Line data Source code
1 : /*-------------------------------------------------------------------------
2 : *
3 : * deadlock.c
4 : * POSTGRES deadlock detection code
5 : *
6 : * See src/backend/storage/lmgr/README for a description of the deadlock
7 : * detection and resolution algorithms.
8 : *
9 : *
10 : * Portions Copyright (c) 1996-2023, PostgreSQL Global Development Group
11 : * Portions Copyright (c) 1994, Regents of the University of California
12 : *
13 : *
14 : * IDENTIFICATION
15 : * src/backend/storage/lmgr/deadlock.c
16 : *
17 : * Interface:
18 : *
19 : * DeadLockCheck()
20 : * DeadLockReport()
21 : * RememberSimpleDeadLock()
22 : * InitDeadLockChecking()
23 : *
24 : *-------------------------------------------------------------------------
25 : */
26 : #include "postgres.h"
27 :
28 : #include "miscadmin.h"
29 : #include "pg_trace.h"
30 : #include "pgstat.h"
31 : #include "storage/lmgr.h"
32 : #include "storage/proc.h"
33 : #include "utils/memutils.h"
34 :
35 :
36 : /*
37 : * One edge in the waits-for graph.
38 : *
39 : * waiter and blocker may or may not be members of a lock group, but if either
40 : * is, it will be the leader rather than any other member of the lock group.
41 : * The group leaders act as representatives of the whole group even though
42 : * those particular processes need not be waiting at all. There will be at
43 : * least one member of the waiter's lock group on the wait queue for the given
44 : * lock, maybe more.
45 : */
46 : typedef struct
47 : {
48 : PGPROC *waiter; /* the leader of the waiting lock group */
49 : PGPROC *blocker; /* the leader of the group it is waiting for */
50 : LOCK *lock; /* the lock being waited for */
51 : int pred; /* workspace for TopoSort */
52 : int link; /* workspace for TopoSort */
53 : } EDGE;
54 :
55 : /* One potential reordering of a lock's wait queue */
56 : typedef struct
57 : {
58 : LOCK *lock; /* the lock whose wait queue is described */
59 : PGPROC **procs; /* array of PGPROC *'s in new wait order */
60 : int nProcs;
61 : } WAIT_ORDER;
62 :
63 : /*
64 : * Information saved about each edge in a detected deadlock cycle. This
65 : * is used to print a diagnostic message upon failure.
66 : *
67 : * Note: because we want to examine this info after releasing the lock
68 : * manager's partition locks, we can't just store LOCK and PGPROC pointers;
69 : * we must extract out all the info we want to be able to print.
70 : */
71 : typedef struct
72 : {
73 : LOCKTAG locktag; /* ID of awaited lock object */
74 : LOCKMODE lockmode; /* type of lock we're waiting for */
75 : int pid; /* PID of blocked backend */
76 : } DEADLOCK_INFO;
77 :
78 :
79 : static bool DeadLockCheckRecurse(PGPROC *proc);
80 : static int TestConfiguration(PGPROC *startProc);
81 : static bool FindLockCycle(PGPROC *checkProc,
82 : EDGE *softEdges, int *nSoftEdges);
83 : static bool FindLockCycleRecurse(PGPROC *checkProc, int depth,
84 : EDGE *softEdges, int *nSoftEdges);
85 : static bool FindLockCycleRecurseMember(PGPROC *checkProc,
86 : PGPROC *checkProcLeader,
87 : int depth, EDGE *softEdges, int *nSoftEdges);
88 : static bool ExpandConstraints(EDGE *constraints, int nConstraints);
89 : static bool TopoSort(LOCK *lock, EDGE *constraints, int nConstraints,
90 : PGPROC **ordering);
91 :
92 : #ifdef DEBUG_DEADLOCK
93 : static void PrintLockQueue(LOCK *lock, const char *info);
94 : #endif
95 :
96 :
97 : /*
98 : * Working space for the deadlock detector
99 : */
100 :
101 : /* Workspace for FindLockCycle */
102 : static PGPROC **visitedProcs; /* Array of visited procs */
103 : static int nVisitedProcs;
104 :
105 : /* Workspace for TopoSort */
106 : static PGPROC **topoProcs; /* Array of not-yet-output procs */
107 : static int *beforeConstraints; /* Counts of remaining before-constraints */
108 : static int *afterConstraints; /* List head for after-constraints */
109 :
110 : /* Output area for ExpandConstraints */
111 : static WAIT_ORDER *waitOrders; /* Array of proposed queue rearrangements */
112 : static int nWaitOrders;
113 : static PGPROC **waitOrderProcs; /* Space for waitOrders queue contents */
114 :
115 : /* Current list of constraints being considered */
116 : static EDGE *curConstraints;
117 : static int nCurConstraints;
118 : static int maxCurConstraints;
119 :
120 : /* Storage space for results from FindLockCycle */
121 : static EDGE *possibleConstraints;
122 : static int nPossibleConstraints;
123 : static int maxPossibleConstraints;
124 : static DEADLOCK_INFO *deadlockDetails;
125 : static int nDeadlockDetails;
126 :
127 : /* PGPROC pointer of any blocking autovacuum worker found */
128 : static PGPROC *blocking_autovacuum_proc = NULL;
129 :
130 :
131 : /*
132 : * InitDeadLockChecking -- initialize deadlock checker during backend startup
133 : *
134 : * This does per-backend initialization of the deadlock checker; primarily,
135 : * allocation of working memory for DeadLockCheck. We do this per-backend
136 : * since there's no percentage in making the kernel do copy-on-write
137 : * inheritance of workspace from the postmaster. We want to allocate the
138 : * space at startup because (a) the deadlock checker might be invoked when
139 : * there's no free memory left, and (b) the checker is normally run inside a
140 : * signal handler, which is a very dangerous place to invoke palloc from.
