1 | // |
---|
2 | // Cforall Version 1.0.0 Copyright (C) 2019 University of Waterloo |
---|
3 | // |
---|
4 | // The contents of this file are covered under the licence agreement in the |
---|
5 | // file "LICENCE" distributed with Cforall. |
---|
6 | // |
---|
7 | // ready_queue.cfa -- |
---|
8 | // |
---|
9 | // Author : Thierry Delisle |
---|
10 | // Created On : Mon Nov dd 16:29:18 2019 |
---|
11 | // Last Modified By : |
---|
12 | // Last Modified On : |
---|
13 | // Update Count : |
---|
14 | // |
---|
15 | |
---|
16 | #define __cforall_thread__ |
---|
17 | // #define __CFA_DEBUG_PRINT_READY_QUEUE__ |
---|
18 | |
---|
19 | #include "bits/defs.hfa" |
---|
20 | #include "kernel_private.hfa" |
---|
21 | |
---|
22 | #define _GNU_SOURCE |
---|
23 | #include "stdlib.hfa" |
---|
24 | #include "math.hfa" |
---|
25 | |
---|
26 | static const size_t cache_line_size = 64; |
---|
27 | |
---|
28 | // No overriden function, no environment variable, no define |
---|
29 | // fall back to a magic number |
---|
30 | #ifndef __CFA_MAX_PROCESSORS__ |
---|
31 | #define __CFA_MAX_PROCESSORS__ 1024 |
---|
32 | #endif |
---|
33 | |
---|
34 | // returns the maximum number of processors the RWLock support |
---|
35 | __attribute__((weak)) unsigned __max_processors() { |
---|
36 | const char * max_cores_s = getenv("CFA_MAX_PROCESSORS"); |
---|
37 | if(!max_cores_s) { |
---|
38 | __cfadbg_print_nolock(ready_queue, "No CFA_MAX_PROCESSORS in ENV\n"); |
---|
39 | return __CFA_MAX_PROCESSORS__; |
---|
40 | } |
---|
41 | |
---|
42 | char * endptr = 0p; |
---|
43 | long int max_cores_l = strtol(max_cores_s, &endptr, 10); |
---|
44 | if(max_cores_l < 1 || max_cores_l > 65535) { |
---|
45 | __cfadbg_print_nolock(ready_queue, "CFA_MAX_PROCESSORS out of range : %ld\n", max_cores_l); |
---|
46 | return __CFA_MAX_PROCESSORS__; |
---|
47 | } |
---|
48 | if('\0' != *endptr) { |
---|
49 | __cfadbg_print_nolock(ready_queue, "CFA_MAX_PROCESSORS not a decimal number : %s\n", max_cores_s); |
---|
50 | return __CFA_MAX_PROCESSORS__; |
---|
51 | } |
---|
52 | |
---|
53 | return max_cores_l; |
---|
54 | } |
---|
55 | |
---|
56 | //======================================================================= |
---|
57 | // Cluster wide reader-writer lock |
---|
58 | //======================================================================= |
---|
59 | void ?{}(__scheduler_RWLock_t & this) { |
---|
60 | this.max = __max_processors(); |
---|
61 | this.alloc = 0; |
---|
62 | this.ready = 0; |
---|
63 | this.lock = false; |
---|
64 | this.data = alloc(this.max); |
---|
65 | |
---|
66 | /*paranoid*/ verify( 0 == (((uintptr_t)(this.data )) % 64) ); |
---|
67 | /*paranoid*/ verify( 0 == (((uintptr_t)(this.data + 1)) % 64) ); |
---|
68 | /*paranoid*/ verify(__atomic_is_lock_free(sizeof(this.alloc), &this.alloc)); |
---|
69 | /*paranoid*/ verify(__atomic_is_lock_free(sizeof(this.ready), &this.ready)); |
---|
70 | |
---|
71 | } |
---|
72 | void ^?{}(__scheduler_RWLock_t & this) { |
---|
73 | free(this.data); |
---|
74 | } |
---|
75 | |
---|
76 | void ?{}( __scheduler_lock_id_t & this, __processor_id_t * proc ) { |
---|
77 | this.handle = proc; |
---|
78 | this.lock = false; |
---|
79 | } |
---|
80 | |
---|
81 | //======================================================================= |
---|
82 | // Lock-Free registering/unregistering of threads |
---|
83 | unsigned doregister( struct __processor_id_t * proc ) with(*__scheduler_lock) { |
---|
84 | __cfadbg_print_safe(ready_queue, "Kernel : Registering proc %p for RW-Lock\n", proc); |
---|
85 | |
---|
86 | // Step - 1 : check if there is already space in the data |
---|
87 | uint_fast32_t s = ready; |
---|
88 | |
---|
89 | // Check among all the ready |
---|
90 | for(uint_fast32_t i = 0; i < s; i++) { |
---|
91 | __processor_id_t * null = 0p; // Re-write every loop since compare thrashes it |
---|
92 | if( __atomic_load_n(&data[i].handle, (int)__ATOMIC_RELAXED) == null |
---|
93 | && __atomic_compare_exchange_n( &data[i].handle, &null, proc, false, __ATOMIC_SEQ_CST, __ATOMIC_SEQ_CST)) { |
---|
94 | /*paranoid*/ verify(i < ready); |
---|
95 | /*paranoid*/ verify(__alignof__(data[i]) == cache_line_size); |
---|
96 | /*paranoid*/ verify((((uintptr_t)&data[i]) % cache_line_size) == 0); |
---|
97 | return i; |
---|
98 | } |
---|
99 | } |
---|
100 | |
---|
101 | if(max <= alloc) abort("Trying to create more than %ud processors", __scheduler_lock->max); |
