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
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
|
/*
* Copyright 2010 Jerome Glisse <glisse@freedesktop.org>
*
* 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
* on 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
* THE AUTHOR(S) AND/OR THEIR 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:
* Jerome Glisse
*/
#ifndef R600_H
#define R600_H
#include <assert.h>
#include <stdint.h>
#include <stdio.h>
#include <util/u_double_list.h>
#include <pipe/p_compiler.h>
#define RADEON_CTX_MAX_PM4 (64 * 1024 / 4)
#define R600_ERR(fmt, args...) \
fprintf(stderr, "EE %s:%d %s - "fmt, __FILE__, __LINE__, __func__, ##args)
typedef uint64_t u64;
typedef uint32_t u32;
typedef uint16_t u16;
typedef uint8_t u8;
struct radeon;
struct winsys_handle;
enum radeon_family {
CHIP_UNKNOWN,
CHIP_R100,
CHIP_RV100,
CHIP_RS100,
CHIP_RV200,
CHIP_RS200,
CHIP_R200,
CHIP_RV250,
CHIP_RS300,
CHIP_RV280,
CHIP_R300,
CHIP_R350,
CHIP_RV350,
CHIP_RV380,
CHIP_R420,
CHIP_R423,
CHIP_RV410,
CHIP_RS400,
CHIP_RS480,
CHIP_RS600,
CHIP_RS690,
CHIP_RS740,
CHIP_RV515,
CHIP_R520,
CHIP_RV530,
CHIP_RV560,
CHIP_RV570,
CHIP_R580,
CHIP_R600,
CHIP_RV610,
CHIP_RV630,
CHIP_RV670,
CHIP_RV620,
CHIP_RV635,
CHIP_RS780,
CHIP_RS880,
CHIP_RV770,
CHIP_RV730,
CHIP_RV710,
CHIP_RV740,
CHIP_CEDAR,
CHIP_REDWOOD,
CHIP_JUNIPER,
CHIP_CYPRESS,
CHIP_HEMLOCK,
CHIP_PALM,
CHIP_SUMO,
CHIP_SUMO2,
CHIP_BARTS,
CHIP_TURKS,
CHIP_CAICOS,
CHIP_CAYMAN,
CHIP_LAST,
};
enum chip_class {
R600,
R700,
EVERGREEN,
CAYMAN,
};
struct r600_tiling_info {
unsigned num_channels;
unsigned num_banks;
unsigned group_bytes;
};
enum radeon_family r600_get_family(struct radeon *rw);
enum chip_class r600_get_family_class(struct radeon *radeon);
struct r600_tiling_info *r600_get_tiling_info(struct radeon *radeon);
unsigned r600_get_clock_crystal_freq(struct radeon *radeon);
unsigned r600_get_minor_version(struct radeon *radeon);
unsigned r600_get_num_backends(struct radeon *radeon);
/* r600_bo.c */
struct r600_bo;
struct r600_bo *r600_bo(struct radeon *radeon,
unsigned size, unsigned alignment,
unsigned binding, unsigned usage);
struct r600_bo *r600_bo_handle(struct radeon *radeon,
unsigned handle, unsigned *array_mode);
void *r600_bo_map(struct radeon *radeon, struct r600_bo *bo, unsigned usage, void *ctx);
void r600_bo_unmap(struct radeon *radeon, struct r600_bo *bo);
void r600_bo_reference(struct radeon *radeon, struct r600_bo **dst,
struct r600_bo *src);
boolean r600_bo_get_winsys_handle(struct radeon *radeon, struct r600_bo *pb_bo,
unsigned stride, struct winsys_handle *whandle);
static INLINE unsigned r600_bo_offset(struct r600_bo *bo)
{
return 0;
}
/* R600/R700 STATES */
#define R600_GROUP_MAX 16
#define R600_BLOCK_MAX_BO 32
#define R600_BLOCK_MAX_REG 128
/* each range covers 9 bits of dword space = 512 dwords = 2k bytes */
/* there is a block entry for each register so 512 blocks */
/* we have no registers to read/write below 0x8000 (0x2000 in dw space) */
/* we use some fake offsets at 0x40000 to do evergreen sampler borders so take 0x42000 as a max bound*/
#define RANGE_OFFSET_START 0x8000