141 : */
142 : void
143 24110 : InitDeadLockChecking(void)
144 : {
145 : MemoryContext oldcxt;
146 :
147 : /* Make sure allocations are permanent */
148 24110 : oldcxt = MemoryContextSwitchTo(TopMemoryContext);
149 :
150 : /*
151 : * FindLockCycle needs at most MaxBackends entries in visitedProcs[] and
152 : * deadlockDetails[].
153 : */
154 24110 : visitedProcs = (PGPROC **) palloc(MaxBackends * sizeof(PGPROC *));
155 24110 : deadlockDetails = (DEADLOCK_INFO *) palloc(MaxBackends * sizeof(DEADLOCK_INFO));
156 :
157 : /*
158 : * TopoSort needs to consider at most MaxBackends wait-queue entries, and
159 : * it needn't run concurrently with FindLockCycle.
160 : */
161 24110 : topoProcs = visitedProcs; /* re-use this space */
162 24110 : beforeConstraints = (int *) palloc(MaxBackends * sizeof(int));
163 24110 : afterConstraints = (int *) palloc(MaxBackends * sizeof(int));
164 :
165 : /*
166 : * We need to consider rearranging at most MaxBackends/2 wait queues
167 : * (since it takes at least two waiters in a queue to create a soft edge),
168 : * and the expanded form of the wait queues can't involve more than
169 : * MaxBackends total waiters.
170 : */
171 24110 : waitOrders = (WAIT_ORDER *)
172 24110 : palloc((MaxBackends / 2) * sizeof(WAIT_ORDER));
173 24110 : waitOrderProcs = (PGPROC **) palloc(MaxBackends * sizeof(PGPROC *));
174 :
175 : /*
176 : * Allow at most MaxBackends distinct constraints in a configuration. (Is
177 : * this enough? In practice it seems it should be, but I don't quite see
178 : * how to prove it. If we run out, we might fail to find a workable wait
179 : * queue rearrangement even though one exists.) NOTE that this number
180 : * limits the maximum recursion depth of DeadLockCheckRecurse. Making it
181 : * really big might potentially allow a stack-overflow problem.
182 : */
183 24110 : maxCurConstraints = MaxBackends;
184 24110 : curConstraints = (EDGE *) palloc(maxCurConstraints * sizeof(EDGE));
185 :
186 : /*
187 : * Allow up to 3*MaxBackends constraints to be saved without having to
188 : * re-run TestConfiguration. (This is probably more than enough, but we
189 : * can survive if we run low on space by doing excess runs of
190 : * TestConfiguration to re-compute constraint lists each time needed.) The
191 : * last MaxBackends entries in possibleConstraints[] are reserved as
192 : * output workspace for FindLockCycle.
193 : */
194 24110 : maxPossibleConstraints = MaxBackends * 4;
195 24110 : possibleConstraints =
196 24110 : (EDGE *) palloc(maxPossibleConstraints * sizeof(EDGE));
197 :
198 24110 : MemoryContextSwitchTo(oldcxt);
199 24110 : }
200 :
201 : /*
202 : * DeadLockCheck -- Checks for deadlocks for a given process
203 : *
204 : * This code looks for deadlocks involving the given process. If any
205 : * are found, it tries to rearrange lock wait queues to resolve the
206 : * deadlock. If resolution is impossible, return DS_HARD_DEADLOCK ---
207 : * the caller is then expected to abort the given proc's transaction.
208 : *
209 : * Caller must already have locked all partitions of the lock tables.
210 : *
211 : * On failure, deadlock details are recorded in deadlockDetails[] for
212 : * subsequent printing by DeadLockReport(). That activity is separate
213 : * because (a) we don't want to do it while holding all those LWLocks,
214 : * and (b) we are typically invoked inside a signal handler.
215 : */
216 : DeadLockState
217 44 : DeadLockCheck(PGPROC *proc)
218 : {
219 : /* Initialize to "no constraints" */
220 44 : nCurConstraints = 0;
221 44 : nPossibleConstraints = 0;
222 44 : nWaitOrders = 0;
223 :
224 : /* Initialize to not blocked by an autovacuum worker */
225 44 : blocking_autovacuum_proc = NULL;
226 :
227 : /* Search for deadlocks and possible fixes */
228 44 : if (DeadLockCheckRecurse(proc))
229 : {
230 : /*
231 : * Call FindLockCycle one more time, to record the correct
232 : * deadlockDetails[] for the basic state with no rearrangements.
233 : */
234 : int nSoftEdges;
235 :
236 : TRACE_POSTGRESQL_DEADLOCK_FOUND();
237 :
238 8 : nWaitOrders = 0;
239 8 : if (!FindLockCycle(proc, possibleConstraints, &nSoftEdges))
240 0 : elog(FATAL, "deadlock seems to have disappeared");
241 :
242 8 : return DS_HARD_DEADLOCK; /* cannot find a non-deadlocked state */
243 : }
244 :
245 : /* Apply any needed rearrangements of wait queues */
246 42 : for (int i = 0; i < nWaitOrders; i++)
247 : {
248 6 : LOCK *lock = waitOrders[i].lock;
249 6 : PGPROC **procs = waitOrders[i].procs;
250 6 : int nProcs = waitOrders[i].nProcs;
251 6 : dclist_head *waitQueue = &lock->waitProcs;
252 :
253 : Assert(nProcs == dclist_count(waitQueue));
254 :
255 : #ifdef DEBUG_DEADLOCK
256 : PrintLockQueue(lock, "DeadLockCheck:");
257 : #endif
258 :
259 : /* Reset the queue and re-add procs in the desired order */
260 6 : dclist_init(waitQueue);
261 24 : for (int j = 0; j < nProcs; j++)
262 18 : dclist_push_tail(waitQueue, &procs[j]->links);
263 :
264 : #ifdef DEBUG_DEADLOCK
265 : PrintLockQueue(lock, "rearranged to:");
266 : #endif
267 :
268 : /* See if any waiters for the lock can be woken up now */
269 6 : ProcLockWakeup(GetLocksMethodTable(lock), lock);
270 : }
271 :
272 : /* Return code tells caller if we had to escape a deadlock or not */
273 36 : if (nWaitOrders > 0)
274 6 : return DS_SOFT_DEADLOCK;
275 30 : else if (blocking_autovacuum_proc != NULL)
276 0 : return DS_BLOCKED_BY_AUTOVACUUM;
277 : else
278 30 : return DS_NO_DEADLOCK;
279 : }
280 :
281 : /*
282 : * Return the PGPROC of the autovacuum that's blocking a process.