---|
102 | |
---|
103 | // Step - 2 : F&A to get a new spot in the array. |
---|
104 | uint_fast32_t n = __atomic_fetch_add(&alloc, 1, __ATOMIC_SEQ_CST); |
---|
105 | if(max <= n) abort("Trying to create more than %ud processors", __scheduler_lock->max); |
---|
106 | |
---|
107 | // Step - 3 : Mark space as used and then publish it. |
---|
108 | __scheduler_lock_id_t * storage = (__scheduler_lock_id_t *)&data[n]; |
---|
109 | (*storage){ proc }; |
---|
110 | while(true) { |
---|
111 | unsigned copy = n; |
---|
112 | if( __atomic_load_n(&ready, __ATOMIC_RELAXED) == n |
---|
113 | && __atomic_compare_exchange_n(&ready, ©, n + 1, true, __ATOMIC_SEQ_CST, __ATOMIC_SEQ_CST)) |
---|
114 | break; |
---|
115 | asm volatile("pause"); |
---|
116 | } |
---|
117 | |
---|
118 | __cfadbg_print_safe(ready_queue, "Kernel : Registering proc %p done, id %lu\n", proc, n); |
---|
119 | |
---|
120 | // Return new spot. |
---|
121 | /*paranoid*/ verify(n < ready); |
---|
122 | /*paranoid*/ verify(__alignof__(data[n]) == cache_line_size); |
---|
123 | /*paranoid*/ verify((((uintptr_t)&data[n]) % cache_line_size) == 0); |
---|
124 | return n; |
---|
125 | } |
---|
126 | |
---|
127 | void unregister( struct __processor_id_t * proc ) with(*__scheduler_lock) { |
---|
128 | unsigned id = proc->id; |
---|
129 | /*paranoid*/ verify(id < ready); |
---|
130 | /*paranoid*/ verify(proc == __atomic_load_n(&data[id].handle, __ATOMIC_RELAXED)); |
---|
131 | __atomic_store_n(&data[id].handle, 0p, __ATOMIC_RELEASE); |
---|
132 | |
---|
133 | __cfadbg_print_safe(ready_queue, "Kernel : Unregister proc %p\n", proc); |
---|
134 | } |
---|
135 | |
---|
136 | //----------------------------------------------------------------------- |
---|
137 | // Writer side : acquire when changing the ready queue, e.g. adding more |
---|
138 | // queues or removing them. |
---|
139 | uint_fast32_t ready_mutate_lock( void ) with(*__scheduler_lock) { |
---|
140 | // Step 1 : lock global lock |
---|
141 | // It is needed to avoid processors that register mid Critical-Section |
---|
142 | // to simply lock their own lock and enter. |
---|
143 | __atomic_acquire( &lock ); |
---|
144 | |
---|
145 | // Step 2 : lock per-proc lock |
---|
146 | // Processors that are currently being registered aren't counted |
---|
147 | // but can't be in read_lock or in the critical section. |
---|
148 | // All other processors are counted |
---|
149 | uint_fast32_t s = ready; |
---|
150 | for(uint_fast32_t i = 0; i < s; i++) { |
---|
151 | __atomic_acquire( &data[i].lock ); |
---|
152 | } |
---|
153 | |
---|
154 | return s; |
---|
155 | } |
---|
156 | |
---|
157 | void ready_mutate_unlock( uint_fast32_t last_s ) with(*__scheduler_lock) { |
---|
158 | // Step 1 : release local locks |
---|
159 | // This must be done while the global lock is held to avoid |
---|
160 | // threads that where created mid critical section |
---|
161 | // to race to lock their local locks and have the writer |
---|
162 | // immidiately unlock them |
---|
163 | // Alternative solution : return s in write_lock and pass it to write_unlock |
---|
164 | for(uint_fast32_t i = 0; i < last_s; i++) { |
---|
165 | verify(data[i].lock); |
---|
166 | __atomic_store_n(&data[i].lock, (bool)false, __ATOMIC_RELEASE); |
---|
167 | } |
---|
168 | |
---|
169 | // Step 2 : release global lock |
---|
170 | /*paranoid*/ assert(true == lock); |
---|
171 | __atomic_store_n(&lock, (bool)false, __ATOMIC_RELEASE); |
---|
172 | } |
---|
173 | |
---|
174 | //======================================================================= |
---|
175 | // Intrusive Queue used by ready queue |
---|
176 | //======================================================================= |
---|
177 | // Intrusives lanes which are used by the relaxed ready queue |
---|
178 | struct __attribute__((aligned(128))) __intrusive_lane_t { |
---|
179 | // spin lock protecting the queue |
---|
180 | volatile bool lock; |
---|
181 | |
---|
182 | // anchor for the head and the tail of the queue |
---|
183 | struct __sentinel_t { |
---|
184 | // Link lists fields |
---|
185 | // instrusive link field for threads |
---|
186 | // must be exactly as in $thread |
---|
187 | __thread_desc_link link; |
---|
188 | } before, after; |
---|
189 | |
---|
190 | // Optional statistic counters |
---|
191 | #if !defined(__CFA_NO_SCHED_STATS__) |
---|
