* parallel.c
* Infrastructure for launching parallel workers
*
- * Portions Copyright (c) 1996-2016, PostgreSQL Global Development Group
+ * Portions Copyright (c) 1996-2019, PostgreSQL Global Development Group
* Portions Copyright (c) 1994, Regents of the University of California
*
* IDENTIFICATION
#include "postgres.h"
+#include "access/nbtree.h"
#include "access/parallel.h"
+#include "access/session.h"
#include "access/xact.h"
#include "access/xlog.h"
+#include "catalog/pg_enum.h"
+#include "catalog/index.h"
#include "catalog/namespace.h"
#include "commands/async.h"
+#include "executor/execParallel.h"
#include "libpq/libpq.h"
#include "libpq/pqformat.h"
#include "libpq/pqmq.h"
#include "miscadmin.h"
-#include "optimizer/planmain.h"
+#include "optimizer/optimizer.h"
#include "pgstat.h"
#include "storage/ipc.h"
+#include "storage/predicate.h"
#include "storage/sinval.h"
#include "storage/spin.h"
#include "tcop/tcopprot.h"
#include "utils/guc.h"
#include "utils/inval.h"
#include "utils/memutils.h"
-#include "utils/resowner.h"
+#include "utils/relmapper.h"
#include "utils/snapmgr.h"
+#include "utils/typcache.h"
/*
#define PARALLEL_MAGIC 0x50477c7c
/*
- * Magic numbers for parallel state sharing. Higher-level code should use
- * smaller values, leaving these very large ones for use by this module.
+ * Magic numbers for per-context parallel state sharing. Higher-level code
+ * should use smaller values, leaving these very large ones for use by this
+ * module.
*/
#define PARALLEL_KEY_FIXED UINT64CONST(0xFFFFFFFFFFFF0001)
#define PARALLEL_KEY_ERROR_QUEUE UINT64CONST(0xFFFFFFFFFFFF0002)
#define PARALLEL_KEY_TRANSACTION_SNAPSHOT UINT64CONST(0xFFFFFFFFFFFF0006)
#define PARALLEL_KEY_ACTIVE_SNAPSHOT UINT64CONST(0xFFFFFFFFFFFF0007)
#define PARALLEL_KEY_TRANSACTION_STATE UINT64CONST(0xFFFFFFFFFFFF0008)
-#define PARALLEL_KEY_EXTENSION_TRAMPOLINE UINT64CONST(0xFFFFFFFFFFFF0009)
+#define PARALLEL_KEY_ENTRYPOINT UINT64CONST(0xFFFFFFFFFFFF0009)
+#define PARALLEL_KEY_SESSION_DSM UINT64CONST(0xFFFFFFFFFFFF000A)
+#define PARALLEL_KEY_REINDEX_STATE UINT64CONST(0xFFFFFFFFFFFF000B)
+#define PARALLEL_KEY_RELMAPPER_STATE UINT64CONST(0xFFFFFFFFFFFF000C)
+#define PARALLEL_KEY_ENUMBLACKLIST UINT64CONST(0xFFFFFFFFFFFF000D)
/* Fixed-size parallel state. */
typedef struct FixedParallelState
Oid database_id;
Oid authenticated_user_id;
Oid current_user_id;
+ Oid outer_user_id;
Oid temp_namespace_id;
Oid temp_toast_namespace_id;
int sec_context;
+ bool is_superuser;
PGPROC *parallel_master_pgproc;
pid_t parallel_master_pid;
BackendId parallel_master_backend_id;
-
- /* Entrypoint for parallel workers. */
- parallel_worker_main_type entrypoint;
+ TimestampTz xact_ts;
+ TimestampTz stmt_ts;
+ SerializableXactHandle serializable_xact_handle;
/* Mutex protects remaining fields. */
slock_t mutex;
/* List of active parallel contexts. */
static dlist_head pcxt_list = DLIST_STATIC_INIT(pcxt_list);
+/* Backend-local copy of data from FixedParallelState. */
+static pid_t ParallelMasterPid;
+
+/*
+ * List of internal parallel worker entry points. We need this for
+ * reasons explained in LookupParallelWorkerFunction(), below.
+ */
+static const struct
+{
+ const char *fn_name;
+ parallel_worker_main_type fn_addr;
+} InternalParallelWorkers[] =
+
+{
+ {
+ "ParallelQueryMain", ParallelQueryMain
+ },
+ {
+ "_bt_parallel_build_main", _bt_parallel_build_main
+ }
+};
+
/* Private functions. */
static void HandleParallelMessage(ParallelContext *pcxt, int i, StringInfo msg);
-static void ParallelExtensionTrampoline(dsm_segment *seg, shm_toc *toc);
-static void ParallelWorkerMain(Datum main_arg);
static void WaitForParallelWorkersToExit(ParallelContext *pcxt);
+static parallel_worker_main_type LookupParallelWorkerFunction(const char *libraryname, const char *funcname);
+static void ParallelWorkerShutdown(int code, Datum arg);
/*
* destroyed before exiting the current subtransaction.
*/
ParallelContext *
-CreateParallelContext(parallel_worker_main_type entrypoint, int nworkers)
+CreateParallelContext(const char *library_name, const char *function_name,
+ int nworkers)
{
MemoryContext oldcontext;
ParallelContext *pcxt;
/* Number of workers should be non-negative. */
Assert(nworkers >= 0);
- /*
- * If dynamic shared memory is not available, we won't be able to use
- * background workers.