283 : *
284 : * We reset the saved pointer as soon as we pass it back.
285 : */
286 : PGPROC *
287 0 : GetBlockingAutoVacuumPgproc(void)
288 : {
289 : PGPROC *ptr;
290 :
291 0 : ptr = blocking_autovacuum_proc;
292 0 : blocking_autovacuum_proc = NULL;
293 :
294 0 : return ptr;
295 : }
296 :
297 : /*
298 : * DeadLockCheckRecurse -- recursively search for valid orderings
299 : *
300 : * curConstraints[] holds the current set of constraints being considered
301 : * by an outer level of recursion. Add to this each possible solution
302 : * constraint for any cycle detected at this level.
303 : *
304 : * Returns true if no solution exists. Returns false if a deadlock-free
305 : * state is attainable, in which case waitOrders[] shows the required
306 : * rearrangements of lock wait queues (if any).
307 : */
308 : static bool
309 50 : DeadLockCheckRecurse(PGPROC *proc)
310 : {
311 : int nEdges;
312 : int oldPossibleConstraints;
313 : bool savedList;
314 : int i;
315 :
316 50 : nEdges = TestConfiguration(proc);
317 50 : if (nEdges < 0)
318 8 : return true; /* hard deadlock --- no solution */
319 42 : if (nEdges == 0)
320 36 : return false; /* good configuration found */
321 6 : if (nCurConstraints >= maxCurConstraints)
322 0 : return true; /* out of room for active constraints? */
323 6 : oldPossibleConstraints = nPossibleConstraints;
324 6 : if (nPossibleConstraints + nEdges + MaxBackends <= maxPossibleConstraints)
325 : {
326 : /* We can save the edge list in possibleConstraints[] */
327 6 : nPossibleConstraints += nEdges;
328 6 : savedList = true;
329 : }
330 : else
331 : {
332 : /* Not room; will need to regenerate the edges on-the-fly */
333 0 : savedList = false;
334 : }
335 :
336 : /*
337 : * Try each available soft edge as an addition to the configuration.
338 : */
339 6 : for (i = 0; i < nEdges; i++)
340 : {
341 6 : if (!savedList && i > 0)
342 : {
343 : /* Regenerate the list of possible added constraints */
344 0 : if (nEdges != TestConfiguration(proc))
345 0 : elog(FATAL, "inconsistent results during deadlock check");
346 : }
347 6 : curConstraints[nCurConstraints] =
348 6 : possibleConstraints[oldPossibleConstraints + i];
349 6 : nCurConstraints++;
350 6 : if (!DeadLockCheckRecurse(proc))
351 6 : return false; /* found a valid solution! */
352 : /* give up on that added constraint, try again */
353 0 : nCurConstraints--;
354 : }
355 0 : nPossibleConstraints = oldPossibleConstraints;
356 0 : return true; /* no solution found */
357 : }
358 :
359 :
360 : /*--------------------
361 : * Test a configuration (current set of constraints) for validity.
362 : *
363 : * Returns:
364 : * 0: the configuration is good (no deadlocks)
365 : * -1: the configuration has a hard deadlock or is not self-consistent
366 : * >0: the configuration has one or more soft deadlocks
367 : *
368 : * In the soft-deadlock case, one of the soft cycles is chosen arbitrarily
369 : * and a list of its soft edges is returned beginning at
370 : * possibleConstraints+nPossibleConstraints. The return value is the
371 : * number of soft edges.
372 : *--------------------
373 : */
374 : static int
375 50 : TestConfiguration(PGPROC *startProc)
376 : {
377 50 : int softFound = 0;
378 50 : EDGE *softEdges = possibleConstraints + nPossibleConstraints;
379 : int nSoftEdges;
380 : int i;
381 :
382 : /*
383 : * Make sure we have room for FindLockCycle's output.
384 : */
385 50 : if (nPossibleConstraints + MaxBackends > maxPossibleConstraints)
386 0 : return -1;
387 :
388 : /*
389 : * Expand current constraint set into wait orderings. Fail if the
390 : * constraint set is not self-consistent.
391 : */
392 50 : if (!ExpandConstraints(curConstraints, nCurConstraints))
393 0 : return -1;
394 :
395 : /*
396 : * Check for cycles involving startProc or any of the procs mentioned in
397 : * constraints. We check startProc last because if it has a soft cycle
398 : * still to be dealt with, we want to deal with that first.
399 : */
400 56 : for (i = 0; i < nCurConstraints; i++)
401 : {
402 6 : if (FindLockCycle(curConstraints[i].waiter, softEdges, &nSoftEdges))
403 : {
404 0 : if (nSoftEdges == 0)
405 0 : return -1; /* hard deadlock detected */
406 0 : softFound = nSoftEdges;
407 : }
408 6 : if (FindLockCycle(curConstraints[i].blocker, softEdges, &nSoftEdges))
409 : {
410 0 : if (nSoftEdges == 0)
411 0 : return -1; /* hard deadlock detected */
412 0 : softFound = nSoftEdges;
413 : }
414 : }
415 50 : if (FindLockCycle(startProc, softEdges, &nSoftEdges))
416 : {
417 14 : if (nSoftEdges == 0)
418 8 : return -1; /* hard deadlock detected */
419 6 : softFound = nSoftEdges;
420 : }
421 42 : return softFound;
422 : }
423 :
424 :
425 : /*
426 : * FindLockCycle -- basic check for deadlock cycles
427 : *
428 : * Scan outward from the given proc to see if there is a cycle in the
429 : * waits-for graph that includes this proc. Return true if a cycle
430 : * is found, else false. If a cycle is found, we return a list of
431 : * the "soft edges", if any, included in the cycle. These edges could
432 : * potentially be eliminated by rearranging wait queues. We also fill
433 : * deadlockDetails[] with information about the detected cycle; this info
434 : * is not used by the deadlock algorithm itself, only to print a useful
435 : * message after failing.
