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
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
|
/*
Copyright (C) Intel Corp. 2006. All Rights Reserved.
Intel funded Tungsten Graphics to
develop this 3D driver.
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 COPYRIGHT OWNER(S) 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.
**********************************************************************/
/*
* Authors:
* Keith Whitwell <keithw@vmware.com>
*/
#include "main/glheader.h"
#include "main/macros.h"
#include "main/enums.h"
#include "main/transformfeedback.h"
#include "intel_batchbuffer.h"
#include "brw_defines.h"
#include "brw_context.h"
#include "brw_eu.h"
#include "brw_util.h"
#include "brw_state.h"
#include "brw_ff_gs.h"
#include "util/ralloc.h"
void
brw_compile_ff_gs_prog(struct brw_context *brw,
struct brw_ff_gs_prog_key *key)
{
struct brw_ff_gs_compile c;
const GLuint *program;
void *mem_ctx;
GLuint program_size;
memset(&c, 0, sizeof(c));
c.key = *key;
c.vue_map = brw->vs.prog_data->base.vue_map;
c.nr_regs = (c.vue_map.num_slots + 1)/2;
mem_ctx = ralloc_context(NULL);
/* Begin the compilation:
*/
brw_init_compile(brw->intelScreen->devinfo, &c.func, mem_ctx);
c.func.single_program_flow = 1;
/* For some reason the thread is spawned with only 4 channels
* unmasked.
*/
brw_set_default_mask_control(&c.func, BRW_MASK_DISABLE);
if (brw->gen >= 6) {
unsigned num_verts;
bool check_edge_flag;
/* On Sandybridge, we use the GS for implementing transform feedback
* (called "Stream Out" in the PRM).
*/
switch (key->primitive) {
case _3DPRIM_POINTLIST:
num_verts = 1;
check_edge_flag = false;
break;
case _3DPRIM_LINELIST:
case _3DPRIM_LINESTRIP:
case _3DPRIM_LINELOOP:
num_verts = 2;
check_edge_flag = false;
break;
case _3DPRIM_TRILIST:
case _3DPRIM_TRIFAN:
case _3DPRIM_TRISTRIP:
case _3DPRIM_RECTLIST:
num_verts = 3;
check_edge_flag = false;
break;
case _3DPRIM_QUADLIST:
case _3DPRIM_QUADSTRIP:
case _3DPRIM_POLYGON:
num_verts = 3;
check_edge_flag = true;
break;
default:
unreachable("Unexpected primitive type in Gen6 SOL program.");
}
gen6_sol_program(&c, key, num_verts, check_edge_flag);
} else {
/* On Gen4-5, we use the GS to decompose certain types of primitives.
* Note that primitives which don't require a GS program have already
* been weeded out by now.
*/
switch (key->primitive) {
case _3DPRIM_QUADLIST:
brw_ff_gs_quads( &c, key );
break;
case _3DPRIM_QUADSTRIP:
brw_ff_gs_quad_strip( &c, key );
break;
case _3DPRIM_LINELOOP:
brw_ff_gs_lines( &c );
break;
default:
ralloc_free(mem_ctx);
return;
}
}
brw_compact_instructions(&c.func, 0, 0, NULL);
/* get the program
*/
program = brw_get_program(&c.func, &program_size);
if (unlikely(INTEL_DEBUG & DEBUG_GS)) {
fprintf(stderr, "gs:\n");
brw_disassemble(brw->intelScreen->devinfo, c.func.store,
0, program_size, stderr);
fprintf(stderr, "\n");
}
brw_upload_cache(&brw->cache, BRW_CACHE_FF_GS_PROG,
&c.key, sizeof(c.key),
program, program_size,
&c.prog_data, sizeof(c.prog_data),
&brw->ff_gs.prog_offset, &brw->ff_gs.prog_data);
ralloc_free(mem_ctx);
}
static bool
brw_ff_gs_state_dirty(struct brw_context *brw)
{
return brw_state_dirty(brw,
_NEW_LIGHT,
BRW_NEW_PRIMITIVE |
BRW_NEW_TRANSFORM_FEEDBACK |
BRW_NEW_VS_PROG_DATA);
}
static void
brw_ff_gs_populate_key(struct brw_context *brw,
struct brw_ff_gs_prog_key *key)
{
static const unsigned swizzle_for_offset[4] = {
BRW_SWIZZLE4(0, 1, 2, 3),
BRW_SWIZZLE4(1, 2, 3, 3),
BRW_SWIZZLE4(2, 3, 3, 3),
BRW_SWIZZLE4(3, 3, 3, 3)
};
struct gl_context *ctx = &brw->ctx;
memset(key, 0, sizeof(*key));
/* BRW_NEW_VS_PROG_DATA (part of VUE map) */
key->attrs = brw->vs.prog_data->base.vue_map.slots_valid;
/* BRW_NEW_PRIMITIVE */
key->primitive = brw->primitive;
/* _NEW_LIGHT */
key->pv_first = (ctx->Light.ProvokingVertex == GL_FIRST_VERTEX_CONVENTION);
if (key->primitive == _3DPRIM_QUADLIST && ctx->Light.ShadeModel != GL_FLAT) {
/* Provide consistent primitive order with brw_set_prim's
* optimization of single quads to trifans.
