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 (c) 2014-2018, NVIDIA CORPORATION. All rights reserved.
*
* 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 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.
*/
#ifndef SEMAPHORE_GK20A_H
#define SEMAPHORE_GK20A_H
#include <nvgpu/log.h>
#include <nvgpu/timers.h>
#include <nvgpu/atomic.h>
#include <nvgpu/bug.h>
#include <nvgpu/kref.h>
#include <nvgpu/list.h>
#include <nvgpu/nvgpu_mem.h>
#include "gk20a/gk20a.h"
#include "gk20a/mm_gk20a.h"
#include "gk20a/channel_gk20a.h"
#define gpu_sema_dbg(g, fmt, args...) \
nvgpu_log(g, gpu_dbg_sema, fmt, ##args)
#define gpu_sema_verbose_dbg(g, fmt, args...) \
nvgpu_log(g, gpu_dbg_sema_v, fmt, ##args)
/*
* Max number of channels that can be used is 512. This of course needs to be
* fixed to be dynamic but still fast.
*/
#define SEMAPHORE_POOL_COUNT 512
#define SEMAPHORE_SIZE 16
#define SEMAPHORE_SEA_GROWTH_RATE 32
struct nvgpu_semaphore_sea;
/*
* Underlying semaphore data structure. This semaphore can be shared amongst
* other semaphore instances.
*/
struct nvgpu_semaphore_int {
int idx; /* Semaphore index. */
u32 offset; /* Offset into the pool. */
nvgpu_atomic_t next_value; /* Next available value. */
struct nvgpu_semaphore_pool *p; /* Pool that owns this sema. */
struct channel_gk20a *ch; /* Channel that owns this sema. */
};
/*
* A semaphore which the rest of the driver actually uses. This consists of a
* pointer to a real semaphore and a value to wait for. This allows one physical
* semaphore to be shared among an essentially infinite number of submits.
*/
struct nvgpu_semaphore {
struct nvgpu_semaphore_int *hw_sema;
nvgpu_atomic_t value;
int incremented;
struct nvgpu_ref ref;
};
/*
* A semaphore pool. Each address space will own exactly one of these.
*/
struct nvgpu_semaphore_pool {
struct nvgpu_list_node pool_list_entry; /* Node for list of pools. */
u64 gpu_va; /* GPU access to the pool. */
u64 gpu_va_ro; /* GPU access to the pool. */
int page_idx; /* Index into sea bitmap. */
DECLARE_BITMAP(semas_alloced, PAGE_SIZE / SEMAPHORE_SIZE);
struct nvgpu_semaphore_sea *sema_sea; /* Sea that owns this pool. */
struct nvgpu_mutex pool_lock;
/*
* This is the address spaces's personal RW table. Other channels will
* ultimately map this page as RO. This is a sub-nvgpu_mem from the
* sea's mem.
*/
struct nvgpu_mem rw_mem;
int mapped;
/*
* Sometimes a channel can be released before other channels are
* done waiting on it. This ref count ensures that the pool doesn't
* go away until all semaphores using this pool are cleaned up first.
*/
struct nvgpu_ref ref;
};
static inline struct nvgpu_semaphore_pool *
nvgpu_semaphore_pool_from_pool_list_entry(struct nvgpu_list_node *node)
{
return (struct nvgpu_semaphore_pool *)
((uintptr_t)node -
offsetof(struct nvgpu_semaphore_pool, pool_list_entry));
};
/*
* A sea of semaphores pools. Each pool is owned by a single VM. Since multiple
* channels can share a VM each channel gets it's own HW semaphore from the
* pool. Channels then allocate regular semaphores - basically just a value that
* signifies when a particular job is done.
*/
struct nvgpu_semaphore_sea {
struct nvgpu_list_node pool_list; /* List of pools in this sea. */
struct gk20a *gk20a;
size_t size; /* Number of pages available. */
u64 gpu_va; /* GPU virtual address of sema sea. */
u64 map_size; /* Size of the mapping. */
/*
* TODO:
* List of pages that we use to back the pools. The number of pages
* can grow dynamically since allocating 512 pages for all channels at
* once would be a tremendous waste.