- */
- if (dynamic_shared_memory_type == DSM_IMPL_NONE)
- nworkers = 0;
-
- /*
- * If we are running under serializable isolation, we can't use parallel
- * workers, at least not until somebody enhances that mechanism to be
- * parallel-aware.
- */
- if (IsolationIsSerializable())
- nworkers = 0;
-
/* We might be running in a short-lived memory context. */
oldcontext = MemoryContextSwitchTo(TopTransactionContext);
pcxt = palloc0(sizeof(ParallelContext));
pcxt->subid = GetCurrentSubTransactionId();
pcxt->nworkers = nworkers;
- pcxt->entrypoint = entrypoint;
+ pcxt->library_name = pstrdup(library_name);
+ pcxt->function_name = pstrdup(function_name);
pcxt->error_context_stack = error_context_stack;
shm_toc_initialize_estimator(&pcxt->estimator);
dlist_push_head(&pcxt_list, &pcxt->node);
return pcxt;
}
-/*
- * Establish a new parallel context that calls a function provided by an
- * extension. This works around the fact that the library might get mapped
- * at a different address in each backend.
- */
-ParallelContext *
-CreateParallelContextForExternalFunction(char *library_name,
- char *function_name,
- int nworkers)
-{
- MemoryContext oldcontext;
- ParallelContext *pcxt;
-
- /* We might be running in a very short-lived memory context. */
- oldcontext = MemoryContextSwitchTo(TopTransactionContext);
-
- /* Create the context. */
- pcxt = CreateParallelContext(ParallelExtensionTrampoline, nworkers);
- pcxt->library_name = pstrdup(library_name);
- pcxt->function_name = pstrdup(function_name);
-
- /* Restore previous memory context. */
- MemoryContextSwitchTo(oldcontext);
-
- return pcxt;
-}
-
/*
* Establish the dynamic shared memory segment for a parallel context and
* copy state and other bookkeeping information that will be needed by
Size tsnaplen = 0;
Size asnaplen = 0;
Size tstatelen = 0;
+ Size reindexlen = 0;
+ Size relmapperlen = 0;
+ Size enumblacklistlen = 0;
Size segsize = 0;
int i;
FixedParallelState *fps;
+ dsm_handle session_dsm_handle = DSM_HANDLE_INVALID;
Snapshot transaction_snapshot = GetTransactionSnapshot();
Snapshot active_snapshot = GetActiveSnapshot();
* Normally, the user will have requested at least one worker process, but
* if by chance they have not, we can skip a bunch of things here.
*/
+ if (pcxt->nworkers > 0)
+ {
+ /* Get (or create) the per-session DSM segment's handle. */
+ session_dsm_handle = GetSessionDsmHandle();
+
+ /*
+ * If we weren't able to create a per-session DSM segment, then we can
+ * continue but we can't safely launch any workers because their
+ * record typmods would be incompatible so they couldn't exchange
+ * tuples.
+ */
+ if (session_dsm_handle == DSM_HANDLE_INVALID)
+ pcxt->nworkers = 0;
+ }
+
if (pcxt->nworkers > 0)
{
/* Estimate space for various kinds of state sharing. */
shm_toc_estimate_chunk(&pcxt->estimator, asnaplen);
tstatelen = EstimateTransactionStateSpace();
shm_toc_estimate_chunk(&pcxt->estimator, tstatelen);
+ shm_toc_estimate_chunk(&pcxt->estimator, sizeof(dsm_handle));
+ reindexlen = EstimateReindexStateSpace();
+ shm_toc_estimate_chunk(&pcxt->estimator, reindexlen);
+ relmapperlen = EstimateRelationMapSpace();
+ shm_toc_estimate_chunk(&pcxt->estimator, relmapperlen);
+ enumblacklistlen = EstimateEnumBlacklistSpace();
+ shm_toc_estimate_chunk(&pcxt->estimator, enumblacklistlen);
/* If you add more chunks here, you probably need to add keys. */
- shm_toc_estimate_keys(&pcxt->estimator, 6);
+ shm_toc_estimate_keys(&pcxt->estimator, 10);
/* Estimate space need for error queues. */
StaticAssertStmt(BUFFERALIGN(PARALLEL_ERROR_QUEUE_SIZE) ==
pcxt->nworkers));
shm_toc_estimate_keys(&pcxt->estimator, 1);
- /* Estimate how much we'll need for extension entrypoint info. */
- if (pcxt->library_name != NULL)
- {
- Assert(pcxt->entrypoint == ParallelExtensionTrampoline);