436 : *
437 : * Since we need to be able to check hypothetical configurations that would
438 : * exist after wait queue rearrangement, the routine pays attention to the
439 : * table of hypothetical queue orders in waitOrders[]. These orders will
440 : * be believed in preference to the actual ordering seen in the locktable.
441 : */
442 : static bool
443 70 : FindLockCycle(PGPROC *checkProc,
444 : EDGE *softEdges, /* output argument */
445 : int *nSoftEdges) /* output argument */
446 : {
447 70 : nVisitedProcs = 0;
448 70 : nDeadlockDetails = 0;
449 70 : *nSoftEdges = 0;
450 70 : return FindLockCycleRecurse(checkProc, 0, softEdges, nSoftEdges);
451 : }
452 :
453 : static bool
454 222 : FindLockCycleRecurse(PGPROC *checkProc,
455 : int depth,
456 : EDGE *softEdges, /* output argument */
457 : int *nSoftEdges) /* output argument */
458 : {
459 : int i;
460 : dlist_iter iter;
461 :
462 : /*
463 : * If this process is a lock group member, check the leader instead. (Note
464 : * that we might be the leader, in which case this is a no-op.)
465 : */
466 222 : if (checkProc->lockGroupLeader != NULL)
467 52 : checkProc = checkProc->lockGroupLeader;
468 :
469 : /*
470 : * Have we already seen this proc?
471 : */
472 494 : for (i = 0; i < nVisitedProcs; i++)
473 : {
474 306 : if (visitedProcs[i] == checkProc)
475 : {
476 : /* If we return to starting point, we have a deadlock cycle */
477 34 : if (i == 0)
478 : {
479 : /*
480 : * record total length of cycle --- outer levels will now fill
481 : * deadlockDetails[]
482 : */
483 : Assert(depth <= MaxBackends);
484 22 : nDeadlockDetails = depth;
485 :
486 22 : return true;
487 : }
488 :
489 : /*
490 : * Otherwise, we have a cycle but it does not include the start
491 : * point, so say "no deadlock".
492 : */
493 12 : return false;
494 : }
495 : }
496 : /* Mark proc as seen */
497 : Assert(nVisitedProcs < MaxBackends);
498 188 : visitedProcs[nVisitedProcs++] = checkProc;
499 :
500 : /*
501 : * If the process is waiting, there is an outgoing waits-for edge to each
502 : * process that blocks it.
503 : */
504 318 : if (checkProc->links.next != NULL && checkProc->waitLock != NULL &&
505 130 : FindLockCycleRecurseMember(checkProc, checkProc, depth, softEdges,
506 : nSoftEdges))
507 74 : return true;
508 :
509 : /*
510 : * If the process is not waiting, there could still be outgoing waits-for
511 : * edges if it is part of a lock group, because other members of the lock
512 : * group might be waiting even though this process is not. (Given lock
513 : * groups {A1, A2} and {B1, B2}, if A1 waits for B1 and B2 waits for A2,
514 : * that is a deadlock even neither of B1 and A2 are waiting for anything.)
515 : */
516 196 : dlist_foreach(iter, &checkProc->lockGroupMembers)
517 : {
518 : PGPROC *memberProc;
519 :
520 90 : memberProc = dlist_container(PGPROC, lockGroupLink, iter.cur);
521 :
522 90 : if (memberProc->links.next != NULL && memberProc->waitLock != NULL &&
523 42 : memberProc != checkProc &&
524 42 : FindLockCycleRecurseMember(memberProc, checkProc, depth, softEdges,
525 : nSoftEdges))
526 8 : return true;
527 : }
528 :
529 106 : return false;
530 : }
531 :
532 : static bool
533 172 : FindLockCycleRecurseMember(PGPROC *checkProc,
534 : PGPROC *checkProcLeader,
535 : int depth,
536 : EDGE *softEdges, /* output argument */
537 : int *nSoftEdges) /* output argument */
538 : {
539 : PGPROC *proc;
540 172 : LOCK *lock = checkProc->waitLock;
541 : dlist_iter proclock_iter;
542 : LockMethod lockMethodTable;
543 : int conflictMask;
544 : int i;
545 : int numLockModes,
546 : lm;
547 :
548 : /*
549 : * The relation extension lock can never participate in actual deadlock
550 : * cycle. See Assert in LockAcquireExtended. So, there is no advantage
551 : * in checking wait edges from it.
552 : */
553 172 : if (LOCK_LOCKTAG(*lock) == LOCKTAG_RELATION_EXTEND)
554 0 : return false;
555 :
556 172 : lockMethodTable = GetLocksMethodTable(lock);
557 172 : numLockModes = lockMethodTable->numLockModes;
558 172 : conflictMask = lockMethodTable->conflictTab[checkProc->waitLockMode];
559 :
560 : /*
561 : * Scan for procs that already hold conflicting locks. These are "hard"
562 : * edges in the waits-for graph.