192 | struct __attribute__((aligned(64))) { |
---|
193 | // difference between number of push and pops |
---|
194 | ssize_t diff; |
---|
195 | |
---|
196 | // total number of pushes and pops |
---|
197 | size_t push; |
---|
198 | size_t pop ; |
---|
199 | } stat; |
---|
200 | #endif |
---|
201 | }; |
---|
202 | |
---|
203 | void ?{}(__intrusive_lane_t & this); |
---|
204 | void ^?{}(__intrusive_lane_t & this); |
---|
205 | |
---|
206 | // Get the head pointer (one before the first element) from the anchor |
---|
207 | static inline $thread * head(const __intrusive_lane_t & this) { |
---|
208 | $thread * rhead = ($thread *)( |
---|
209 | (uintptr_t)( &this.before ) - offsetof( $thread, link ) |
---|
210 | ); |
---|
211 | /* paranoid */ verify(rhead); |
---|
212 | return rhead; |
---|
213 | } |
---|
214 | |
---|
215 | // Get the tail pointer (one after the last element) from the anchor |
---|
216 | static inline $thread * tail(const __intrusive_lane_t & this) { |
---|
217 | $thread * rtail = ($thread *)( |
---|
218 | (uintptr_t)( &this.after ) - offsetof( $thread, link ) |
---|
219 | ); |
---|
220 | /* paranoid */ verify(rtail); |
---|
221 | return rtail; |
---|
222 | } |
---|
223 | |
---|
224 | // Ctor |
---|
225 | void ?{}( __intrusive_lane_t & this ) { |
---|
226 | this.lock = false; |
---|
227 | |
---|
228 | this.before.link.prev = 0p; |
---|
229 | this.before.link.next = tail(this); |
---|
230 | this.before.link.ts = 0; |
---|
231 | |
---|
232 | this.after .link.prev = head(this); |
---|
233 | this.after .link.next = 0p; |
---|
234 | this.after .link.ts = 0; |
---|
235 | |
---|
236 | #if !defined(__CFA_NO_SCHED_STATS__) |
---|
237 | this.stat.diff = 0; |
---|
238 | this.stat.push = 0; |
---|
239 | this.stat.pop = 0; |
---|
240 | #endif |
---|
241 | |
---|
242 | // We add a boat-load of assertions here because the anchor code is very fragile |
---|
243 | /* paranoid */ verify(((uintptr_t)( head(this) ) + offsetof( $thread, link )) == (uintptr_t)(&this.before)); |
---|
244 | /* paranoid */ verify(((uintptr_t)( tail(this) ) + offsetof( $thread, link )) == (uintptr_t)(&this.after )); |
---|
245 | /* paranoid */ verify(head(this)->link.prev == 0p ); |
---|
246 | /* paranoid */ verify(head(this)->link.next == tail(this) ); |
---|
247 | /* paranoid */ verify(tail(this)->link.next == 0p ); |
---|
248 | /* paranoid */ verify(tail(this)->link.prev == head(this) ); |
---|
249 | /* paranoid */ verify(&head(this)->link.prev == &this.before.link.prev ); |
---|
250 | /* paranoid */ verify(&head(this)->link.next == &this.before.link.next ); |
---|
251 | /* paranoid */ verify(&tail(this)->link.prev == &this.after .link.prev ); |
---|
252 | /* paranoid */ verify(&tail(this)->link.next == &this.after .link.next ); |
---|
253 | /* paranoid */ verify(sizeof(__intrusive_lane_t) == 128); |
---|
254 | /* paranoid */ verify(sizeof(this) == 128); |
---|
255 | /* paranoid */ verify(__alignof__(__intrusive_lane_t) == 128); |
---|
256 | /* paranoid */ verify(__alignof__(this) == 128); |
---|
257 | /* paranoid */ verifyf(((intptr_t)(&this) % 128) == 0, "Expected address to be aligned %p %% 128 == %zd", &this, ((intptr_t)(&this) % 128)); |
---|
258 | } |
---|
259 | |
---|
260 | // Dtor is trivial |
---|
261 | void ^?{}( __intrusive_lane_t & this ) { |
---|
262 | // Make sure the list is empty |
---|
263 | /* paranoid */ verify(head(this)->link.prev == 0p ); |
---|
264 | /* paranoid */ verify(head(this)->link.next == tail(this) ); |
---|
265 | /* paranoid */ verify(tail(this)->link.next == 0p ); |
---|
266 | /* paranoid */ verify(tail(this)->link.prev == head(this) ); |
---|
267 | } |
---|
268 | |
---|
269 | // Push a thread onto this lane |
---|
270 | // returns true of lane was empty before push, false otherwise |
---|
271 | bool push(__intrusive_lane_t & this, $thread * node) { |
---|
272 | #if defined(__CFA_WITH_VERIFY__) |
---|
273 | /* paranoid */ verify(this.lock); |
---|
274 | /* paranoid */ verify(node->link.ts != 0); |
---|
275 | /* paranoid */ verify(node->link.next == 0p); |
---|
276 | /* paranoid */ verify(node->link.prev == 0p); |
---|
277 | /* paranoid */ verify(tail(this)->link.next == 0p); |
---|
278 | /* paranoid */ verify(head(this)->link.prev == 0p); |
---|
279 | |
---|
280 | if(this.before.link.ts == 0l) { |
---|
281 | /* paranoid */ verify(tail(this)->link.prev == head(this)); |
---|
282 | /* paranoid */ verify(head(this)->link.next == tail(this)); |
