summaryrefslogtreecommitdiffstats
path: root/src/mesa/drivers/dri/i965/brw_bufmgr.h
blob: ee913240432df56692b65b2d4d7deadab091c888 (plain)
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
341
342
343
344
345
346
347
348
/*
 * Copyright © 2008-2012 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.
 *
 * Authors:
 *    Eric Anholt <eric@anholt.net>
 *
 */

/**
 * @file brw_bufmgr.h
 *
 * Public definitions of Intel-specific bufmgr functions.
 */

#ifndef INTEL_BUFMGR_H
#define INTEL_BUFMGR_H

#include <stdbool.h>
#include <stdint.h>
#include <stdio.h>
#include "util/u_atomic.h"
#include "util/list.h"

#if defined(__cplusplus)
extern "C" {
#endif

struct gen_device_info;
struct brw_context;

struct brw_bo {
   /**
    * Size in bytes of the buffer object.
    *
    * The size may be larger than the size originally requested for the
    * allocation, such as being aligned to page size.
    */
   uint64_t size;

   /**
    * Alignment requirement for object
    *
    * Used for GTT mapping & pinning the object.
    */
   uint64_t align;

   /** Buffer manager context associated with this buffer object */
   struct brw_bufmgr *bufmgr;

   /** The GEM handle for this buffer object. */
   uint32_t gem_handle;

   /**
    * Offset of the buffer inside the Graphics Translation Table.
    *
    * This is effectively our GPU address for the buffer and we use it
    * as our base for all state pointers into the buffer. However, since the
    * kernel may be forced to move it around during the course of the
    * buffer's lifetime, we can only know where the buffer was on the last
    * execbuf. We presume, and are usually right, that the buffer will not
    * move and so we use that last offset for the next batch and by doing
    * so we can avoid having the kernel perform a relocation fixup pass as
    * our pointers inside the batch will be using the correct base offset.
    *
    * Since we do use it as a base address for the next batch of pointers,
    * the kernel treats our offset as a request, and if possible will
    * arrange the buffer to placed at that address (trying to balance
    * the cost of buffer migration versus the cost of performing
    * relocations). Furthermore, we can force the kernel to place the buffer,
    * or report a failure if we specified a conflicting offset, at our chosen
    * offset by specifying EXEC_OBJECT_PINNED.
    *
    * Note the GTT may be either per context, or shared globally across the
    * system. On a shared system, our buffers have to contend for address
    * space with both aperture mappings and framebuffers and so are more
    * likely to be moved. On a full ppGTT system, each batch exists in its
    * own GTT, and so each buffer may have their own offset within each
    * context.
    */
   uint64_t gtt_offset;

   /**
    * The validation list index for this buffer, or -1 when not in a batch.
    * Note that a single buffer may be in multiple batches (contexts), and
    * this is a global field, which refers to the last batch using the BO.
    * It should not be considered authoritative, but can be used to avoid a
    * linear walk of the validation list in the common case by guessing that
    * exec_bos[bo->index] == bo and confirming whether that's the case.
    */
   unsigned index;

   /**
    * Boolean of whether the GPU is definitely not accessing the buffer.
    *
    * This is only valid when reusable, since non-reusable
    * buffers are those that have been shared with other
    * processes, so we don't know their state.
    */
   bool idle;

   int refcount;
   const char *name;

#ifndef EXEC_OBJECT_CAPTURE
#define EXEC_OBJECT_CAPTURE            (1<<7)
#endif
   uint64_t kflags;

   /**
    * Kenel-assigned global name for this object
    *
    * List contains both flink named and prime fd'd objects
    */
   unsigned int global_name;

   /**
    * Current tiling mode
    */
   uint32_t tiling_mode;
   uint32_t swizzle_mode;
   uint32_t stride;

   time_t free_time;

   /** Mapped address for the buffer, saved across map/unmap cycles */
   void *map_cpu;
   /** GTT virtual address for the buffer, saved across map/unmap cycles */
   void *map_gtt;
   /** WC CPU address for the buffer, saved across map/unmap cycles */
   void *map_wc;

   /** BO cache list */
   struct list_head head;

   /**
    * Boolean of whether this buffer can be re-used
    */
   bool reusable;

   /**
    * Boolean of whether this buffer has been shared with an external client.
    */
   bool external;

   /**
    * Boolean of whether this buffer is cache coherent
    */
   bool cache_coherent;
};

#define BO_ALLOC_BUSY       (1<<0)
#define BO_ALLOC_ZEROED     (1<<1)

/**
 * Allocate a buffer object.
 *
 * Buffer objects are not necessarily initially mapped into CPU virtual
 * address space or graphics device aperture.  They must be mapped
 * using brw_bo_map() to be used by the CPU.
 */
struct brw_bo *brw_bo_alloc(struct brw_bufmgr *bufmgr, const char *name,
                            uint64_t size, uint64_t alignment);

/**
 * Allocate a tiled buffer object.
 *
 * Alignment for tiled objects is set automatically; the 'flags'
 * argument provides a hint about how the object will be used initially.
 *
 * Valid tiling formats are:
 *  I915_TILING_NONE
 *  I915_TILING_X
 *  I915_TILING_Y
 */
struct brw_bo *brw_bo_alloc_tiled(struct brw_bufmgr *bufmgr,
                                  const char *name,
                                  uint64_t size,
                                  uint32_t tiling_mode,
                                  uint32_t pitch,
                                  unsigned flags);