563 : */
564 512 : dlist_foreach(proclock_iter, &lock->procLocks)
565 : {
566 412 : PROCLOCK *proclock = dlist_container(PROCLOCK, lockLink, proclock_iter.cur);
567 : PGPROC *leader;
568 :
569 412 : proc = proclock->tag.myProc;
570 412 : leader = proc->lockGroupLeader == NULL ? proc : proc->lockGroupLeader;
571 :
572 : /* A proc never blocks itself or any other lock group member */
573 412 : if (leader != checkProcLeader)
574 : {
575 2110 : for (lm = 1; lm <= numLockModes; lm++)
576 : {
577 1962 : if ((proclock->holdMask & LOCKBIT_ON(lm)) &&
578 : (conflictMask & LOCKBIT_ON(lm)))
579 : {
580 : /* This proc hard-blocks checkProc */
581 120 : if (FindLockCycleRecurse(proc, depth + 1,
582 : softEdges, nSoftEdges))
583 : {
584 : /* fill deadlockDetails[] */
585 72 : DEADLOCK_INFO *info = &deadlockDetails[depth];
586 :
587 72 : info->locktag = lock->tag;
588 72 : info->lockmode = checkProc->waitLockMode;
589 72 : info->pid = checkProc->pid;
590 :
591 72 : return true;
592 : }
593 :
594 : /*
595 : * No deadlock here, but see if this proc is an autovacuum
596 : * that is directly hard-blocking our own proc. If so,
597 : * report it so that the caller can send a cancel signal
598 : * to it, if appropriate. If there's more than one such
599 : * proc, it's indeterminate which one will be reported.
600 : *
601 : * We don't touch autovacuums that are indirectly blocking
602 : * us; it's up to the direct blockee to take action. This
603 : * rule simplifies understanding the behavior and ensures
604 : * that an autovacuum won't be canceled with less than
605 : * deadlock_timeout grace period.
606 : *
607 : * Note we read statusFlags without any locking. This is
608 : * OK only for checking the PROC_IS_AUTOVACUUM flag,
609 : * because that flag is set at process start and never
610 : * reset. There is logic elsewhere to avoid canceling an
611 : * autovacuum that is working to prevent XID wraparound
612 : * problems (which needs to read a different statusFlags
613 : * bit), but we don't do that here to avoid grabbing
614 : * ProcArrayLock.
615 : */
616 48 : if (checkProc == MyProc &&
617 28 : proc->statusFlags & PROC_IS_AUTOVACUUM)
618 0 : blocking_autovacuum_proc = proc;
619 :
620 : /* We're done looking at this proclock */
621 48 : break;
622 : }
623 : }
624 : }
625 : }
626 :
627 : /*
628 : * Scan for procs that are ahead of this one in the lock's wait queue.
629 : * Those that have conflicting requests soft-block this one. This must be
630 : * done after the hard-block search, since if another proc both hard- and
631 : * soft-blocks this one, we want to call it a hard edge.
632 : *
633 : * If there is a proposed re-ordering of the lock's wait order, use that
634 : * rather than the current wait order.
635 : */
636 118 : for (i = 0; i < nWaitOrders; i++)
637 : {
638 54 : if (waitOrders[i].lock == lock)
639 36 : break;
640 : }
641 :
642 100 : if (i < nWaitOrders)
643 : {
644 : /* Use the given hypothetical wait queue order */
645 36 : PGPROC **procs = waitOrders[i].procs;
646 36 : int queue_size = waitOrders[i].nProcs;
647 :
648 46 : for (i = 0; i < queue_size; i++)
649 : {
650 : PGPROC *leader;
651 :
652 46 : proc = procs[i];
653 46 : leader = proc->lockGroupLeader == NULL ? proc :
654 : proc->lockGroupLeader;
655 :
656 : /*
657 : * TopoSort will always return an ordering with group members
658 : * adjacent to each other in the wait queue (see comments
659 : * therein). So, as soon as we reach a process in the same lock
660 : * group as checkProc, we know we've found all the conflicts that
661 : * precede any member of the lock group lead by checkProcLeader.
662 : */
663 46 : if (leader == checkProcLeader)
664 36 : break;
665 :
666 : /* Is there a conflict with this guy's request? */
667 10 : if ((LOCKBIT_ON(proc->waitLockMode) & conflictMask) != 0)
668 : {
669 : /* This proc soft-blocks checkProc */
670 10 : if (FindLockCycleRecurse(proc, depth + 1,
671 : softEdges, nSoftEdges))
672 : {
673 : /* fill deadlockDetails[] */
674 0 : DEADLOCK_INFO *info = &deadlockDetails[depth];
675 :
676 0 : info->locktag = lock->tag;
677 0 : info->lockmode = checkProc->waitLockMode;
678 0 : info->pid = checkProc->pid;
679 :
680 : /*
681 : * Add this edge to the list of soft edges in the cycle
682 : */
683 : Assert(*nSoftEdges < MaxBackends);
684 0 : softEdges[*nSoftEdges].waiter = checkProcLeader;
685 0 : softEdges[*nSoftEdges].blocker = leader;
686 0 : softEdges[*nSoftEdges].lock = lock;
687 0 : (*nSoftEdges)++;
688 0 : return true;
689 : }
690 : }
691 : }
692 : }
693 : else
694 : {
695 64 : PGPROC *lastGroupMember = NULL;
696 : dlist_iter proc_iter;
697 : dclist_head *waitQueue;
698 :
699 : /* Use the true lock wait queue order */
700 64 : waitQueue = &lock->waitProcs;
701 :
702 : /*
703 : * Find the last member of the lock group that is present in the wait
704 : * queue. Anything after this is not a soft lock conflict. If group
705 : * locking is not in use, then we know immediately which process we're
706 : * looking for, but otherwise we've got to search the wait queue to
707 : * find the last process actually present.
708 : */
709 64 : if (checkProc->lockGroupLeader == NULL)
710 46 : lastGroupMember = checkProc;
711 : else
712 : {
713 64 : dclist_foreach(proc_iter, waitQueue)
714 : {
715 46 : proc = dlist_container(PGPROC, links, proc_iter.cur);
716 :
717 46 : if (proc->lockGroupLeader == checkProcLeader)
718 26 : lastGroupMember = proc;
719 : }
720 : Assert(lastGroupMember != NULL);
721 : }
722 :
723 : /*
724 : * OK, now rescan (or scan) the queue to identify the soft conflicts.
