/*
* Primary entry point for VACUUM and ANALYZE commands.
*
* relid is normally InvalidOid; if it is not, then it provides the relation
* OID to be processed, and vacstmt->relation is ignored. (The non-invalid
* case is currently only used by autovacuum.)
*
* do_toast is passed as FALSE by autovacuum, because it processes TOAST
* tables separately.
*
* for_wraparound is used by autovacuum to let us know when it's forcing
* a vacuum for wraparound, which should not be auto-cancelled.
*
* bstrategy is normally given as NULL, but in autovacuum it can be passed
* in to use the same buffer strategy object across multiple vacuum() calls.
*
* isTopLevel should be passed down from ProcessUtility.
*
* It is the caller's responsibility that vacstmt and bstrategy
* (if given) be allocated in a memory context that won't disappear
* at transaction commit.
*/
void
vacuum(VacuumStmt *vacstmt, Oid relid, bool do_toast,
BufferAccessStrategy bstrategy, bool for_wraparound, bool isTopLevel)
{
const char *stmttype;
volatile bool all_rels,
in_outer_xact,
use_own_xacts;
List *relations;
/* sanity checks on options */
Assert(vacstmt->options & (VACOPT_VACUUM | VACOPT_ANALYZE));
Assert((vacstmt->options & VACOPT_VACUUM) ||
!(vacstmt->options & (VACOPT_FULL | VACOPT_FREEZE)));
Assert((vacstmt->options & VACOPT_ANALYZE) || vacstmt->va_cols == NIL);
stmttype = (vacstmt->options & VACOPT_VACUUM) ? "VACUUM" : "ANALYZE";
/*
* We cannot run VACUUM inside a user transaction block; if we were inside
* a transaction, then our commit- and start-transaction-command calls
* would not have the intended effect! There are numerous other subtle
* dependencies on this, too.
*
* ANALYZE (without VACUUM) can run either way.
*/
if (vacstmt->options & VACOPT_VACUUM)
{
PreventTransactionChain(isTopLevel, stmttype);
in_outer_xact = false;
}
else
in_outer_xact = IsInTransactionChain(isTopLevel);
/*
* Send info about dead objects to the statistics collector, unless we are
* in autovacuum --- autovacuum.c does this for itself.
*/
if ((vacstmt->options & VACOPT_VACUUM) && !IsAutoVacuumWorkerProcess())
pgstat_vacuum_stat();
/*
* Create special memory context for cross-transaction storage.
*
* Since it is a child of PortalContext, it will go away eventually even
* if we suffer an error; there's no need for special abort cleanup logic.
*/
vac_context = AllocSetContextCreate(PortalContext,
"Vacuum",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
/*
* If caller didn't give us a buffer strategy object, make one in the
* cross-transaction memory context.
*/
if (bstrategy == NULL)
{
MemoryContext old_context = MemoryContextSwitchTo(vac_context);
bstrategy = GetAccessStrategy(BAS_VACUUM);
MemoryContextSwitchTo(old_context);
}
vac_strategy = bstrategy;
/* Remember whether we are processing everything in the DB */
all_rels = (!OidIsValid(relid) && vacstmt->relation == NULL);
/*
* Build list of relations to process, unless caller gave us one. (If we
* build one, we put it in vac_context for safekeeping.)
*/
relations = get_rel_oids(relid, vacstmt->relation);
/*
* Decide whether we need to start/commit our own transactions.
*
* For VACUUM (with or without ANALYZE): always do so, so that we can
* release locks as soon as possible. (We could possibly use the outer
* transaction for a one-table VACUUM, but handling TOAST tables would be
* problematic.)
*
* For ANALYZE (no VACUUM): if inside a transaction block, we cannot
* start/commit our own transactions. Also, there's no need to do so if
* only processing one relation. For multiple relations when not within a
* transaction block, and also in an autovacuum worker, use own
* transactions so we can release locks sooner.
*/
if (vacstmt->options & VACOPT_VACUUM)
use_own_xacts = true;
else
{
Assert(vacstmt->options & VACOPT_ANALYZE);
if (IsAutoVacuumWorkerProcess())
use_own_xacts = true;
else if (in_outer_xact)
use_own_xacts = false;
else if (list_length(relations) > 1)
use_own_xacts = true;
else
use_own_xacts = false;
}
/*
* vacuum_rel expects to be entered with no transaction active; it will
* start and commit its own transaction. But we are called by an SQL
* command, and so we are executing inside a transaction already. We
* commit the transaction started in PostgresMain() here, and start
* another one before exiting to match the commit waiting for us back in
* PostgresMain().
*/
if (use_own_xacts)
{
/* ActiveSnapshot is not set by autovacuum */
if (ActiveSnapshotSet())
PopActiveSnapshot();
/* matches the StartTransaction in PostgresMain() */
CommitTransactionCommand();
}
/* Turn vacuum cost accounting on or off */
PG_TRY();
{
ListCell *cur;
VacuumCostActive = (VacuumCostDelay > 0);
VacuumCostBalance = 0;
/*
* Loop to process each selected relation.