*/
key->pv_first = true;
}
if (brw->gen >= 7) {
/* On Gen7 and later, we don't use GS (yet). */
key->need_gs_prog = false;
} else if (brw->gen == 6) {
/* On Gen6, GS is used for transform feedback. */
/* BRW_NEW_TRANSFORM_FEEDBACK */
if (_mesa_is_xfb_active_and_unpaused(ctx)) {
const struct gl_shader_program *shaderprog =
ctx->_Shader->CurrentProgram[MESA_SHADER_VERTEX];
const struct gl_transform_feedback_info *linked_xfb_info =
&shaderprog->LinkedTransformFeedback;
int i;
/* Make sure that the VUE slots won't overflow the unsigned chars in
* key->transform_feedback_bindings[].
*/
STATIC_ASSERT(BRW_VARYING_SLOT_COUNT <= 256);
/* Make sure that we don't need more binding table entries than we've
* set aside for use in transform feedback. (We shouldn't, since we
* set aside enough binding table entries to have one per component).
*/
assert(linked_xfb_info->NumOutputs <= BRW_MAX_SOL_BINDINGS);
key->need_gs_prog = true;
key->num_transform_feedback_bindings = linked_xfb_info->NumOutputs;
for (i = 0; i < key->num_transform_feedback_bindings; ++i) {
key->transform_feedback_bindings[i] =
linked_xfb_info->Outputs[i].OutputRegister;
key->transform_feedback_swizzles[i] =
swizzle_for_offset[linked_xfb_info->Outputs[i].ComponentOffset];
}
}
} else {
/* Pre-gen6, GS is used to transform QUADLIST, QUADSTRIP, and LINELOOP
* into simpler primitives.
*/
key->need_gs_prog = (brw->primitive == _3DPRIM_QUADLIST ||
brw->primitive == _3DPRIM_QUADSTRIP ||
brw->primitive == _3DPRIM_LINELOOP);
}
}
/* Calculate interpolants for triangle and line rasterization.
*/
void
brw_upload_ff_gs_prog(struct brw_context *brw)
{
struct brw_ff_gs_prog_key key;
if (!brw_ff_gs_state_dirty(brw))
return;
/* Populate the key:
*/
brw_ff_gs_populate_key(brw, &key);
if (brw->ff_gs.prog_active != key.need_gs_prog) {
brw->ctx.NewDriverState |= BRW_NEW_FF_GS_PROG_DATA;
brw->ff_gs.prog_active = key.need_gs_prog;
}
if (brw->ff_gs.prog_active) {
if (!brw_search_cache(&brw->cache, BRW_CACHE_FF_GS_PROG,
&key, sizeof(key),
&brw->ff_gs.prog_offset, &brw->ff_gs.prog_data)) {
brw_compile_ff_gs_prog(brw, &key);
}
}
}
void gen6_brw_upload_ff_gs_prog(struct brw_context *brw)
{
brw_upload_ff_gs_prog(brw);
}
|