725 : */
726 86 : dclist_foreach(proc_iter, waitQueue)
727 : {
728 : PGPROC *leader;
729 :
730 86 : proc = dlist_container(PGPROC, links, proc_iter.cur);
731 :
732 86 : leader = proc->lockGroupLeader == NULL ? proc :
733 : proc->lockGroupLeader;
734 :
735 : /* Done when we reach the target proc */
736 86 : if (proc == lastGroupMember)
737 54 : break;
738 :
739 : /* Is there a conflict with this guy's request? */
740 32 : if ((LOCKBIT_ON(proc->waitLockMode) & conflictMask) != 0 &&
741 : leader != checkProcLeader)
742 : {
743 : /* This proc soft-blocks checkProc */
744 22 : if (FindLockCycleRecurse(proc, depth + 1,
745 : softEdges, nSoftEdges))
746 : {
747 : /* fill deadlockDetails[] */
748 10 : DEADLOCK_INFO *info = &deadlockDetails[depth];
749 :
750 10 : info->locktag = lock->tag;
751 10 : info->lockmode = checkProc->waitLockMode;
752 10 : info->pid = checkProc->pid;
753 :
754 : /*
755 : * Add this edge to the list of soft edges in the cycle
756 : */
757 : Assert(*nSoftEdges < MaxBackends);
758 10 : softEdges[*nSoftEdges].waiter = checkProcLeader;
759 10 : softEdges[*nSoftEdges].blocker = leader;
760 10 : softEdges[*nSoftEdges].lock = lock;
761 10 : (*nSoftEdges)++;
762 10 : return true;
763 : }
764 : }
765 : }
766 : }
767 :
768 : /*
769 : * No conflict detected here.
770 : */
771 90 : return false;
772 : }
773 :
774 :
775 : /*
776 : * ExpandConstraints -- expand a list of constraints into a set of
777 : * specific new orderings for affected wait queues
778 : *
779 : * Input is a list of soft edges to be reversed. The output is a list
780 : * of nWaitOrders WAIT_ORDER structs in waitOrders[], with PGPROC array
781 : * workspace in waitOrderProcs[].
782 : *
783 : * Returns true if able to build an ordering that satisfies all the
784 : * constraints, false if not (there are contradictory constraints).
785 : */
786 : static bool
787 50 : ExpandConstraints(EDGE *constraints,
788 : int nConstraints)
789 : {
790 50 : int nWaitOrderProcs = 0;
791 : int i,
792 : j;
793 :
794 50 : nWaitOrders = 0;
795 :
796 : /*
797 : * Scan constraint list backwards. This is because the last-added
798 : * constraint is the only one that could fail, and so we want to test it
799 : * for inconsistency first.
800 : */
801 56 : for (i = nConstraints; --i >= 0;)
802 : {
803 6 : LOCK *lock = constraints[i].lock;
804 :
805 : /* Did we already make a list for this lock? */
806 6 : for (j = nWaitOrders; --j >= 0;)
807 : {
808 0 : if (waitOrders[j].lock == lock)
809 0 : break;
810 : }
811 6 : if (j >= 0)
812 0 : continue;
813 : /* No, so allocate a new list */
814 6 : waitOrders[nWaitOrders].lock = lock;
815 6 : waitOrders[nWaitOrders].procs = waitOrderProcs + nWaitOrderProcs;
816 6 : waitOrders[nWaitOrders].nProcs = dclist_count(&lock->waitProcs);
817 6 : nWaitOrderProcs += dclist_count(&lock->waitProcs);
818 : Assert(nWaitOrderProcs <= MaxBackends);
819 :
820 : /*
821 : * Do the topo sort. TopoSort need not examine constraints after this
822 : * one, since they must be for different locks.
823 : */
824 6 : if (!TopoSort(lock, constraints, i + 1,
825 6 : waitOrders[nWaitOrders].procs))
826 0 : return false;
827 6 : nWaitOrders++;
828 : }
829 50 : return true;
830 : }
831 :
832 :
833 : /*
834 : * TopoSort -- topological sort of a wait queue
835 : *
836 : * Generate a re-ordering of a lock's wait queue that satisfies given
837 : * constraints about certain procs preceding others. (Each such constraint
838 : * is a fact of a partial ordering.) Minimize rearrangement of the queue
839 : * not needed to achieve the partial ordering.
840 : *
841 : * This is a lot simpler and slower than, for example, the topological sort
842 : * algorithm shown in Knuth's Volume 1. However, Knuth's method doesn't
843 : * try to minimize the damage to the existing order. In practice we are
844 : * not likely to be working with more than a few constraints, so the apparent
845 : * slowness of the algorithm won't really matter.
846 : *
847 : * The initial queue ordering is taken directly from the lock's wait queue.
848 : * The output is an array of PGPROC pointers, of length equal to the lock's
849 : * wait queue length (the caller is responsible for providing this space).
850 : * The partial order is specified by an array of EDGE structs. Each EDGE
851 : * is one that we need to reverse, therefore the "waiter" must appear before
852 : * the "blocker" in the output array. The EDGE array may well contain
853 : * edges associated with other locks; these should be ignored.
854 : *
855 : * Returns true if able to build an ordering that satisfies all the
856 : * constraints, false if not (there are contradictory constraints).