---|
283 | } else { |
---|
284 | /* paranoid */ verify(tail(this)->link.prev != head(this)); |
---|
285 | /* paranoid */ verify(head(this)->link.next != tail(this)); |
---|
286 | } |
---|
287 | #endif |
---|
288 | |
---|
289 | // Get the relevant nodes locally |
---|
290 | $thread * tail = tail(this); |
---|
291 | $thread * prev = tail->link.prev; |
---|
292 | |
---|
293 | // Do the push |
---|
294 | node->link.next = tail; |
---|
295 | node->link.prev = prev; |
---|
296 | prev->link.next = node; |
---|
297 | tail->link.prev = node; |
---|
298 | |
---|
299 | // Update stats |
---|
300 | #if !defined(__CFA_NO_SCHED_STATS__) |
---|
301 | this.stat.diff++; |
---|
302 | this.stat.push++; |
---|
303 | #endif |
---|
304 | |
---|
305 | verify(node->link.next == tail(this)); |
---|
306 | |
---|
307 | // Check if the queue used to be empty |
---|
308 | if(this.before.link.ts == 0l) { |
---|
309 | this.before.link.ts = node->link.ts; |
---|
310 | /* paranoid */ verify(node->link.prev == head(this)); |
---|
311 | return true; |
---|
312 | } |
---|
313 | return false; |
---|
314 | } |
---|
315 | |
---|
316 | // Pop a thread from this lane (must be non-empty) |
---|
317 | // returns popped |
---|
318 | // returns true of lane was empty before push, false otherwise |
---|
319 | [$thread *, bool] pop(__intrusive_lane_t & this) { |
---|
320 | /* paranoid */ verify(this.lock); |
---|
321 | /* paranoid */ verify(this.before.link.ts != 0ul); |
---|
322 | |
---|
323 | // Get anchors locally |
---|
324 | $thread * head = head(this); |
---|
325 | $thread * tail = tail(this); |
---|
326 | |
---|
327 | // Get the relevant nodes locally |
---|
328 | $thread * node = head->link.next; |
---|
329 | $thread * next = node->link.next; |
---|
330 | |
---|
331 | /* paranoid */ verify(node != tail); |
---|
332 | /* paranoid */ verify(node); |
---|
333 | |
---|
334 | // Do the pop |
---|
335 | head->link.next = next; |
---|
336 | next->link.prev = head; |
---|
337 | node->link.[next, prev] = 0p; |
---|
338 | |
---|
339 | // Update head time stamp |
---|
340 | this.before.link.ts = next->link.ts; |
---|
341 | |
---|
342 | // Update stats |
---|
343 | #ifndef __CFA_NO_SCHED_STATS__ |
---|
344 | this.stat.diff--; |
---|
345 | this.stat.pop ++; |
---|
346 | #endif |
---|
347 | |
---|
348 | // Check if we emptied list and return accordingly |
---|
349 | /* paranoid */ verify(tail(this)->link.next == 0p); |
---|
350 | /* paranoid */ verify(head(this)->link.prev == 0p); |
---|
351 | if(next == tail) { |
---|
352 | /* paranoid */ verify(this.before.link.ts == 0); |
---|
353 | /* paranoid */ verify(tail(this)->link.prev == head(this)); |
---|
354 | /* paranoid */ verify(head(this)->link.next == tail(this)); |
---|
355 | return [node, true]; |
---|
356 | } |
---|
357 | else { |
---|
358 | /* paranoid */ verify(next->link.ts != 0); |
---|
359 | /* paranoid */ verify(tail(this)->link.prev != head(this)); |
---|
360 | /* paranoid */ verify(head(this)->link.next != tail(this)); |
---|
361 | /* paranoid */ verify(this.before.link.ts != 0); |
---|
362 | return [node, false]; |
---|
363 | } |
---|
364 | } |
---|
365 | |
---|
366 | // Check whether or not list is empty |
---|
367 | static inline bool is_empty(__intrusive_lane_t & this) { |
---|
368 | // Cannot verify here since it may not be locked |
---|
369 | return this.before.link.ts == 0; |
---|
370 | } |
---|
371 | |
---|
372 | // Return the timestamp |
---|
373 | static inline unsigned long long ts(__intrusive_lane_t & this) { |
---|
374 | // Cannot verify here since it may not be locked |
---|
375 | return this.before.link.ts; |
---|
376 | } |
---|
377 | |
---|
378 | //======================================================================= |
---|
379 | // Scalable Non-Zero counter |
---|
380 | //======================================================================= |
---|
381 | |
---|
382 | union __snzi_val_t { |
---|
383 | uint64_t _all; |
---|
384 | struct __attribute__((packed)) { |
---|
385 | char cnt; |
---|
386 | uint64_t ver:56; |
---|
387 | }; |
---|
388 | }; |
---|
389 | |
---|
390 | bool cas(volatile __snzi_val_t & self, __snzi_val_t & exp, char _cnt, uint64_t _ver) { |
---|
391 | __snzi_val_t t; |
---|
392 | t.ver = _ver; |
---|
393 | t.cnt = _cnt; |
---|
394 | /* paranoid */ verify(t._all == ((_ver << 8) | ((unsigned char)_cnt))); |
---|
