rast->blocks[i].mask);
}
|