#define HASH_SHIFT 9
#define NUM_RANGES (0x42000 - RANGE_OFFSET_START) / (4 << HASH_SHIFT) /* 128 << 9 = 64k */
#define CTX_RANGE_ID(offset) ((((offset - RANGE_OFFSET_START) >> 2) >> HASH_SHIFT) & 255)
#define CTX_BLOCK_ID(offset) (((offset - RANGE_OFFSET_START) >> 2) & ((1 << HASH_SHIFT) - 1))
struct r600_pipe_reg {
u32 value;
u32 mask;
struct r600_block *block;
struct r600_bo *bo;
u32 id;
};
struct r600_pipe_state {
unsigned id;
unsigned nregs;
struct r600_pipe_reg regs[R600_BLOCK_MAX_REG];
};
#define R600_BLOCK_STATUS_ENABLED (1 << 0)
#define R600_BLOCK_STATUS_DIRTY (1 << 1)
struct r600_block_reloc {
struct r600_bo *bo;
unsigned flush_flags;
unsigned flush_mask;
unsigned bo_pm4_index;
};
struct r600_block {
struct list_head list;
unsigned status;
unsigned flags;
unsigned start_offset;
unsigned pm4_ndwords;
unsigned pm4_flush_ndwords;
unsigned nbo;
u16 nreg;
u16 nreg_dirty;
u32 *reg;
u32 pm4[R600_BLOCK_MAX_REG];
unsigned pm4_bo_index[R600_BLOCK_MAX_REG];
struct r600_block_reloc reloc[R600_BLOCK_MAX_BO];
};
struct r600_range {
struct r600_block **blocks;
};
/*
* relocation
*/
#pragma pack(1)
struct r600_reloc {
uint32_t handle;
uint32_t read_domain;
uint32_t write_domain;
uint32_t flags;
};
#pragma pack()
/*
* query
*/
struct r600_query {
u64 result;
/* The kind of query. Currently only OQ is supported. */
unsigned type;
/* How many results have been written, in dwords. It's incremented
* after end_query and flush. */
unsigned num_results;
/* if we've flushed the query */
unsigned state;
/* The buffer where query results are stored. */
struct r600_bo *buffer;
unsigned buffer_size;
/* linked list of queries */
struct list_head list;
};
#define R600_QUERY_STATE_STARTED (1 << 0)
#define R600_QUERY_STATE_ENDED (1 << 1)
#define R600_QUERY_STATE_SUSPENDED (1 << 2)
#define R600_CONTEXT_DRAW_PENDING (1 << 0)
#define R600_CONTEXT_DST_CACHES_DIRTY (1 << 1)
#define R600_CONTEXT_CHECK_EVENT_FLUSH (1 << 2)
struct r600_context {
struct radeon *radeon;
struct r600_range *range;
unsigned nblocks;
struct r600_block **blocks;
struct list_head dirty;
unsigned pm4_ndwords;
unsigned pm4_cdwords;
unsigned pm4_dirty_cdwords;
unsigned ctx_pm4_ndwords;
unsigned nreloc;
unsigned creloc;
struct r600_reloc *reloc;
struct radeon_bo **bo;
u32 *pm4;
struct list_head query_list;
unsigned num_query_running;
struct list_head fenced_bo;
unsigned max_db; /* for OQ */
unsigned num_dest_buffers;
unsigned flags;
boolean predicate_drawing;
};
struct r600_draw {
u32 vgt_num_indices;
u32 vgt_num_instances;
u32 vgt_index_type;
u32 vgt_draw_initiator;
u32 indices_bo_offset;
struct r600_bo *indices;
};
int r600_context_init(struct r600_context *ctx, struct radeon *radeon);
void r600_context_fini(struct r600_context *ctx);
void r600_context_pipe_state_set(struct r600_context *ctx, struct r600_pipe_state *state);
void r600_context_pipe_state_set_ps_resource(struct r600_context *ctx, struct r600_pipe_state *state, unsigned rid);
void r600_context_pipe_state_set_vs_resource(struct r600_context *ctx, struct r600_pipe_state *state, unsigned rid);
void r600_context_pipe_state_set_fs_resource(struct r600_context *ctx, struct r600_pipe_state *state, unsigned rid);