- Assert(pcxt->function_name != NULL);
- shm_toc_estimate_chunk(&pcxt->estimator, strlen(pcxt->library_name)
- + strlen(pcxt->function_name) + 2);
- shm_toc_estimate_keys(&pcxt->estimator, 1);
- }
+ /* Estimate how much we'll need for the entrypoint info. */
+ shm_toc_estimate_chunk(&pcxt->estimator, strlen(pcxt->library_name) +
+ strlen(pcxt->function_name) + 2);
+ shm_toc_estimate_keys(&pcxt->estimator, 1);
}
/*
shm_toc_allocate(pcxt->toc, sizeof(FixedParallelState));
fps->database_id = MyDatabaseId;
fps->authenticated_user_id = GetAuthenticatedUserId();
+ fps->outer_user_id = GetCurrentRoleId();
+ fps->is_superuser = session_auth_is_superuser;
GetUserIdAndSecContext(&fps->current_user_id, &fps->sec_context);
GetTempNamespaceState(&fps->temp_namespace_id,
&fps->temp_toast_namespace_id);
fps->parallel_master_pgproc = MyProc;
fps->parallel_master_pid = MyProcPid;
fps->parallel_master_backend_id = MyBackendId;
- fps->entrypoint = pcxt->entrypoint;
+ fps->xact_ts = GetCurrentTransactionStartTimestamp();
+ fps->stmt_ts = GetCurrentStatementStartTimestamp();
+ fps->serializable_xact_handle = ShareSerializableXact();
SpinLockInit(&fps->mutex);
fps->last_xlog_end = 0;
shm_toc_insert(pcxt->toc, PARALLEL_KEY_FIXED, fps);
char *tsnapspace;
char *asnapspace;
char *tstatespace;
+ char *reindexspace;
+ char *relmapperspace;
char *error_queue_space;
+ char *session_dsm_handle_space;
+ char *entrypointstate;
+ char *enumblacklistspace;
+ Size lnamelen;
/* Serialize shared libraries we have loaded. */
libraryspace = shm_toc_allocate(pcxt->toc, library_len);
SerializeSnapshot(active_snapshot, asnapspace);
shm_toc_insert(pcxt->toc, PARALLEL_KEY_ACTIVE_SNAPSHOT, asnapspace);
+ /* Provide the handle for per-session segment. */
+ session_dsm_handle_space = shm_toc_allocate(pcxt->toc,
+ sizeof(dsm_handle));
+ *(dsm_handle *) session_dsm_handle_space = session_dsm_handle;
+ shm_toc_insert(pcxt->toc, PARALLEL_KEY_SESSION_DSM,
+ session_dsm_handle_space);
+
/* Serialize transaction state. */
tstatespace = shm_toc_allocate(pcxt->toc, tstatelen);
SerializeTransactionState(tstatelen, tstatespace);
shm_toc_insert(pcxt->toc, PARALLEL_KEY_TRANSACTION_STATE, tstatespace);
+ /* Serialize reindex state. */
+ reindexspace = shm_toc_allocate(pcxt->toc, reindexlen);
+ SerializeReindexState(reindexlen, reindexspace);
+ shm_toc_insert(pcxt->toc, PARALLEL_KEY_REINDEX_STATE, reindexspace);
+
+ /* Serialize relmapper state. */
+ relmapperspace = shm_toc_allocate(pcxt->toc, relmapperlen);
+ SerializeRelationMap(relmapperlen, relmapperspace);
+ shm_toc_insert(pcxt->toc, PARALLEL_KEY_RELMAPPER_STATE,
+ relmapperspace);
+
+ /* Serialize enum blacklist state. */
+ enumblacklistspace = shm_toc_allocate(pcxt->toc, enumblacklistlen);
+ SerializeEnumBlacklist(enumblacklistspace, enumblacklistlen);
+ shm_toc_insert(pcxt->toc, PARALLEL_KEY_ENUMBLACKLIST,
+ enumblacklistspace);
+
/* Allocate space for worker information. */
pcxt->worker = palloc0(sizeof(ParallelWorkerInfo) * pcxt->nworkers);
}
shm_toc_insert(pcxt->toc, PARALLEL_KEY_ERROR_QUEUE, error_queue_space);
- /* Serialize extension entrypoint information. */
- if (pcxt->library_name != NULL)
- {
- Size lnamelen = strlen(pcxt->library_name);
- char *extensionstate;
-
- extensionstate = shm_toc_allocate(pcxt->toc, lnamelen
- + strlen(pcxt->function_name) + 2);
- strcpy(extensionstate, pcxt->library_name);
- strcpy(extensionstate + lnamelen + 1, pcxt->function_name);
- shm_toc_insert(pcxt->toc, PARALLEL_KEY_EXTENSION_TRAMPOLINE,
- extensionstate);
- }
+ /*
+ * Serialize entrypoint information. It's unsafe to pass function
+ * pointers across processes, as the function pointer may be different
+ * in each process in EXEC_BACKEND builds, so we always pass library
+ * and function name. (We use library name "postgres" for functions
+ * in the core backend.)