/**
 * Allocate a tiled buffer object.
 *
 * Alignment for tiled objects is set automatically; the 'flags'
 * argument provides a hint about how the object will be used initially.
 *
 * Valid tiling formats are:
 *  I915_TILING_NONE
 *  I915_TILING_X
 *  I915_TILING_Y
 *
 * Note the tiling format may be rejected; callers should check the
 * 'tiling_mode' field on return, as well as the pitch value, which
 * may have been rounded up to accommodate for tiling restrictions.
 */
struct brw_bo *brw_bo_alloc_tiled_2d(struct brw_bufmgr *bufmgr,
                                     const char *name,
                                     int x, int y, int cpp,
                                     uint32_t tiling_mode,
                                     uint32_t *pitch,
                                     unsigned flags);

/** Takes a reference on a buffer object */
static inline void
brw_bo_reference(struct brw_bo *bo)
{
   p_atomic_inc(&bo->refcount);
}

/**
 * Releases a reference on a buffer object, freeing the data if
 * no references remain.
 */
void brw_bo_unreference(struct brw_bo *bo);

/* Must match MapBufferRange interface (for convenience) */
#define MAP_READ        GL_MAP_READ_BIT
#define MAP_WRITE       GL_MAP_WRITE_BIT
#define MAP_ASYNC       GL_MAP_UNSYNCHRONIZED_BIT
#define MAP_PERSISTENT  GL_MAP_PERSISTENT_BIT
#define MAP_COHERENT    GL_MAP_COHERENT_BIT
/* internal */
#define MAP_INTERNAL_MASK       (0xff << 24)
#define MAP_RAW                 (0x01 << 24)

/**
 * Maps the buffer into userspace.
 *
 * This function will block waiting for any existing execution on the
 * buffer to complete, first.  The resulting mapping is returned.
 */
MUST_CHECK void *brw_bo_map(struct brw_context *brw, struct brw_bo *bo, unsigned flags);

/**
 * Reduces the refcount on the userspace mapping of the buffer
 * object.
 */
static inline int brw_bo_unmap(struct brw_bo *bo) { return 0; }

/** Write data into an object. */
int brw_bo_subdata(struct brw_bo *bo, uint64_t offset,
                   uint64_t size, const void *data);
/**
 * Waits for rendering to an object by the GPU to have completed.
 *
 * This is not required for any access to the BO by bo_map,
 * bo_subdata, etc.  It is merely a way for the driver to implement
 * glFinish.
 */
void brw_bo_wait_rendering(struct brw_bo *bo);

/**
 * Tears down the buffer manager instance.
 */
void brw_bufmgr_destroy(struct brw_bufmgr *bufmgr);

/**
 * Get the current tiling (and resulting swizzling) mode for the bo.
 *
 * \param buf Buffer to get tiling mode for
 * \param tiling_mode returned tiling mode
 * \param swizzle_mode returned swizzling mode
 */
int brw_bo_get_tiling(struct brw_bo *bo, uint32_t *tiling_mode,
                      uint32_t *swizzle_mode);

/**
 * Create a visible name for a buffer which can be used by other apps
 *
 * \param buf Buffer to create a name for
 * \param name Returned name
 */
int brw_bo_flink(struct brw_bo *bo, uint32_t *name);

/**
 * Returns 1 if mapping the buffer for write could cause the process
 * to block, due to the object being active in the GPU.
 */
int brw_bo_busy(struct brw_bo *bo);

/**
 * Specify the volatility of the buffer.
 * \param bo Buffer to create a name for
 * \param madv The purgeable status
 *
 * Use I915_MADV_DONTNEED to mark the buffer as purgeable, and it will be
 * reclaimed under memory pressure. If you subsequently require the buffer,
 * then you must pass I915_MADV_WILLNEED to mark the buffer as required.
 *
 * Returns 1 if the buffer was retained, or 0 if it was discarded whilst
 * marked as I915_MADV_DONTNEED.
 */
int brw_bo_madvise(struct brw_bo *bo, int madv);

/* drm_bacon_bufmgr_gem.c */
struct brw_bufmgr *brw_bufmgr_init(struct gen_device_info *devinfo, int fd);
struct brw_bo *brw_bo_gem_create_from_name(struct brw_bufmgr *bufmgr,
                                           const char *name,
                                           unsigned int handle);
void brw_bufmgr_enable_reuse(struct brw_bufmgr *bufmgr);

int brw_bo_wait(struct brw_bo *bo, int64_t timeout_ns);

uint32_t brw_create_hw_context(struct brw_bufmgr *bufmgr);

#define BRW_CONTEXT_LOW_PRIORITY ((I915_CONTEXT_MIN_USER_PRIORITY-1)/2)
#define BRW_CONTEXT_MEDIUM_PRIORITY (I915_CONTEXT_DEFAULT_PRIORITY)
#define BRW_CONTEXT_HIGH_PRIORITY ((I915_CONTEXT_MAX_USER_PRIORITY+1)/2)

int brw_hw_context_set_priority(struct brw_bufmgr *bufmgr,
                                uint32_t ctx_id,
                                int priority);

void brw_destroy_hw_context(struct brw_bufmgr *bufmgr, uint32_t ctx_id);

int brw_bo_gem_export_to_prime(struct brw_bo *bo, int *prime_fd);
struct brw_bo *brw_bo_gem_create_from_prime(struct brw_bufmgr *bufmgr,
                                            int prime_fd);

int brw_reg_read(struct brw_bufmgr *bufmgr, uint32_t offset,
                 uint64_t *result);

/** @{ */

#if defined(__cplusplus)
}
#endif
#endif /* INTEL_BUFMGR_H */