395 | return __atomic_compare_exchange_n(&self._all, &exp._all, t._all, false, __ATOMIC_SEQ_CST, __ATOMIC_SEQ_CST); |
---|
396 | } |
---|
397 | |
---|
398 | bool cas(volatile __snzi_val_t & self, __snzi_val_t & exp, const __snzi_val_t & tar) { |
---|
399 | return __atomic_compare_exchange_n(&self._all, &exp._all, tar._all, false, __ATOMIC_SEQ_CST, __ATOMIC_SEQ_CST); |
---|
400 | } |
---|
401 | |
---|
402 | void ?{}( __snzi_val_t & this ) { this._all = 0; } |
---|
403 | void ?{}( __snzi_val_t & this, const volatile __snzi_val_t & o) { this._all = o._all; } |
---|
404 | |
---|
405 | struct __attribute__((aligned(128))) __snzi_node_t { |
---|
406 | volatile __snzi_val_t value; |
---|
407 | struct __snzi_node_t * parent; |
---|
408 | bool is_root; |
---|
409 | }; |
---|
410 | |
---|
411 | static inline void arrive( __snzi_node_t & ); |
---|
412 | static inline void depart( __snzi_node_t & ); |
---|
413 | |
---|
414 | #define __snzi_half -1 |
---|
415 | |
---|
416 | //-------------------------------------------------- |
---|
417 | // Root node |
---|
418 | static void arrive_r( __snzi_node_t & this ) { |
---|
419 | /* paranoid */ verify( this.is_root ); |
---|
420 | __atomic_fetch_add(&this.value._all, 1, __ATOMIC_SEQ_CST); |
---|
421 | } |
---|
422 | |
---|
423 | static void depart_r( __snzi_node_t & this ) { |
---|
424 | /* paranoid */ verify( this.is_root ); |
---|
425 | __atomic_fetch_sub(&this.value._all, 1, __ATOMIC_SEQ_CST); |
---|
426 | } |
---|
427 | |
---|
428 | //-------------------------------------------------- |
---|
429 | // Hierarchical node |
---|
430 | static void arrive_h( __snzi_node_t & this ) { |
---|
431 | int undoArr = 0; |
---|
432 | bool success = false; |
---|
433 | while(!success) { |
---|
434 | __snzi_val_t x = { this.value }; |
---|
435 | /* paranoid */ verify(x.cnt <= 120); |
---|
436 | if( x.cnt >= 1 ) { |
---|
437 | if( cas( this.value, x, x.cnt + 1, x.ver ) ) { |
---|
438 | success = true; |
---|
439 | } |
---|
440 | } |
---|
441 | /* paranoid */ verify(x.cnt <= 120); |
---|
442 | if( x.cnt == 0 ) { |
---|
443 | if( cas( this.value, x, __snzi_half, x.ver + 1) ) { |
---|
444 | success = true; |
---|
445 | x.cnt = __snzi_half; |
---|
446 | x.ver = x.ver + 1; |
---|
447 | } |
---|
448 | } |
---|
449 | /* paranoid */ verify(x.cnt <= 120); |
---|
450 | if( x.cnt == __snzi_half ) { |
---|
451 | /* paranoid */ verify( this.parent); |
---|
452 | arrive( *this.parent ); |
---|
453 | if( !cas( this.value, x, 1, x.ver) ) { |
---|
454 | undoArr = undoArr + 1; |
---|
455 | } |
---|
456 | } |
---|
457 | } |
---|
458 | |
---|
459 | for(int i = 0; i < undoArr; i++) { |
---|
460 | /* paranoid */ verify( this.parent ); |
---|
461 | depart( *this.parent ); |
---|
462 | } |
---|
463 | } |
---|
464 | |
---|
465 | static void depart_h( __snzi_node_t & this ) { |
---|
466 | while(true) { |
---|
467 | const __snzi_val_t x = { this.value }; |
---|
468 | /* paranoid */ verifyf(x.cnt >= 1, "%d", x.cnt); |
---|
469 | if( cas( this.value, x, x.cnt - 1, x.ver ) ) { |
---|
470 | if( x.cnt == 1 ) { |
---|
471 | /* paranoid */ verify( this.parent ); |
---|
472 | depart( *this.parent ); |
---|
473 | } |
---|
474 | return; |
---|
475 | } |
---|
476 | } |
---|
477 | } |
---|
478 | |
---|
479 | //-------------------------------------------------- |
---|
480 | // All nodes |
---|
481 | static inline void arrive( __snzi_node_t & this ) { |
---|
482 | if(this.is_root) arrive_r( this ); |
---|
483 | else arrive_h( this ); |
---|
484 | } |
---|
485 | |
---|
486 | static inline void depart( __snzi_node_t & this ) { |
---|
487 | if(this.is_root) depart_r( this ); |
---|
488 | else depart_h( this ); |
---|
489 | } |
---|
490 | |
---|
491 | static inline bool query( __snzi_node_t & this ) { |
---|
492 | /* paranoid */ verify( this.is_root ); |
---|
493 | return this.value._all > 0; |
---|
494 | } |
---|
495 | |
---|
496 | //-------------------------------------------------- |
---|
497 | // SNZI object |
---|
498 | void ?{}( __snzi_t & this, unsigned depth ) with( this ) { |
---|
499 | mask = (1 << depth) - 1; |
---|
500 | root = (1 << (depth + 1)) - 2; |
---|
501 | nodes = alloc( root + 1 ); |
---|
502 | |
---|
503 | int width = 1 << depth; |
---|
504 | for(int i = 0; i < root; i++) { |
---|
505 | nodes[i].value._all = 0; |
---|
506 | nodes[i].parent = &nodes[(i / 2) + width ]; |
---|
507 | nodes[i].is_root = false; |
---|
508 | } |
---|
509 | |