+ */
+ lnamelen = strlen(pcxt->library_name);
+ entrypointstate = shm_toc_allocate(pcxt->toc, lnamelen +
+ strlen(pcxt->function_name) + 2);
+ strcpy(entrypointstate, pcxt->library_name);
+ strcpy(entrypointstate + lnamelen + 1, pcxt->function_name);
+ shm_toc_insert(pcxt->toc, PARALLEL_KEY_ENTRYPOINT, entrypointstate);
}
/* Restore previous memory context. */
ReinitializeParallelDSM(ParallelContext *pcxt)
{
FixedParallelState *fps;
- char *error_queue_space;
- int i;
/* Wait for any old workers to exit. */
if (pcxt->nworkers_launched > 0)
WaitForParallelWorkersToFinish(pcxt);
WaitForParallelWorkersToExit(pcxt);
pcxt->nworkers_launched = 0;
+ if (pcxt->known_attached_workers)
+ {
+ pfree(pcxt->known_attached_workers);
+ pcxt->known_attached_workers = NULL;
+ pcxt->nknown_attached_workers = 0;
+ }
}
/* Reset a few bits of fixed parallel state to a clean state. */
- fps = shm_toc_lookup(pcxt->toc, PARALLEL_KEY_FIXED);
+ fps = shm_toc_lookup(pcxt->toc, PARALLEL_KEY_FIXED, false);
fps->last_xlog_end = 0;
- /* Recreate error queues. */
- error_queue_space =
- shm_toc_lookup(pcxt->toc, PARALLEL_KEY_ERROR_QUEUE);
- for (i = 0; i < pcxt->nworkers; ++i)
+ /* Recreate error queues (if they exist). */
+ if (pcxt->nworkers > 0)
{
- char *start;
- shm_mq *mq;
+ char *error_queue_space;
+ int i;
- start = error_queue_space + i * PARALLEL_ERROR_QUEUE_SIZE;
- mq = shm_mq_create(start, PARALLEL_ERROR_QUEUE_SIZE);
- shm_mq_set_receiver(mq, MyProc);
- pcxt->worker[i].error_mqh = shm_mq_attach(mq, pcxt->seg, NULL);
+ error_queue_space =
+ shm_toc_lookup(pcxt->toc, PARALLEL_KEY_ERROR_QUEUE, false);
+ for (i = 0; i < pcxt->nworkers; ++i)
+ {
+ char *start;
+ shm_mq *mq;
+
+ start = error_queue_space + i * PARALLEL_ERROR_QUEUE_SIZE;
+ mq = shm_mq_create(start, PARALLEL_ERROR_QUEUE_SIZE);
+ shm_mq_set_receiver(mq, MyProc);
+ pcxt->worker[i].error_mqh = shm_mq_attach(mq, pcxt->seg, NULL);
+ }
}
}
oldcontext = MemoryContextSwitchTo(TopTransactionContext);
/* Configure a worker. */
+ memset(&worker, 0, sizeof(worker));
snprintf(worker.bgw_name, BGW_MAXLEN, "parallel worker for PID %d",
MyProcPid);
+ snprintf(worker.bgw_type, BGW_MAXLEN, "parallel worker");
worker.bgw_flags =
- BGWORKER_SHMEM_ACCESS | BGWORKER_BACKEND_DATABASE_CONNECTION;
+ BGWORKER_SHMEM_ACCESS | BGWORKER_BACKEND_DATABASE_CONNECTION
+ | BGWORKER_CLASS_PARALLEL;
worker.bgw_start_time = BgWorkerStart_ConsistentState;
worker.bgw_restart_time = BGW_NEVER_RESTART;
- worker.bgw_main = ParallelWorkerMain;
+ sprintf(worker.bgw_library_name, "postgres");
+ sprintf(worker.bgw_function_name, "ParallelWorkerMain");
worker.bgw_main_arg = UInt32GetDatum(dsm_segment_handle(pcxt->seg));
worker.bgw_notify_pid = MyProcPid;
- memset(&worker.bgw_extra, 0, BGW_EXTRALEN);
/*
* Start workers.
*/
any_registrations_failed = true;
pcxt->worker[i].bgwhandle = NULL;
- pfree(pcxt->worker[i].error_mqh);
+ shm_mq_detach(pcxt->worker[i].error_mqh);
pcxt->worker[i].error_mqh = NULL;
}
}
+ /*
+ * Now that nworkers_launched has taken its final value, we can initialize
+ * known_attached_workers.
+ */
+ if (pcxt->nworkers_launched > 0)
+ {
+ pcxt->known_attached_workers =
+ palloc0(sizeof(bool) * pcxt->nworkers_launched);
+ pcxt->nknown_attached_workers = 0;
+ }
+
/* Restore previous memory context. */
MemoryContextSwitchTo(oldcontext);
}
+/*
+ * Wait for all workers to attach to their error queues, and throw an error if
+ * any worker fails to do this.
+ *
+ * Callers can assume that if this function returns successfully, then the
+ * number of workers given by pcxt->nworkers_launched have initialized and
+ * attached to their error queues. Whether or not these workers are guaranteed
+ * to still be running depends on what code the caller asked them to run;
+ * this function does not guarantee that they have not exited. However, it
+ * does guarantee that any workers which exited must have done so cleanly and
+ * after successfully performing the work with which they were tasked.
+ *
+ * If this function is not called, then some of the workers that were launched
+ * may not have been started due to a fork() failure, or may have exited during
+ * early startup prior to attaching to the error queue, so nworkers_launched
+ * cannot be viewed as completely reliable. It will never be less than the
+ * number of workers which actually started, but it might be more. Any workers
+ * that failed to start will still be discovered by
+ * WaitForParallelWorkersToFinish and an error will be thrown at that time,
+ * provided that function is eventually reached.
+ *
+ * In general, the leader process should do as much work as possible before
+ * calling this function. fork() failures and other early-startup failures
+ * are very uncommon, and having the leader sit idle when it could be doing
+ * useful work is undesirable. However, if the leader needs to wait for
+ * all of its workers or for a specific worker, it may want to call this
+ * function before doing so. If not, it must make some other provision for
+ * the failure-to-start case, lest it wait forever. On the other hand, a
+ * leader which never waits for a worker that might not be started yet, or
+ * at least never does so prior to WaitForParallelWorkersToFinish(), need not
+ * call this function at all.