*/
int page_count; /* Pages allocated to pools. */
/*
* The read-only memory for the entire semaphore sea. Each semaphore
* pool needs a sub-nvgpu_mem that will be mapped as RW in its address
* space. This sea_mem cannot be freed until all semaphore_pools have
* been freed.
*/
struct nvgpu_mem sea_mem;
/*
* Can't use a regular allocator here since the full range of pools are
* not always allocated. Instead just use a bitmap.
*/
DECLARE_BITMAP(pools_alloced, SEMAPHORE_POOL_COUNT);
struct nvgpu_mutex sea_lock; /* Lock alloc/free calls. */
};
/*
* Semaphore sea functions.
*/
struct nvgpu_semaphore_sea *nvgpu_semaphore_sea_create(struct gk20a *gk20a);
void nvgpu_semaphore_sea_destroy(struct gk20a *g);
int nvgpu_semaphore_sea_map(struct nvgpu_semaphore_pool *sea,
struct vm_gk20a *vm);
void nvgpu_semaphore_sea_unmap(struct nvgpu_semaphore_pool *sea,
struct vm_gk20a *vm);
struct nvgpu_semaphore_sea *nvgpu_semaphore_get_sea(struct gk20a *g);
/*
* Semaphore pool functions.
*/
struct nvgpu_semaphore_pool *nvgpu_semaphore_pool_alloc(
struct nvgpu_semaphore_sea *sea);
int nvgpu_semaphore_pool_map(struct nvgpu_semaphore_pool *pool,
struct vm_gk20a *vm);
void nvgpu_semaphore_pool_unmap(struct nvgpu_semaphore_pool *pool,
struct vm_gk20a *vm);
u64 __nvgpu_semaphore_pool_gpu_va(struct nvgpu_semaphore_pool *p, bool global);
void nvgpu_semaphore_pool_get(struct nvgpu_semaphore_pool *p);
void nvgpu_semaphore_pool_put(struct nvgpu_semaphore_pool *p);
/*
* Semaphore functions.
*/
struct nvgpu_semaphore *nvgpu_semaphore_alloc(struct channel_gk20a *ch);
void nvgpu_semaphore_put(struct nvgpu_semaphore *s);
void nvgpu_semaphore_get(struct nvgpu_semaphore *s);
void nvgpu_semaphore_free_hw_sema(struct channel_gk20a *ch);
/*
* Return the address of a specific semaphore.
*
* Don't call this on a semaphore you don't own - the VA returned will make no
* sense in your specific channel's VM.
*/
static inline u64 nvgpu_semaphore_gpu_rw_va(struct nvgpu_semaphore *s)
{
return __nvgpu_semaphore_pool_gpu_va(s->hw_sema->p, false) +
s->hw_sema->offset;
}
/*
* Get the global RO address for the semaphore. Can be called on any semaphore
* regardless of whether you own it.
*/
static inline u64 nvgpu_semaphore_gpu_ro_va(struct nvgpu_semaphore *s)
{
return __nvgpu_semaphore_pool_gpu_va(s->hw_sema->p, true) +
s->hw_sema->offset;
}
static inline u64 nvgpu_hw_sema_addr(struct nvgpu_semaphore_int *hw_sema)
{
return __nvgpu_semaphore_pool_gpu_va(hw_sema->p, true) +
hw_sema->offset;
}
static inline u32 __nvgpu_semaphore_read(struct nvgpu_semaphore_int *hw_sema)
{
return nvgpu_mem_rd(hw_sema->ch->g,
&hw_sema->p->rw_mem, hw_sema->offset);
}
/*
* Read the underlying value from a semaphore.
*/
static inline u32 nvgpu_semaphore_read(struct nvgpu_semaphore *s)
{
return __nvgpu_semaphore_read(s->hw_sema);
}
/*
* Check if "racer" is over "goal" with wraparound handling.