---|
510 | nodes[ root ].value._all = 0; |
---|
511 | nodes[ root ].parent = 0p; |
---|
512 | nodes[ root ].is_root = true; |
---|
513 | } |
---|
514 | |
---|
515 | void ^?{}( __snzi_t & this ) { |
---|
516 | free( this.nodes ); |
---|
517 | } |
---|
518 | |
---|
519 | static inline void arrive( __snzi_t & this, int idx) { |
---|
520 | idx &= this.mask; |
---|
521 | arrive( this.nodes[idx] ); |
---|
522 | } |
---|
523 | |
---|
524 | static inline void depart( __snzi_t & this, int idx) { |
---|
525 | idx &= this.mask; |
---|
526 | depart( this.nodes[idx] ); |
---|
527 | } |
---|
528 | |
---|
529 | static inline bool query( const __snzi_t & this ) { |
---|
530 | return query( this.nodes[ this.root ] ); |
---|
531 | } |
---|
532 | |
---|
533 | //======================================================================= |
---|
534 | // Cforall Reqdy Queue used by ready queue |
---|
535 | //======================================================================= |
---|
536 | |
---|
537 | void ?{}(__ready_queue_t & this) with (this) { |
---|
538 | |
---|
539 | lanes.data = alloc(4); |
---|
540 | for( i; 4 ) { |
---|
541 | (lanes.data[i]){}; |
---|
542 | } |
---|
543 | lanes.count = 4; |
---|
544 | snzi{ log2( lanes.count / 8 ) }; |
---|
545 | } |
---|
546 | |
---|
547 | void ^?{}(__ready_queue_t & this) with (this) { |
---|
548 | verify( 4 == lanes.count ); |
---|
549 | verify( !query( snzi ) ); |
---|
550 | |
---|
551 | ^(snzi){}; |
---|
552 | |
---|
553 | for( i; 4 ) { |
---|
554 | ^(lanes.data[i]){}; |
---|
555 | } |
---|
556 | free(lanes.data); |
---|
557 | } |
---|
558 | |
---|
559 | //----------------------------------------------------------------------- |
---|
560 | __attribute__((hot)) bool push(struct cluster * cltr, struct $thread * thrd) with (cltr->ready_queue) { |
---|
561 | __cfadbg_print_safe(ready_queue, "Kernel : Pushing %p on cluster %p\n", thrd, cltr); |
---|
562 | |
---|
563 | // write timestamp |
---|
564 | thrd->link.ts = rdtscl(); |
---|
565 | |
---|
566 | // Try to pick a lane and lock it |
---|
567 | unsigned i; |
---|
568 | do { |
---|
569 | // Pick the index of a lane |
---|
570 | i = __tls_rand() % lanes.count; |
---|
571 | |
---|
572 | #if !defined(__CFA_NO_STATISTICS__) |
---|
573 | __tls_stats()->ready.pick.push.attempt++; |
---|
574 | #endif |
---|
575 | |
---|
576 | // If we can't lock it retry |
---|
577 | } while( !__atomic_try_acquire( &lanes.data[i].lock ) ); |
---|
578 | |
---|
579 | bool first = false; |
---|
580 | |
---|
581 | // Actually push it |
---|
582 | bool lane_first = push(lanes.data[i], thrd); |
---|
583 | |
---|
584 | // If this lane used to be empty we need to do more |
---|
585 | if(lane_first) { |
---|
586 | // Check if the entire queue used to be empty |
---|
587 | first = !query(snzi); |
---|
588 | |
---|
589 | // Update the snzi |
---|
590 | arrive( snzi, i ); |
---|
591 | } |
---|
592 | |
---|
593 | // Unlock and return |
---|
594 | __atomic_unlock( &lanes.data[i].lock ); |
---|
595 | |
---|
596 | __cfadbg_print_safe(ready_queue, "Kernel : Pushed %p on cluster %p (idx: %u, mask %llu, first %d)\n", thrd, cltr, i, used.mask[0], lane_first); |
---|
597 | |
---|
598 | // Update statistics |
---|
599 | #if !defined(__CFA_NO_STATISTICS__) |
---|
600 | __tls_stats()->ready.pick.push.success++; |
---|
601 | #endif |
---|
602 | |
---|
603 | // return whether or not the list was empty before this push |
---|
604 | return first; |
---|
605 | } |
---|
606 | |
---|
607 | //----------------------------------------------------------------------- |
---|
608 | // Given 2 indexes, pick the list with the oldest push an try to pop from it |
---|
609 | static struct $thread * try_pop(struct cluster * cltr, unsigned i, unsigned j) with (cltr->ready_queue) { |
---|
610 | #if !defined(__CFA_NO_STATISTICS__) |
---|
611 | __tls_stats()->ready.pick.pop.attempt++; |
---|
612 | #endif |
---|
613 | |
---|
614 | // Pick the bet list |
---|
615 | int w = i; |
---|
616 | if( __builtin_expect(!is_empty(lanes.data[j]), true) ) { |
---|
617 | w = (ts(lanes.data[i]) < ts(lanes.data[j])) ? i : j; |
---|
618 | } |
---|
619 | |
---|
620 | // Get relevant elements locally |
---|
621 | __intrusive_lane_t & lane = lanes.data[w]; |
---|
622 | |
---|
623 | // If list looks empty retry |
---|
624 | if( is_empty(lane) ) return 0p; |
---|
625 | |
---|
626 | // If we can't get the lock retry |
---|