857 : */
858 : static bool
859 6 : TopoSort(LOCK *lock,
860 : EDGE *constraints,
861 : int nConstraints,
862 : PGPROC **ordering) /* output argument */
863 : {
864 6 : dclist_head *waitQueue = &lock->waitProcs;
865 6 : int queue_size = dclist_count(waitQueue);
866 : PGPROC *proc;
867 : int i,
868 : j,
869 : jj,
870 : k,
871 : kk,
872 : last;
873 : dlist_iter proc_iter;
874 :
875 : /* First, fill topoProcs[] array with the procs in their current order */
876 6 : i = 0;
877 24 : dclist_foreach(proc_iter, waitQueue)
878 : {
879 18 : proc = dlist_container(PGPROC, links, proc_iter.cur);
880 18 : topoProcs[i++] = proc;
881 : }
882 : Assert(i == queue_size);
883 :
884 : /*
885 : * Scan the constraints, and for each proc in the array, generate a count
886 : * of the number of constraints that say it must be before something else,
887 : * plus a list of the constraints that say it must be after something
888 : * else. The count for the j'th proc is stored in beforeConstraints[j],
889 : * and the head of its list in afterConstraints[j]. Each constraint
890 : * stores its list link in constraints[i].link (note any constraint will
891 : * be in just one list). The array index for the before-proc of the i'th
892 : * constraint is remembered in constraints[i].pred.
893 : *
894 : * Note that it's not necessarily the case that every constraint affects
895 : * this particular wait queue. Prior to group locking, a process could be
896 : * waiting for at most one lock. But a lock group can be waiting for
897 : * zero, one, or multiple locks. Since topoProcs[] is an array of the
898 : * processes actually waiting, while constraints[] is an array of group
899 : * leaders, we've got to scan through topoProcs[] for each constraint,
900 : * checking whether both a waiter and a blocker for that group are
901 : * present. If so, the constraint is relevant to this wait queue; if not,
902 : * it isn't.
903 : */
904 12 : MemSet(beforeConstraints, 0, queue_size * sizeof(int));
905 12 : MemSet(afterConstraints, 0, queue_size * sizeof(int));
906 12 : for (i = 0; i < nConstraints; i++)
907 : {
908 : /*
909 : * Find a representative process that is on the lock queue and part of
910 : * the waiting lock group. This may or may not be the leader, which
911 : * may or may not be waiting at all. If there are any other processes
912 : * in the same lock group on the queue, set their number of
913 : * beforeConstraints to -1 to indicate that they should be emitted
914 : * with their groupmates rather than considered separately.
915 : *
916 : * In this loop and the similar one just below, it's critical that we
917 : * consistently select the same representative member of any one lock
918 : * group, so that all the constraints are associated with the same
919 : * proc, and the -1's are only associated with not-representative
920 : * members. We select the last one in the topoProcs array.
921 : */
922 6 : proc = constraints[i].waiter;
923 : Assert(proc != NULL);
924 6 : jj = -1;
925 24 : for (j = queue_size; --j >= 0;)
926 : {
927 18 : PGPROC *waiter = topoProcs[j];
928 :
929 18 : if (waiter == proc || waiter->lockGroupLeader == proc)
930 : {
931 : Assert(waiter->waitLock == lock);
932 10 : if (jj == -1)
933 6 : jj = j;
934 : else
935 : {
936 : Assert(beforeConstraints[j] <= 0);
937 4 : beforeConstraints[j] = -1;
938 : }
939 : }
940 : }
941 :
942 : /* If no matching waiter, constraint is not relevant to this lock. */
943 6 : if (jj < 0)
944 0 : continue;
945 :
946 : /*
947 : * Similarly, find a representative process that is on the lock queue
948 : * and waiting for the blocking lock group. Again, this could be the
949 : * leader but does not need to be.
950 : */
951 6 : proc = constraints[i].blocker;
952 : Assert(proc != NULL);
953 6 : kk = -1;
954 24 : for (k = queue_size; --k >= 0;)
955 : {
956 18 : PGPROC *blocker = topoProcs[k];
957 :
958 18 : if (blocker == proc || blocker->lockGroupLeader == proc)
959 : {
960 : Assert(blocker->waitLock == lock);
961 6 : if (kk == -1)
962 6 : kk = k;
963 : else
964 : {
965 : Assert(beforeConstraints[k] <= 0);
966 0 : beforeConstraints[k] = -1;
967 : }
968 : }
969 : }
970 :
971 : /* If no matching blocker, constraint is not relevant to this lock. */
972 6 : if (kk < 0)
973 0 : continue;
974 :
975 : Assert(beforeConstraints[jj] >= 0);
976 6 : beforeConstraints[jj]++; /* waiter must come before */
977 : /* add this constraint to list of after-constraints for blocker */
978 6 : constraints[i].pred = jj;
979 6 : constraints[i].link = afterConstraints[kk];
980 6 : afterConstraints[kk] = i + 1;
981 : }
982 :
983 : /*--------------------
984 : * Now scan the topoProcs array backwards. At each step, output the
985 : * last proc that has no remaining before-constraints plus any other
986 : * members of the same lock group; then decrease the beforeConstraints
987 : * count of each of the procs it was constrained against.
988 : * i = index of ordering[] entry we want to output this time
989 : * j = search index for topoProcs[]
990 : * k = temp for scanning constraint list for proc j
991 : * last = last non-null index in topoProcs (avoid redundant searches)
992 : *--------------------
993 : */
994 6 : last = queue_size - 1;
995 20 : for (i = queue_size - 1; i >= 0;)
996 : {
997 : int c;
998 14 : int nmatches = 0;
999 :
1000 : /* Find next candidate to output */
1001 14 : while (topoProcs[last] == NULL)
1002 0 : last--;
1003 28 : for (j = last; j >= 0; j--)
1004 : {
1005 28 : if (topoProcs[j] != NULL && beforeConstraints[j] == 0)
1006 14 : break;
1007 : }
1008 :
1009 : /* If no available candidate, topological sort fails */
1010 14 : if (j < 0)
1011 0 : return false;
1012 :
1013 : /*
1014 : * Output everything in the lock group. There's no point in
1015 : * outputting an ordering where members of the same lock group are not
1016 : * consecutive on the wait queue: if some other waiter is between two
1017 : * requests that belong to the same group, then either it conflicts
1018 : * with both of them and is certainly not a solution; or it conflicts
1019 : * with at most one of them and is thus isomorphic to an ordering
1020 : * where the group members are consecutive.
