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
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
|
/*
* Copyright © 2018 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.
*
*/
#include "main/mtypes.h"
#include "glsl_types.h"
#include "linker_util.h"
#include "util/bitscan.h"
#include "util/set.h"
#include "ir_uniform.h" /* for gl_uniform_storage */
/* Utility methods shared between the GLSL IR and the NIR */
/* From the OpenGL 4.6 specification, 7.3.1.1 Naming Active Resources:
*
* "For an active shader storage block member declared as an array of an
* aggregate type, an entry will be generated only for the first array
* element, regardless of its type. Such block members are referred to as
* top-level arrays. If the block member is an aggregate type, the
* enumeration rules are then applied recursively."
*/
bool
link_util_should_add_buffer_variable(struct gl_shader_program *prog,
struct gl_uniform_storage *uniform,
int top_level_array_base_offset,
int top_level_array_size_in_bytes,
int second_element_offset,
int block_index)
{
/* If the uniform is not a shader storage buffer or is not an array return
* true.
*/
if (!uniform->is_shader_storage || top_level_array_size_in_bytes == 0)
return true;
int after_top_level_array = top_level_array_base_offset +
top_level_array_size_in_bytes;
/* Check for a new block, or that we are not dealing with array elements of
* a top member array other than the first element.
*/
if (block_index != uniform->block_index ||
uniform->offset >= after_top_level_array ||
uniform->offset < second_element_offset) {
return true;
}
return false;
}
bool
link_util_add_program_resource(struct gl_shader_program *prog,
struct set *resource_set,
GLenum type, const void *data, uint8_t stages)
{
assert(data);
/* If resource already exists, do not add it again. */
if (_mesa_set_search(resource_set, data))
return true;
prog->data->ProgramResourceList =
reralloc(prog->data,
prog->data->ProgramResourceList,
gl_program_resource,
prog->data->NumProgramResourceList + 1);
if (!prog->data->ProgramResourceList) {
linker_error(prog, "Out of memory during linking.\n");
return false;
}
struct gl_program_resource *res =
&prog->data->ProgramResourceList[prog->data->NumProgramResourceList];
res->Type = type;
res->Data = data;
res->StageReferences = stages;
prog->data->NumProgramResourceList++;
_mesa_set_add(resource_set, data);
return true;
}
/**
* Search through the list of empty blocks to find one that fits the current
* uniform.
*/
int
link_util_find_empty_block(struct gl_shader_program *prog,
struct gl_uniform_storage *uniform)
{
const unsigned entries = MAX2(1, uniform->array_elements);
foreach_list_typed(struct empty_uniform_block, block, link,
&prog->EmptyUniformLocations) {
/* Found a block with enough slots to fit the uniform */
if (block->slots == entries) {
unsigned start = block->start;
exec_node_remove(&block->link);
ralloc_free(block);
return start;
/* Found a block with more slots than needed. It can still be used. */
} else if (block->slots > entries) {
unsigned start = block->start;
block->start += entries;
block->slots -= entries;
return start;
}
}
return -1;
}
void
link_util_update_empty_uniform_locations(struct gl_shader_program *prog)
{
struct empty_uniform_block *current_block = NULL;
for (unsigned i = 0; i < prog->NumUniformRemapTable; i++) {
/* We found empty space in UniformRemapTable. */
if (prog->UniformRemapTable[i] == NULL) {
/* We've found the beginning of a new continous block of empty slots */
if (!current_block || current_block->start + current_block->slots != i) {
current_block = rzalloc(prog, struct empty_uniform_block);
current_block->start = i;
exec_list_push_tail(&prog->EmptyUniformLocations,
¤t_block->link);
}
/* The current block continues, so we simply increment its slots */
current_block->slots++;
}
}
}
void