627 | if( !__atomic_try_acquire(&lane.lock) ) return 0p; |
---|
628 | |
---|
629 | |
---|
630 | // If list is empty, unlock and retry |
---|
631 | if( is_empty(lane) ) { |
---|
632 | __atomic_unlock(&lane.lock); |
---|
633 | return 0p; |
---|
634 | } |
---|
635 | |
---|
636 | // Actually pop the list |
---|
637 | struct $thread * thrd; |
---|
638 | bool emptied; |
---|
639 | [thrd, emptied] = pop(lane); |
---|
640 | |
---|
641 | /* paranoid */ verify(thrd); |
---|
642 | /* paranoid */ verify(lane.lock); |
---|
643 | |
---|
644 | // If this was the last element in the lane |
---|
645 | if(emptied) { |
---|
646 | depart( snzi, w ); |
---|
647 | } |
---|
648 | |
---|
649 | // Unlock and return |
---|
650 | __atomic_unlock(&lane.lock); |
---|
651 | |
---|
652 | // Update statistics |
---|
653 | #if !defined(__CFA_NO_STATISTICS__) |
---|
654 | __tls_stats()->ready.pick.pop.success++; |
---|
655 | #endif |
---|
656 | |
---|
657 | // return the popped thread |
---|
658 | return thrd; |
---|
659 | } |
---|
660 | |
---|
661 | // Pop from the ready queue from a given cluster |
---|
662 | __attribute__((hot)) $thread * pop(struct cluster * cltr) with (cltr->ready_queue) { |
---|
663 | /* paranoid */ verify( lanes.count > 0 ); |
---|
664 | |
---|
665 | // As long as the list is not empty, try finding a lane that isn't empty and pop from it |
---|
666 | while( query(snzi) ) { |
---|
667 | // Pick two lists at random |
---|
668 | int i = __tls_rand() % __atomic_load_n( &lanes.count, __ATOMIC_RELAXED ); |
---|
669 | int j = __tls_rand() % __atomic_load_n( &lanes.count, __ATOMIC_RELAXED ); |
---|
670 | |
---|
671 | // try popping from the 2 picked lists |
---|
672 | struct $thread * thrd = try_pop(cltr, i, j); |
---|
673 | if(thrd) return thrd; |
---|
674 | } |
---|
675 | |
---|
676 | // All lanes where empty return 0p |
---|
677 | return 0p; |
---|
678 | } |
---|
679 | |
---|
680 | //----------------------------------------------------------------------- |
---|
681 | |
---|
682 | static void check( __ready_queue_t & q ) with (q) { |
---|
683 | #if defined(__CFA_WITH_VERIFY__) |
---|
684 | { |
---|
685 | for( idx ; lanes.count ) { |
---|
686 | __intrusive_lane_t & sl = lanes.data[idx]; |
---|
687 | assert(!lanes.data[idx].lock); |
---|
688 | |
---|
689 | assert(head(sl)->link.prev == 0p ); |
---|
690 | assert(head(sl)->link.next->link.prev == head(sl) ); |
---|
691 | assert(tail(sl)->link.next == 0p ); |
---|
692 | assert(tail(sl)->link.prev->link.next == tail(sl) ); |
---|
693 | |
---|
694 | if(sl.before.link.ts == 0l) { |
---|
695 | assert(tail(sl)->link.prev == head(sl)); |
---|
696 | assert(head(sl)->link.next == tail(sl)); |
---|
697 | } else { |
---|
698 | assert(tail(sl)->link.prev != head(sl)); |
---|
699 | assert(head(sl)->link.next != tail(sl)); |
---|
700 | } |
---|
701 | } |
---|
702 | } |
---|
703 | #endif |
---|
704 | } |
---|
705 | |
---|
706 | // Call this function of the intrusive list was moved using memcpy |
---|
707 | // fixes the list so that the pointers back to anchors aren't left dangling |
---|
708 | static inline void fix(__intrusive_lane_t & ll) { |
---|
709 | // if the list is not empty then follow he pointer and fix its reverse |
---|
710 | if(!is_empty(ll)) { |
---|
711 | head(ll)->link.next->link.prev = head(ll); |
---|
712 | tail(ll)->link.prev->link.next = tail(ll); |
---|
713 | } |
---|
714 | // Otherwise just reset the list |
---|
715 | else { |
---|
716 | verify(tail(ll)->link.next == 0p); |
---|
717 | tail(ll)->link.prev = head(ll); |
---|
718 | head(ll)->link.next = tail(ll); |
---|
719 | verify(head(ll)->link.prev == 0p); |
---|
720 | } |
---|
721 | } |
---|
722 | |
---|
723 | // Grow the ready queue |
---|
724 | void ready_queue_grow (struct cluster * cltr) { |
---|
725 | // Lock the RWlock so no-one pushes/pops while we are changing the queue |
---|
726 | uint_fast32_t last_size = ready_mutate_lock(); |
---|
727 | |
---|
728 | __cfadbg_print_safe(ready_queue, "Kernel : Growing ready queue\n"); |
---|
729 | |
---|
730 | // Make sure that everything is consistent |
---|
731 | /* paranoid */ check( cltr->ready_queue ); |
---|
732 | |
---|
733 | // grow the ready queue |
---|
734 | with( cltr->ready_queue ) { |
---|
735 | ^(snzi){}; |
---|
736 | |
---|
737 | size_t ncount = lanes.count; |
---|
738 | |
---|
739 | // increase count |
---|
740 | ncount += 4; |
---|
741 | |
---|