1021 : */
1022 14 : proc = topoProcs[j];
1023 14 : if (proc->lockGroupLeader != NULL)
1024 4 : proc = proc->lockGroupLeader;
1025 : Assert(proc != NULL);
1026 56 : for (c = 0; c <= last; ++c)
1027 : {
1028 42 : if (topoProcs[c] == proc || (topoProcs[c] != NULL &&
1029 22 : topoProcs[c]->lockGroupLeader == proc))
1030 : {
1031 18 : ordering[i - nmatches] = topoProcs[c];
1032 18 : topoProcs[c] = NULL;
1033 18 : ++nmatches;
1034 : }
1035 : }
1036 : Assert(nmatches > 0);
1037 14 : i -= nmatches;
1038 :
1039 : /* Update beforeConstraints counts of its predecessors */
1040 20 : for (k = afterConstraints[j]; k > 0; k = constraints[k - 1].link)
1041 6 : beforeConstraints[constraints[k - 1].pred]--;
1042 : }
1043 :
1044 : /* Done */
1045 6 : return true;
1046 : }
1047 :
1048 : #ifdef DEBUG_DEADLOCK
1049 : static void
1050 : PrintLockQueue(LOCK *lock, const char *info)
1051 : {
1052 : dclist_head *waitQueue = &lock->waitProcs;
1053 : dlist_iter proc_iter;
1054 :
1055 : printf("%s lock %p queue ", info, lock);
1056 :
1057 : dclist_foreach(proc_iter, waitQueue)
1058 : {
1059 : PGPROC *proc = dlist_container(PGPROC, links, proc_iter.cur);
1060 :
1061 : printf(" %d", proc->pid);
1062 : }
1063 : printf("\n");
1064 : fflush(stdout);
1065 : }
1066 : #endif
1067 :
1068 : /*
1069 : * Report a detected deadlock, with available details.
1070 : */
1071 : void
1072 10 : DeadLockReport(void)
1073 : {
1074 : StringInfoData clientbuf; /* errdetail for client */
1075 : StringInfoData logbuf; /* errdetail for server log */
1076 : StringInfoData locktagbuf;
1077 : int i;
1078 :
1079 10 : initStringInfo(&clientbuf);
1080 10 : initStringInfo(&logbuf);
1081 10 : initStringInfo(&locktagbuf);
1082 :
1083 : /* Generate the "waits for" lines sent to the client */
1084 44 : for (i = 0; i < nDeadlockDetails; i++)
1085 : {
1086 34 : DEADLOCK_INFO *info = &deadlockDetails[i];
1087 : int nextpid;
1088 :
1089 : /* The last proc waits for the first one... */
1090 34 : if (i < nDeadlockDetails - 1)
1091 24 : nextpid = info[1].pid;
1092 : else
1093 10 : nextpid = deadlockDetails[0].pid;
1094 :
1095 : /* reset locktagbuf to hold next object description */
1096 34 : resetStringInfo(&locktagbuf);
1097 :
1098 34 : DescribeLockTag(&locktagbuf, &info->locktag);
1099 :
1100 34 : if (i > 0)
1101 24 : appendStringInfoChar(&clientbuf, '\n');
1102 :
1103 68 : appendStringInfo(&clientbuf,
1104 34 : _("Process %d waits for %s on %s; blocked by process %d."),
1105 : info->pid,
1106 34 : GetLockmodeName(info->locktag.locktag_lockmethodid,
1107 : info->lockmode),
1108 : locktagbuf.data,
1109 : nextpid);
1110 : }
1111 :
1112 : /* Duplicate all the above for the server ... */
1113 10 : appendBinaryStringInfo(&logbuf, clientbuf.data, clientbuf.len);
1114 :
1115 : /* ... and add info about query strings */
1116 44 : for (i = 0; i < nDeadlockDetails; i++)
1117 : {
1118 34 : DEADLOCK_INFO *info = &deadlockDetails[i];
1119 :
1120 34 : appendStringInfoChar(&logbuf, '\n');
1121 :
1122 34 : appendStringInfo(&logbuf,
1123 34 : _("Process %d: %s"),
1124 : info->pid,
1125 : pgstat_get_backend_current_activity(info->pid, false));
1126 : }
1127 :
1128 10 : pgstat_report_deadlock();
1129 :
1130 10 : ereport(ERROR,
1131 : (errcode(ERRCODE_T_R_DEADLOCK_DETECTED),
1132 : errmsg("deadlock detected"),
1133 : errdetail_internal("%s", clientbuf.data),
1134 : errdetail_log("%s", logbuf.data),
1135 : errhint("See server log for query details.")));
1136 : }
1137 :
1138 : /*
1139 : * RememberSimpleDeadLock: set up info for DeadLockReport when ProcSleep
1140 : * detects a trivial (two-way) deadlock. proc1 wants to block for lockmode
1141 : * on lock, but proc2 is already waiting and would be blocked by proc1.
1142 : */
1143 : void
1144 2 : RememberSimpleDeadLock(PGPROC *proc1,
1145 : LOCKMODE lockmode,
1146 : LOCK *lock,
1147 : PGPROC *proc2)
1148 : {
1149 2 : DEADLOCK_INFO *info = &deadlockDetails[0];
1150 :
1151 2 : info->locktag = lock->tag;
1152 2 : info->lockmode = lockmode;
1153 2 : info->pid = proc1->pid;
1154 2 : info++;
1155 2 : info->locktag = proc2->waitLock->tag;
1156 2 : info->lockmode = proc2->waitLockMode;
1157 2 : info->pid = proc2->pid;
1158 2 : nDeadlockDetails = 2;
1159 2 : }
|