+ */
+void
+WaitForParallelWorkersToAttach(ParallelContext *pcxt)
+{
+ int i;
+
+ /* Skip this if we have no launched workers. */
+ if (pcxt->nworkers_launched == 0)
+ return;
+
+ for (;;)
+ {
+ /*
+ * This will process any parallel messages that are pending and it may
+ * also throw an error propagated from a worker.
+ */
+ CHECK_FOR_INTERRUPTS();
+
+ for (i = 0; i < pcxt->nworkers_launched; ++i)
+ {
+ BgwHandleStatus status;
+ shm_mq *mq;
+ int rc;
+ pid_t pid;
+
+ if (pcxt->known_attached_workers[i])
+ continue;
+
+ /*
+ * If error_mqh is NULL, then the worker has already exited
+ * cleanly.
+ */
+ if (pcxt->worker[i].error_mqh == NULL)
+ {
+ pcxt->known_attached_workers[i] = true;
+ ++pcxt->nknown_attached_workers;
+ continue;
+ }
+
+ status = GetBackgroundWorkerPid(pcxt->worker[i].bgwhandle, &pid);
+ if (status == BGWH_STARTED)
+ {
+ /* Has the worker attached to the error queue? */
+ mq = shm_mq_get_queue(pcxt->worker[i].error_mqh);
+ if (shm_mq_get_sender(mq) != NULL)
+ {
+ /* Yes, so it is known to be attached. */
+ pcxt->known_attached_workers[i] = true;
+ ++pcxt->nknown_attached_workers;
+ }
+ }
+ else if (status == BGWH_STOPPED)
+ {
+ /*
+ * If the worker stopped without attaching to the error queue,
+ * throw an error.
+ */
+ mq = shm_mq_get_queue(pcxt->worker[i].error_mqh);
+ if (shm_mq_get_sender(mq) == NULL)
+ ereport(ERROR,
+ (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
+ errmsg("parallel worker failed to initialize"),
+ errhint("More details may be available in the server log.")));
+
+ pcxt->known_attached_workers[i] = true;
+ ++pcxt->nknown_attached_workers;
+ }
+ else
+ {
+ /*
+ * Worker not yet started, so we must wait. The postmaster
+ * will notify us if the worker's state changes. Our latch
+ * might also get set for some other reason, but if so we'll
+ * just end up waiting for the same worker again.
+ */
+ rc = WaitLatch(MyLatch,
+ WL_LATCH_SET | WL_EXIT_ON_PM_DEATH,
+ -1, WAIT_EVENT_BGWORKER_STARTUP);
+
+ if (rc & WL_LATCH_SET)
+ ResetLatch(MyLatch);
+ }
+ }
+
+ /* If all workers are known to have started, we're done. */
+ if (pcxt->nknown_attached_workers >= pcxt->nworkers_launched)
+ {
+ Assert(pcxt->nknown_attached_workers == pcxt->nworkers_launched);
+ break;
+ }
+ }
+}
+
/*
* Wait for all workers to finish computing.
*
for (;;)
{
bool anyone_alive = false;
+ int nfinished = 0;
int i;
/*
for (i = 0; i < pcxt->nworkers_launched; ++i)
{
- if (pcxt->worker[i].error_mqh != NULL)
+ /*
+ * If error_mqh is NULL, then the worker has already exited
+ * cleanly. If we have received a message through error_mqh from
+ * the worker, we know it started up cleanly, and therefore we're
+ * certain to be notified when it exits.
+ */
+ if (pcxt->worker[i].error_mqh == NULL)
+ ++nfinished;
+ else if (pcxt->known_attached_workers[i])
{
anyone_alive = true;
break;
}
if (!anyone_alive)
- break;
+ {
+ /* If all workers are known to have finished, we're done. */
+ if (nfinished >= pcxt->nworkers_launched)
+ {
+ Assert(nfinished == pcxt->nworkers_launched);
+ break;
+ }
+
+ /*
+ * We didn't detect any living workers, but not all workers are
+ * known to have exited cleanly. Either not all workers have
+ * launched yet, or maybe some of them failed to start or
+ * terminated abnormally.
+ */
+ for (i = 0; i < pcxt->nworkers_launched; ++i)
+ {
+ pid_t pid;
+ shm_mq *mq;
+
+ /*
+ * If the worker is BGWH_NOT_YET_STARTED or BGWH_STARTED, we
+ * should just keep waiting. If it is BGWH_STOPPED, then
+ * further investigation is needed.
+ */
+ if (pcxt->worker[i].error_mqh == NULL ||
+ pcxt->worker[i].bgwhandle == NULL ||
+ GetBackgroundWorkerPid(pcxt->worker[i].bgwhandle,
+ &pid) != BGWH_STOPPED)
+ continue;
+
+ /*
+ * Check whether the worker ended up stopped without ever
+ * attaching to the error queue. If so, the postmaster was
+ * unable to fork the worker or it exited without initializing
+ * properly. We must throw an error, since the caller may
+ * have been expecting the worker to do some work before
+ * exiting.
+ */
+ mq = shm_mq_get_queue(pcxt->worker[i].error_mqh);
+ if (shm_mq_get_sender(mq) == NULL)
+ ereport(ERROR,
+ (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
+ errmsg("parallel worker failed to initialize"),
+ errhint("More details may be available in the server log.")));
- WaitLatch(&MyProc->procLatch, WL_LATCH_SET, -1,
- WAIT_EVENT_PARALLEL_FINISH);
- ResetLatch(&MyProc->procLatch);
+ /*
+ * The worker is stopped, but is attached to the error queue.