*/
foreach(cur, relations)
{
Oid relid = lfirst_oid(cur);
bool scanned_all = false;
if (vacstmt->options & VACOPT_VACUUM)
vacuum_rel(relid, vacstmt, do_toast, for_wraparound,
&scanned_all);
if (vacstmt->options & VACOPT_ANALYZE)
{
/*
* If using separate xacts, start one for analyze. Otherwise,
* we can use the outer transaction.
*/
if (use_own_xacts)
{
StartTransactionCommand();
/* functions in indexes may want a snapshot set */
PushActiveSnapshot(GetTransactionSnapshot());
}
analyze_rel(relid, vacstmt, vac_strategy, !scanned_all);
if (use_own_xacts)
{
PopActiveSnapshot();
CommitTransactionCommand();
}
}
}
}
/* ----------------------------------------------------------------
* ExecInitSetOp
*
* This initializes the setop node state structures and
* the node's subplan.
* ----------------------------------------------------------------
*/
SetOpState *
ExecInitSetOp(SetOp *node, EState *estate, int eflags)
{
SetOpState *setopstate;
/* check for unsupported flags */
Assert(!(eflags & (EXEC_FLAG_BACKWARD | EXEC_FLAG_MARK)));
/*
* create state structure
*/
setopstate = makeNode(SetOpState);
setopstate->ps.plan = (Plan *) node;
setopstate->ps.state = estate;
setopstate->ps.ps_OuterTupleSlot = NULL;
setopstate->subplan_done = false;
setopstate->numOutput = 0;
/*
* Miscellaneous initialization
*
* SetOp nodes have no ExprContext initialization because they never call
* ExecQual or ExecProject. But they do need a per-tuple memory context
* anyway for calling execTuplesMatch.
*/
setopstate->tempContext =
AllocSetContextCreate(CurrentMemoryContext,
"SetOp",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
#define SETOP_NSLOTS 1
/*
* Tuple table initialization
*/
ExecInitResultTupleSlot(estate, &setopstate->ps);
/*
* then initialize outer plan
*/
outerPlanState(setopstate) = ExecInitNode(outerPlan(node), estate, eflags);
/*
* setop nodes do no projections, so initialize projection info for this
* node appropriately
*/
ExecAssignResultTypeFromTL(&setopstate->ps);
setopstate->ps.ps_ProjInfo = NULL;
/*
* Precompute fmgr lookup data for inner loop
*/
setopstate->eqfunctions =
execTuplesMatchPrepare(ExecGetResultType(&setopstate->ps),
node->numCols,
node->dupColIdx);
return setopstate;
}
/*
* hash_create -- create a new dynamic hash table
*
* tabname: a name for the table (for debugging purposes)
* nelem: maximum number of elements expected
* *info: additional table parameters, as indicated by flags
* flags: bitmask indicating which parameters to take from *info
*
* Note: for a shared-memory hashtable, nelem needs to be a pretty good
* estimate, since we can't expand the table on the fly. But an unshared
* hashtable can be expanded on-the-fly, so it's better for nelem to be
* on the small side and let the table grow if it's exceeded. An overly
* large nelem will penalize hash_seq_search speed without buying much.
*/
HTAB *
hash_create(const char *tabname, long nelem, HASHCTL *info, int flags)
{
HTAB *hashp;
HASHHDR *hctl;
/*
* For shared hash tables, we have a local hash header (HTAB struct) that
* we allocate in TopMemoryContext; all else is in shared memory.
*
* For non-shared hash tables, everything including the hash header is in
* a memory context created specially for the hash table --- this makes
* hash_destroy very simple. The memory context is made a child of either
* a context specified by the caller, or TopMemoryContext if nothing is
* specified.
*/
if (flags & HASH_SHARED_MEM)
{
/* Set up to allocate the hash header */
CurrentDynaHashCxt = TopMemoryContext;
}
else
{
/* Create the hash table's private memory context */
if (flags & HASH_CONTEXT)
CurrentDynaHashCxt = info->hcxt;
else
CurrentDynaHashCxt = TopMemoryContext;
CurrentDynaHashCxt = AllocSetContextCreate(CurrentDynaHashCxt,
tabname,
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
}
/* Initialize the hash header, plus a copy of the table name */
hashp = (HTAB *) DynaHashAlloc(sizeof(HTAB) + strlen(tabname) +1);
MemSet(hashp, 0, sizeof(HTAB));
hashp->tabname = (char *) (hashp + 1);
strcpy(hashp->tabname, tabname);
if (flags & HASH_FUNCTION)
hashp->hash = info->hash;
else
hashp->hash = string_hash; /* default hash function */
/*
* If you don't specify a match function, it defaults to string_compare if
* you used string_hash (either explicitly or by default) and to memcmp
* otherwise. (Prior to PostgreSQL 7.4, memcmp was always used.)