*/
static inline bool __nvgpu_semaphore_value_released(u32 goal, u32 racer)
{
/*
* Handle wraparound with the same heuristic as the hardware does:
* although the integer will eventually wrap around, consider a sema
* released against a threshold if its value has passed that threshold
* but has not wrapped over half of the u32 range over that threshold;
* such wrapping is unlikely to happen during a sema lifetime.
*
* Values for [goal, goal + 0x7fffffff] are considered signaled; that's
* precisely half of the 32-bit space. If racer == goal + 0x80000000,
* then it needs 0x80000000 increments to wrap again and signal.
*
* Unsigned arithmetic is used because it's well-defined. This is
* effectively the same as: signed_racer - signed_goal > 0.
*/
return racer - goal < 0x80000000;
}
static inline u32 nvgpu_semaphore_get_value(struct nvgpu_semaphore *s)
{
return (u32)nvgpu_atomic_read(&s->value);
}
static inline bool nvgpu_semaphore_is_released(struct nvgpu_semaphore *s)
{
u32 sema_val = nvgpu_semaphore_read(s);
u32 wait_payload = nvgpu_semaphore_get_value(s);
return __nvgpu_semaphore_value_released(wait_payload, sema_val);
}
static inline bool nvgpu_semaphore_is_acquired(struct nvgpu_semaphore *s)
{
return !nvgpu_semaphore_is_released(s);
}
static inline u32 nvgpu_semaphore_next_value(struct nvgpu_semaphore *s)
{
return (u32)nvgpu_atomic_read(&s->hw_sema->next_value);
}
/*
* If @force is set then this will not wait for the underlying semaphore to
* catch up to the passed semaphore threshold.
*/
static inline void __nvgpu_semaphore_release(struct nvgpu_semaphore *s,
bool force)
{
struct nvgpu_semaphore_int *hw_sema = s->hw_sema;
u32 current_val;
u32 threshold = nvgpu_semaphore_get_value(s);
int attempts = 0;
/*
* Wait until the sema value is 1 less than the write value. That
* way this function is essentially an increment.
*
* TODO: tune the wait a little better.
*/
while (!__nvgpu_semaphore_value_released(threshold - 1,
current_val = nvgpu_semaphore_read(s))) {
if (force)
break;
nvgpu_msleep(100);
attempts += 1;
if (attempts > 100) {
WARN(1, "Stall on sema release!");
return;
}
}
/*
* If the semaphore has already passed the value we would write then
* this is really just a NO-OP.
*/
if (__nvgpu_semaphore_value_released(threshold, current_val))
return;
nvgpu_mem_wr(hw_sema->ch->g, &hw_sema->p->rw_mem, hw_sema->offset,
threshold);
gpu_sema_verbose_dbg(hw_sema->p->sema_sea->gk20a,
"(c=%d) WRITE %u", hw_sema->ch->chid, threshold);
}
static inline void nvgpu_semaphore_release(struct nvgpu_semaphore *s)
{
__nvgpu_semaphore_release(s, false);
}
/*
* Configure a software based increment on this semaphore. This is useful for
* when we want the GPU to wait on a SW event before processing a channel.
* Another way to describe this is when the GPU needs to wait on a SW pre-fence.
* The pre-fence signals SW which in turn calls nvgpu_semaphore_release() which
* then allows the GPU to continue.
*
* Also used to prep a semaphore for an INCR by the GPU.
*/
static inline void nvgpu_semaphore_incr(struct nvgpu_semaphore *s)
{
BUG_ON(s->incremented);
nvgpu_atomic_set(&s->value, nvgpu_atomic_add_return(1, &s->hw_sema->next_value));
s->incremented = 1;
gpu_sema_verbose_dbg(s->hw_sema->p->sema_sea->gk20a,
"INCR sema for c=%d (%u)",
s->hw_sema->ch->chid,
nvgpu_semaphore_next_value(s));
}
#endif
|