+ * Unless there's a bug somewhere, this will only happen when
+ * the worker writes messages and terminates after the
+ * CHECK_FOR_INTERRUPTS() near the top of this function and
+ * before the call to GetBackgroundWorkerPid(). In that case,
+ * or latch should have been set as well and the right things
+ * will happen on the next pass through the loop.
+ */
+ }
+ }
+
+ (void) WaitLatch(MyLatch, WL_LATCH_SET | WL_EXIT_ON_PM_DEATH, -1,
+ WAIT_EVENT_PARALLEL_FINISH);
+ ResetLatch(MyLatch);
}
if (pcxt->toc != NULL)
{
FixedParallelState *fps;
- fps = shm_toc_lookup(pcxt->toc, PARALLEL_KEY_FIXED);
+ fps = shm_toc_lookup(pcxt->toc, PARALLEL_KEY_FIXED, false);
if (fps->last_xlog_end > XactLastRecEnd)
XactLastRecEnd = fps->last_xlog_end;
}
if (status == BGWH_POSTMASTER_DIED)
ereport(FATAL,
(errcode(ERRCODE_ADMIN_SHUTDOWN),
- errmsg("postmaster exited during a parallel transaction")));
+ errmsg("postmaster exited during a parallel transaction")));
/* Release memory. */
pfree(pcxt->worker[i].bgwhandle);
{
TerminateBackgroundWorker(pcxt->worker[i].bgwhandle);
- pfree(pcxt->worker[i].error_mqh);
+ shm_mq_detach(pcxt->worker[i].error_mqh);
pcxt->worker[i].error_mqh = NULL;
}
}
}
/* Free memory. */
+ pfree(pcxt->library_name);
+ pfree(pcxt->function_name);
pfree(pcxt);
}
}
else
ereport(ERROR,
- (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
- errmsg("lost connection to parallel worker")));
+ (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
+ errmsg("lost connection to parallel worker")));
}
}
}
{
char msgtype;
+ if (pcxt->known_attached_workers != NULL &&
+ !pcxt->known_attached_workers[i])
+ {
+ pcxt->known_attached_workers[i] = true;
+ pcxt->nknown_attached_workers++;
+ }
+
msgtype = pq_getmsgbyte(msg);
switch (msgtype)
case 'X': /* Terminate, indicating clean exit */
{
- pfree(pcxt->worker[i].error_mqh);
+ shm_mq_detach(pcxt->worker[i].error_mqh);
pcxt->worker[i].error_mqh = NULL;
break;
}
/*
* Main entrypoint for parallel workers.
*/
-static void
+void
ParallelWorkerMain(Datum main_arg)
{
dsm_segment *seg;
shm_mq *mq;
shm_mq_handle *mqh;
char *libraryspace;
+ char *entrypointstate;
+ char *library_name;
+ char *function_name;
+ parallel_worker_main_type entrypt;
char *gucspace;
char *combocidspace;
char *tsnapspace;
char *asnapspace;
char *tstatespace;
+ char *reindexspace;
+ char *relmapperspace;
+ char *enumblacklistspace;
StringInfoData msgbuf;
+ char *session_dsm_handle_space;
/* Set flag to indicate that we're initializing a parallel worker. */
InitializingParallelWorker = true;
Assert(ParallelWorkerNumber == -1);
memcpy(&ParallelWorkerNumber, MyBgworkerEntry->bgw_extra, sizeof(int));
- /* Set up a memory context and resource owner. */
- Assert(CurrentResourceOwner == NULL);
- CurrentResourceOwner = ResourceOwnerCreate(NULL, "parallel toplevel");
+ /* Set up a memory context to work in, just for cleanliness. */
CurrentMemoryContext = AllocSetContextCreate(TopMemoryContext,
"Parallel worker",
ALLOCSET_DEFAULT_SIZES);
/*
- * Now that we have a resource owner, we can attach to the dynamic shared
- * memory segment and read the table of contents.
+ * Attach to the dynamic shared memory segment for the parallel query, and
+ * find its table of contents.
+ *
+ * Note: at this point, we have not created any ResourceOwner in this
+ * process. This will result in our DSM mapping surviving until process
+ * exit, which is fine. If there were a ResourceOwner, it would acquire
+ * ownership of the mapping, but we have no need for that.
*/
seg = dsm_attach(DatumGetUInt32(main_arg));
if (seg == NULL)
if (toc == NULL)
ereport(ERROR,
(errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
- errmsg("invalid magic number in dynamic shared memory segment")));
+ errmsg("invalid magic number in dynamic shared memory segment")));
/* Look up fixed parallel state. */
- fps = shm_toc_lookup(toc, PARALLEL_KEY_FIXED);
- Assert(fps != NULL);
+ fps = shm_toc_lookup(toc, PARALLEL_KEY_FIXED, false);
MyFixedParallelState = fps;
+ /* Arrange to signal the leader if we exit. */
+ ParallelMasterPid = fps->parallel_master_pid;
+ ParallelMasterBackendId = fps->parallel_master_backend_id;
+ on_shmem_exit(ParallelWorkerShutdown, (Datum) 0);
+
/*
- * Now that we have a worker number, we can find and attach to the error
- * queue provided for us. That's good, because until we do that, any
- * errors that happen here will not be reported back to the process that
- * requested that this worker be launched.