void r600_context_pipe_state_set_ps_sampler(struct r600_context *ctx, struct r600_pipe_state *state, unsigned id);
void r600_context_pipe_state_set_vs_sampler(struct r600_context *ctx, struct r600_pipe_state *state, unsigned id);
void r600_context_flush(struct r600_context *ctx);
void r600_context_dump_bof(struct r600_context *ctx, const char *file);
void r600_context_draw(struct r600_context *ctx, const struct r600_draw *draw);
struct r600_query *r600_context_query_create(struct r600_context *ctx, unsigned query_type);
void r600_context_query_destroy(struct r600_context *ctx, struct r600_query *query);
boolean r600_context_query_result(struct r600_context *ctx,
struct r600_query *query,
boolean wait, void *vresult);
void r600_query_begin(struct r600_context *ctx, struct r600_query *query);
void r600_query_end(struct r600_context *ctx, struct r600_query *query);
void r600_context_queries_suspend(struct r600_context *ctx);
void r600_context_queries_resume(struct r600_context *ctx);
void r600_query_predication(struct r600_context *ctx, struct r600_query *query, int operation,
int flag_wait);
void r600_context_emit_fence(struct r600_context *ctx, struct r600_bo *fence,
unsigned offset, unsigned value);
void r600_context_flush_all(struct r600_context *ctx, unsigned flush_flags);
void r600_context_flush_dest_caches(struct r600_context *ctx);
int evergreen_context_init(struct r600_context *ctx, struct radeon *radeon);
void evergreen_context_draw(struct r600_context *ctx, const struct r600_draw *draw);
void evergreen_context_flush_dest_caches(struct r600_context *ctx);
void evergreen_context_pipe_state_set_ps_resource(struct r600_context *ctx, struct r600_pipe_state *state, unsigned rid);
void evergreen_context_pipe_state_set_vs_resource(struct r600_context *ctx, struct r600_pipe_state *state, unsigned rid);
void evergreen_context_pipe_state_set_fs_resource(struct r600_context *ctx, struct r600_pipe_state *state, unsigned rid);
void evergreen_context_pipe_state_set_ps_sampler(struct r600_context *ctx, struct r600_pipe_state *state, unsigned id);
void evergreen_context_pipe_state_set_vs_sampler(struct r600_context *ctx, struct r600_pipe_state *state, unsigned id);
struct radeon *radeon_decref(struct radeon *radeon);
void _r600_pipe_state_add_reg(struct r600_context *ctx,
struct r600_pipe_state *state,
u32 offset, u32 value, u32 mask,
u32 range_id, u32 block_id,
struct r600_bo *bo);
void r600_pipe_state_add_reg_noblock(struct r600_pipe_state *state,
u32 offset, u32 value, u32 mask,
struct r600_bo *bo);
#define r600_pipe_state_add_reg(state, offset, value, mask, bo) _r600_pipe_state_add_reg(&rctx->ctx, state, offset, value, mask, CTX_RANGE_ID(offset), CTX_BLOCK_ID(offset), bo)
static inline void r600_pipe_state_mod_reg(struct r600_pipe_state *state,
u32 value)
{
state->regs[state->nregs].value = value;
state->nregs++;
}
static inline void r600_pipe_state_mod_reg_bo(struct r600_pipe_state *state,
u32 value, struct r600_bo *bo)
{
state->regs[state->nregs].value = value;
state->regs[state->nregs].bo = bo;
state->nregs++;
}
#endif
|