link_util_check_subroutine_resources(struct gl_shader_program *prog)
{
unsigned mask = prog->data->linked_stages;
while (mask) {
const int i = u_bit_scan(&mask);
struct gl_program *p = prog->_LinkedShaders[i]->Program;
if (p->sh.NumSubroutineUniformRemapTable > MAX_SUBROUTINE_UNIFORM_LOCATIONS) {
linker_error(prog, "Too many %s shader subroutine uniforms\n",
_mesa_shader_stage_to_string(i));
}
}
}
/**
* Validate uniform resources used by a program versus the implementation limits
*/
void
link_util_check_uniform_resources(struct gl_context *ctx,
struct gl_shader_program *prog)
{
unsigned total_uniform_blocks = 0;
unsigned total_shader_storage_blocks = 0;
for (unsigned i = 0; i < MESA_SHADER_STAGES; i++) {
struct gl_linked_shader *sh = prog->_LinkedShaders[i];
if (sh == NULL)
continue;
if (sh->num_uniform_components >
ctx->Const.Program[i].MaxUniformComponents) {
if (ctx->Const.GLSLSkipStrictMaxUniformLimitCheck) {
linker_warning(prog, "Too many %s shader default uniform block "
"components, but the driver will try to optimize "
"them out; this is non-portable out-of-spec "
"behavior\n",
_mesa_shader_stage_to_string(i));
} else {
linker_error(prog, "Too many %s shader default uniform block "
"components\n",
_mesa_shader_stage_to_string(i));
}
}
if (sh->num_combined_uniform_components >
ctx->Const.Program[i].MaxCombinedUniformComponents) {
if (ctx->Const.GLSLSkipStrictMaxUniformLimitCheck) {
linker_warning(prog, "Too many %s shader uniform components, "
"but the driver will try to optimize them out; "
"this is non-portable out-of-spec behavior\n",
_mesa_shader_stage_to_string(i));
} else {
linker_error(prog, "Too many %s shader uniform components\n",
_mesa_shader_stage_to_string(i));
}
}
total_shader_storage_blocks += sh->Program->info.num_ssbos;
total_uniform_blocks += sh->Program->info.num_ubos;
}
if (total_uniform_blocks > ctx->Const.MaxCombinedUniformBlocks) {
linker_error(prog, "Too many combined uniform blocks (%d/%d)\n",
total_uniform_blocks, ctx->Const.MaxCombinedUniformBlocks);
}
if (total_shader_storage_blocks > ctx->Const.MaxCombinedShaderStorageBlocks) {
linker_error(prog, "Too many combined shader storage blocks (%d/%d)\n",
total_shader_storage_blocks,
ctx->Const.MaxCombinedShaderStorageBlocks);
}
for (unsigned i = 0; i < prog->data->NumUniformBlocks; i++) {
if (prog->data->UniformBlocks[i].UniformBufferSize >
ctx->Const.MaxUniformBlockSize) {
linker_error(prog, "Uniform block %s too big (%d/%d)\n",
prog->data->UniformBlocks[i].Name,
prog->data->UniformBlocks[i].UniformBufferSize,
ctx->Const.MaxUniformBlockSize);
}
}
for (unsigned i = 0; i < prog->data->NumShaderStorageBlocks; i++) {
if (prog->data->ShaderStorageBlocks[i].UniformBufferSize >
ctx->Const.MaxShaderStorageBlockSize) {
linker_error(prog, "Shader storage block %s too big (%d/%d)\n",
prog->data->ShaderStorageBlocks[i].Name,
prog->data->ShaderStorageBlocks[i].UniformBufferSize,
ctx->Const.MaxShaderStorageBlockSize);
}
}
}
void
link_util_calculate_subroutine_compat(struct gl_shader_program *prog)
{
unsigned mask = prog->data->linked_stages;
while (mask) {
const int i = u_bit_scan(&mask);
struct gl_program *p = prog->_LinkedShaders[i]->Program;
for (unsigned j = 0; j < p->sh.NumSubroutineUniformRemapTable; j++) {
if (p->sh.SubroutineUniformRemapTable[j] == INACTIVE_UNIFORM_EXPLICIT_LOCATION)
continue;
struct gl_uniform_storage *uni = p->sh.SubroutineUniformRemapTable[j];
if (!uni)
continue;
int count = 0;
if (p->sh.NumSubroutineFunctions == 0) {
linker_error(prog, "subroutine uniform %s defined but no valid functions found\n", uni->type->name);
continue;
}
for (unsigned f = 0; f < p->sh.NumSubroutineFunctions; f++) {
struct gl_subroutine_function *fn = &p->sh.SubroutineFunctions[f];
for (int k = 0; k < fn->num_compat_types; k++) {
if (fn->types[k] == uni->type) {
count++;
break;
}
}
}
uni->num_compatible_subroutines = count;
}
}
}
|