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
|
/*
* Copyright © 2015 Broadcom
*
* 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.
*/
#include "vc4_qir.h"
#include "tgsi/tgsi_info.h"
#include "glsl/nir/nir_builder.h"
/**
* Walks the NIR generated by TGSI-to-NIR to lower its io intrinsics into
* something amenable to the VC4 architecture.
*
* Currently, it split outputs into scalars, and drops any non-position values
* in coordinate shaders.
*/
static void
vc4_nir_lower_output(struct vc4_compile *c, nir_builder *b,
nir_intrinsic_instr *intr)
{
nir_variable *output_var = NULL;
foreach_list_typed(nir_variable, var, node, &c->s->outputs) {
if (var->data.driver_location == intr->const_index[0]) {
output_var = var;
break;
}
}
assert(output_var);
unsigned semantic_name = output_var->data.location;
if (c->stage == QSTAGE_COORD &&
(semantic_name != TGSI_SEMANTIC_POSITION &&
semantic_name != TGSI_SEMANTIC_PSIZE)) {
nir_instr_remove(&intr->instr);
return;
}
/* All TGSI-to-NIR outputs are VEC4. */
assert(intr->num_components == 4);
nir_builder_insert_before_instr(b, &intr->instr);
for (unsigned i = 0; i < intr->num_components; i++) {
nir_intrinsic_instr *intr_comp =
nir_intrinsic_instr_create(c->s, nir_intrinsic_store_output);
intr_comp->num_components = 1;
intr_comp->const_index[0] = intr->const_index[0] * 4 + i;
assert(intr->src[0].is_ssa);
intr_comp->src[0] = nir_src_for_ssa(nir_swizzle(b,
intr->src[0].ssa,
&i, 1, false));
nir_builder_instr_insert(b, &intr_comp->instr);
}
nir_instr_remove(&intr->instr);
}
static void
vc4_nir_lower_io_instr(struct vc4_compile *c, nir_builder *b,
struct nir_instr *instr)
{
if (instr->type != nir_instr_type_intrinsic)
return;
nir_intrinsic_instr *intr = nir_instr_as_intrinsic(instr);
switch (intr->intrinsic) {
case nir_intrinsic_store_output:
vc4_nir_lower_output(c, b, intr);
break;
default:
break;
}
}
static bool
vc4_nir_lower_io_block(nir_block *block, void *arg)
{
struct vc4_compile *c = arg;
nir_function_impl *impl =
nir_cf_node_get_function(&block->cf_node);
nir_builder b;
nir_builder_init(&b, impl);
nir_foreach_instr_safe(block, instr)
vc4_nir_lower_io_instr(c, &b, instr);
return true;
}
static bool
vc4_nir_lower_io_impl(struct vc4_compile *c, nir_function_impl *impl)
{
nir_foreach_block(impl, vc4_nir_lower_io_block, c);
nir_metadata_preserve(impl, nir_metadata_block_index |
nir_metadata_dominance);
return true;
}
void
vc4_nir_lower_io(struct vc4_compile *c)
{
nir_foreach_overload(c->s, overload) {
if (overload->impl)
vc4_nir_lower_io_impl(c, overload->impl);
}
}
|