+ * Now we can find and attach to the error queue provided for us. That's
+ * good, because until we do that, any errors that happen here will not be
+ * reported back to the process that requested that this worker be
+ * launched.
*/
- error_queue_space = shm_toc_lookup(toc, PARALLEL_KEY_ERROR_QUEUE);
+ error_queue_space = shm_toc_lookup(toc, PARALLEL_KEY_ERROR_QUEUE, false);
mq = (shm_mq *) (error_queue_space +
ParallelWorkerNumber * PARALLEL_ERROR_QUEUE_SIZE);
shm_mq_set_sender(mq, MyProc);
* in this case.
*/
pq_beginmessage(&msgbuf, 'K');
- pq_sendint(&msgbuf, (int32) MyProcPid, sizeof(int32));
- pq_sendint(&msgbuf, (int32) MyCancelKey, sizeof(int32));
+ pq_sendint32(&msgbuf, (int32) MyProcPid);
+ pq_sendint32(&msgbuf, (int32) MyCancelKey);
pq_endmessage(&msgbuf);
/*
return;
/*
- * Load libraries that were loaded by original backend. We want to do
- * this before restoring GUCs, because the libraries might define custom
- * variables.
+ * Restore transaction and statement start-time timestamps. This must
+ * happen before anything that would start a transaction, else asserts in
+ * xact.c will fire.
*/
- libraryspace = shm_toc_lookup(toc, PARALLEL_KEY_LIBRARY);
- Assert(libraryspace != NULL);
- RestoreLibraryState(libraryspace);
+ SetParallelStartTimestamps(fps->xact_ts, fps->stmt_ts);
+
+ /*
+ * Identify the entry point to be called. In theory this could result in
+ * loading an additional library, though most likely the entry point is in
+ * the core backend or in a library we just loaded.
+ */
+ entrypointstate = shm_toc_lookup(toc, PARALLEL_KEY_ENTRYPOINT, false);
+ library_name = entrypointstate;
+ function_name = entrypointstate + strlen(library_name) + 1;
+
+ entrypt = LookupParallelWorkerFunction(library_name, function_name);
/* Restore database connection. */
BackgroundWorkerInitializeConnectionByOid(fps->database_id,
- fps->authenticated_user_id);
+ fps->authenticated_user_id,
+ 0);
/*
* Set the client encoding to the database encoding, since that is what
*/
SetClientEncoding(GetDatabaseEncoding());
- /* Restore GUC values from launching backend. */
- gucspace = shm_toc_lookup(toc, PARALLEL_KEY_GUC);
- Assert(gucspace != NULL);
+ /*
+ * Load libraries that were loaded by original backend. We want to do
+ * this before restoring GUCs, because the libraries might define custom
+ * variables.
+ */
+ libraryspace = shm_toc_lookup(toc, PARALLEL_KEY_LIBRARY, false);
StartTransactionCommand();
+ RestoreLibraryState(libraryspace);
+
+ /* Restore GUC values from launching backend. */
+ gucspace = shm_toc_lookup(toc, PARALLEL_KEY_GUC, false);
RestoreGUCState(gucspace);
CommitTransactionCommand();
/* Crank up a transaction state appropriate to a parallel worker. */
- tstatespace = shm_toc_lookup(toc, PARALLEL_KEY_TRANSACTION_STATE);
+ tstatespace = shm_toc_lookup(toc, PARALLEL_KEY_TRANSACTION_STATE, false);
StartParallelWorkerTransaction(tstatespace);
/* Restore combo CID state. */
- combocidspace = shm_toc_lookup(toc, PARALLEL_KEY_COMBO_CID);
- Assert(combocidspace != NULL);
+ combocidspace = shm_toc_lookup(toc, PARALLEL_KEY_COMBO_CID, false);
RestoreComboCIDState(combocidspace);
+ /* Attach to the per-session DSM segment and contained objects. */
+ session_dsm_handle_space =
+ shm_toc_lookup(toc, PARALLEL_KEY_SESSION_DSM, false);
+ AttachSession(*(dsm_handle *) session_dsm_handle_space);
+
/* Restore transaction snapshot. */
- tsnapspace = shm_toc_lookup(toc, PARALLEL_KEY_TRANSACTION_SNAPSHOT);
- Assert(tsnapspace != NULL);
+ tsnapspace = shm_toc_lookup(toc, PARALLEL_KEY_TRANSACTION_SNAPSHOT, false);
RestoreTransactionSnapshot(RestoreSnapshot(tsnapspace),
fps->parallel_master_pgproc);
/* Restore active snapshot. */
- asnapspace = shm_toc_lookup(toc, PARALLEL_KEY_ACTIVE_SNAPSHOT);
- Assert(asnapspace != NULL);
+ asnapspace = shm_toc_lookup(toc, PARALLEL_KEY_ACTIVE_SNAPSHOT, false);
PushActiveSnapshot(RestoreSnapshot(asnapspace));
/*
*/
InvalidateSystemCaches();
+ /*
+ * Restore current role id. Skip verifying whether session user is
+ * allowed to become this role and blindly restore the leader's state for
+ * current role.