742 | // Allocate new array (uses realloc and memcpies the data) |
---|
743 | lanes.data = alloc(lanes.data, ncount); |
---|
744 | |
---|
745 | // Fix the moved data |
---|
746 | for( idx; (size_t)lanes.count ) { |
---|
747 | fix(lanes.data[idx]); |
---|
748 | } |
---|
749 | |
---|
750 | // Construct new data |
---|
751 | for( idx; (size_t)lanes.count ~ ncount) { |
---|
752 | (lanes.data[idx]){}; |
---|
753 | } |
---|
754 | |
---|
755 | // Update original |
---|
756 | lanes.count = ncount; |
---|
757 | |
---|
758 | // Re-create the snzi |
---|
759 | snzi{ log2( lanes.count / 8 ) }; |
---|
760 | for( idx; (size_t)lanes.count ) { |
---|
761 | if( !is_empty(lanes.data[idx]) ) { |
---|
762 | arrive(snzi, idx); |
---|
763 | } |
---|
764 | } |
---|
765 | } |
---|
766 | |
---|
767 | // Make sure that everything is consistent |
---|
768 | /* paranoid */ check( cltr->ready_queue ); |
---|
769 | |
---|
770 | __cfadbg_print_safe(ready_queue, "Kernel : Growing ready queue done\n"); |
---|
771 | |
---|
772 | // Unlock the RWlock |
---|
773 | ready_mutate_unlock( last_size ); |
---|
774 | } |
---|
775 | |
---|
776 | // Shrink the ready queue |
---|
777 | void ready_queue_shrink(struct cluster * cltr) { |
---|
778 | // Lock the RWlock so no-one pushes/pops while we are changing the queue |
---|
779 | uint_fast32_t last_size = ready_mutate_lock(); |
---|
780 | |
---|
781 | __cfadbg_print_safe(ready_queue, "Kernel : Shrinking ready queue\n"); |
---|
782 | |
---|
783 | // Make sure that everything is consistent |
---|
784 | /* paranoid */ check( cltr->ready_queue ); |
---|
785 | |
---|
786 | with( cltr->ready_queue ) { |
---|
787 | ^(snzi){}; |
---|
788 | |
---|
789 | size_t ocount = lanes.count; |
---|
790 | // Check that we have some space left |
---|
791 | if(ocount < 8) abort("Program attempted to destroy more Ready Queues than were created"); |
---|
792 | |
---|
793 | // reduce the actual count so push doesn't use the old queues |
---|
794 | lanes.count -= 4; |
---|
795 | verify(ocount > lanes.count); |
---|
796 | |
---|
797 | // for printing count the number of displaced threads |
---|
798 | #if defined(__CFA_DEBUG_PRINT__) || defined(__CFA_DEBUG_PRINT_READY_QUEUE__) |
---|
799 | __attribute__((unused)) size_t displaced = 0; |
---|
800 | #endif |
---|
801 | |
---|
802 | // redistribute old data |
---|
803 | for( idx; (size_t)lanes.count ~ ocount) { |
---|
804 | // Lock is not strictly needed but makes checking invariants much easier |
---|
805 | __attribute__((unused)) bool locked = __atomic_try_acquire(&lanes.data[idx].lock); |
---|
806 | verify(locked); |
---|
807 | |
---|
808 | // As long as we can pop from this lane to push the threads somewhere else in the queue |
---|
809 | while(!is_empty(lanes.data[idx])) { |
---|
810 | struct $thread * thrd; |
---|
811 | __attribute__((unused)) bool _; |
---|
812 | [thrd, _] = pop(lanes.data[idx]); |
---|
813 | |
---|
814 | push(cltr, thrd); |
---|
815 | |
---|
816 | // for printing count the number of displaced threads |
---|
817 | #if defined(__CFA_DEBUG_PRINT__) || defined(__CFA_DEBUG_PRINT_READY_QUEUE__) |
---|
818 | displaced++; |
---|
819 | #endif |
---|
820 | } |
---|
821 | |
---|
822 | // Unlock the lane |
---|
823 | __atomic_unlock(&lanes.data[idx].lock); |
---|
824 | |
---|
825 | // TODO print the queue statistics here |
---|
826 | |
---|
827 | ^(lanes.data[idx]){}; |
---|
828 | } |
---|
829 | |
---|
830 | __cfadbg_print_safe(ready_queue, "Kernel : Shrinking ready queue displaced %zu threads\n", displaced); |
---|
831 | |
---|
832 | // Allocate new array (uses realloc and memcpies the data) |
---|
833 | lanes.data = alloc(lanes.data, lanes.count); |
---|
834 | |
---|
835 | // Fix the moved data |
---|
836 | for( idx; (size_t)lanes.count ) { |
---|
837 | fix(lanes.data[idx]); |
---|
838 | } |
---|
839 | |
---|
840 | // Re-create the snzi |
---|
841 | snzi{ log2( lanes.count / 8 ) }; |
---|
842 | for( idx; (size_t)lanes.count ) { |
---|
843 | if( !is_empty(lanes.data[idx]) ) { |
---|
844 | arrive(snzi, idx); |
---|
845 | } |
---|
846 | } |
---|
847 | } |
---|
848 | |
---|
849 | // Make sure that everything is consistent |
---|
850 | /* paranoid */ check( cltr->ready_queue ); |
---|
851 | |
---|
852 | __cfadbg_print_safe(ready_queue, "Kernel : Shrinking ready queue done\n"); |
---|
853 | |
---|
854 | // Unlock the RWlock |
---|
855 | ready_mutate_unlock( last_size ); |
---|
856 | } |
---|