+ */
+ SetCurrentRoleId(fps->outer_user_id, fps->is_superuser);
+
/* Restore user ID and security context. */
SetUserIdAndSecContext(fps->current_user_id, fps->sec_context);
SetTempNamespaceState(fps->temp_namespace_id,
fps->temp_toast_namespace_id);
- /* Set ParallelMasterBackendId so we know how to address temp relations. */
- ParallelMasterBackendId = fps->parallel_master_backend_id;
+ /* Restore reindex state. */
+ reindexspace = shm_toc_lookup(toc, PARALLEL_KEY_REINDEX_STATE, false);
+ RestoreReindexState(reindexspace);
+
+ /* Restore relmapper state. */
+ relmapperspace = shm_toc_lookup(toc, PARALLEL_KEY_RELMAPPER_STATE, false);
+ RestoreRelationMap(relmapperspace);
+
+ /* Restore enum blacklist. */
+ enumblacklistspace = shm_toc_lookup(toc, PARALLEL_KEY_ENUMBLACKLIST,
+ false);
+ RestoreEnumBlacklist(enumblacklistspace);
+
+ /* Attach to the leader's serializable transaction, if SERIALIZABLE. */
+ AttachSerializableXact(fps->serializable_xact_handle);
/*
* We've initialized all of our state now; nothing should change
/*
* Time to do the real work: invoke the caller-supplied code.
- *
- * If you get a crash at this line, see the comments for
- * ParallelExtensionTrampoline.
*/
- fps->entrypoint(seg, toc);
+ entrypt(seg, toc);
/* Must exit parallel mode to pop active snapshot. */
ExitParallelMode();
- /* Must pop active snapshot so resowner.c doesn't complain. */
+ /* Must pop active snapshot so snapmgr.c doesn't complain. */
PopActiveSnapshot();
/* Shut down the parallel-worker transaction. */
EndParallelWorkerTransaction();
+ /* Detach from the per-session DSM segment. */
+ DetachSession();
+
/* Report success. */
pq_putmessage('X', NULL, 0);
}
-/*
- * It's unsafe for the entrypoint invoked by ParallelWorkerMain to be a
- * function living in a dynamically loaded module, because the module might
- * not be loaded in every process, or might be loaded but not at the same
- * address. To work around that problem, CreateParallelContextForExtension()
- * arranges to call this function rather than calling the extension-provided
- * function directly; and this function then looks up the real entrypoint and
- * calls it.
- */
-static void
-ParallelExtensionTrampoline(dsm_segment *seg, shm_toc *toc)
-{
- char *extensionstate;
- char *library_name;
- char *function_name;
- parallel_worker_main_type entrypt;
-
- extensionstate = shm_toc_lookup(toc, PARALLEL_KEY_EXTENSION_TRAMPOLINE);
- Assert(extensionstate != NULL);
- library_name = extensionstate;
- function_name = extensionstate + strlen(library_name) + 1;
-
- entrypt = (parallel_worker_main_type)
- load_external_function(library_name, function_name, true, NULL);
- entrypt(seg, toc);
-}
-
/*
* Update shared memory with the ending location of the last WAL record we
* wrote, if it's greater than the value already stored there.
fps->last_xlog_end = last_xlog_end;
SpinLockRelease(&fps->mutex);
}
+
+/*
+ * Make sure the leader tries to read from our error queue one more time.
+ * This guards against the case where we exit uncleanly without sending an
+ * ErrorResponse to the leader, for example because some code calls proc_exit
+ * directly.
+ */
+static void
+ParallelWorkerShutdown(int code, Datum arg)
+{
+ SendProcSignal(ParallelMasterPid,
+ PROCSIG_PARALLEL_MESSAGE,
+ ParallelMasterBackendId);
+}
+
+/*
+ * Look up (and possibly load) a parallel worker entry point function.
+ *
+ * For functions contained in the core code, we use library name "postgres"
+ * and consult the InternalParallelWorkers array. External functions are
+ * looked up, and loaded if necessary, using load_external_function().
+ *
+ * The point of this is to pass function names as strings across process
+ * boundaries. We can't pass actual function addresses because of the
+ * possibility that the function has been loaded at a different address
+ * in a different process. This is obviously a hazard for functions in
+ * loadable libraries, but it can happen even for functions in the core code
+ * on platforms using EXEC_BACKEND (e.g., Windows).
+ *
+ * At some point it might be worthwhile to get rid of InternalParallelWorkers[]
+ * in favor of applying load_external_function() for core functions too;
+ * but that raises portability issues that are not worth addressing now.
+ */
+static parallel_worker_main_type
+LookupParallelWorkerFunction(const char *libraryname, const char *funcname)
+{
+ /*
+ * If the function is to be loaded from postgres itself, search the
+ * InternalParallelWorkers array.
+ */
+ if (strcmp(libraryname, "postgres") == 0)
+ {
+ int i;
+
+ for (i = 0; i < lengthof(InternalParallelWorkers); i++)
+ {
+ if (strcmp(InternalParallelWorkers[i].fn_name, funcname) == 0)
+ return InternalParallelWorkers[i].fn_addr;
+ }
+
+ /* We can only reach this by programming error. */
+ elog(ERROR, "internal function \"%s\" not found", funcname);
+ }
+
+ /* Otherwise load from external library. */
+ return (parallel_worker_main_type)
+ load_external_function(libraryname, funcname, true, NULL);
+}
projects / postgresql / blobdiff