*/
if (flags & HASH_COMPARE)
hashp->match = info->match;
else if (hashp->hash == string_hash)
hashp->match = (HashCompareFunc) string_compare;
else
hashp->match = memcmp;
/*
* Similarly, the key-copying function defaults to strlcpy or memcpy.
*/
if (flags & HASH_KEYCOPY)
hashp->keycopy = info->keycopy;
else if (hashp->hash == string_hash)
hashp->keycopy = (HashCopyFunc) strlcpy;
else
hashp->keycopy = memcpy;
if (flags & HASH_ALLOC)
hashp->alloc = info->alloc;
else
hashp->alloc = DynaHashAlloc;
if (flags & HASH_SHARED_MEM)
{
/*
* ctl structure and directory are preallocated for shared memory
* tables. Note that HASH_DIRSIZE and HASH_ALLOC had better be set as
* well.
*/
hashp->hctl = info->hctl;
hashp->dir = (HASHSEGMENT *) (((char *) info->hctl) + sizeof(HASHHDR));
hashp->hcxt = NULL;
hashp->isshared = true;
/* hash table already exists, we're just attaching to it */
if (flags & HASH_ATTACH)
{
/* make local copies of some heavily-used values */
hctl = hashp->hctl;
hashp->keysize = hctl->keysize;
hashp->ssize = hctl->ssize;
hashp->sshift = hctl->sshift;
return hashp;
}
}
else
{
/* setup hash table defaults */
hashp->hctl = NULL;
hashp->dir = NULL;
hashp->hcxt = CurrentDynaHashCxt;
hashp->isshared = false;
}
if (!hashp->hctl)
{
hashp->hctl = (HASHHDR *) hashp->alloc(sizeof(HASHHDR));
if (!hashp->hctl)
ereport(ERROR,
(errcode(ERRCODE_OUT_OF_MEMORY),
errmsg("out of memory")));
}
hashp->frozen = false;
hdefault(hashp);
hctl = hashp->hctl;
if (flags & HASH_PARTITION)
{
/* Doesn't make sense to partition a local hash table */
Assert(flags & HASH_SHARED_MEM);
/*
* The number of partitions had better be a power of 2. Also, it must
* be less than INT_MAX (see init_htab()), so call the int version of
* next_pow2.
*/
Assert(info->num_partitions == next_pow2_int(info->num_partitions));
hctl->num_partitions = info->num_partitions;
}
if (flags & HASH_SEGMENT)
{
hctl->ssize = info->ssize;
hctl->sshift = my_log2(info->ssize);
/* ssize had better be a power of 2 */
Assert(hctl->ssize == (1L << hctl->sshift));
}
if (flags & HASH_FFACTOR)
hctl->ffactor = info->ffactor;
/*
* SHM hash tables have fixed directory size passed by the caller.
*/
if (flags & HASH_DIRSIZE)
{
hctl->max_dsize = info->max_dsize;
hctl->dsize = info->dsize;
}
/*
* hash table now allocates space for key and data but you have to say how
* much space to allocate
*/
if (flags & HASH_ELEM)
{
Assert(info->entrysize >= info->keysize);
hctl->keysize = info->keysize;
hctl->entrysize = info->entrysize;
}
/* make local copies of heavily-used constant fields */
hashp->keysize = hctl->keysize;
hashp->ssize = hctl->ssize;
hashp->sshift = hctl->sshift;
/* Build the hash directory structure */
if (!init_htab(hashp, nelem))
elog(ERROR, "failed to initialize hash table \"%s\"", hashp->tabname);
/*
* For a shared hash table, preallocate the requested number of elements.
* This reduces problems with run-time out-of-shared-memory conditions.
*
* For a non-shared hash table, preallocate the requested number of
* elements if it's less than our chosen nelem_alloc. This avoids wasting
* space if the caller correctly estimates a small table size.
*/
if ((flags & HASH_SHARED_MEM) ||
nelem < hctl->nelem_alloc)
{
if (!element_alloc(hashp, (int) nelem))
ereport(ERROR,
(errcode(ERRCODE_OUT_OF_MEMORY),
errmsg("out of memory")));
}
if (flags & HASH_FIXED_SIZE)
hashp->isfixed = true;
return hashp;
}
/*
* Move tuples from pending pages into regular GIN structure.
*
* This can be called concurrently by multiple backends, so it must cope.
* On first glance it looks completely not concurrent-safe and not crash-safe
* either. The reason it's okay is that multiple insertion of the same entry
* is detected and treated as a no-op by gininsert.c. If we crash after
* posting entries to the main index and before removing them from the
* pending list, it's okay because when we redo the posting later on, nothing
* bad will happen. Likewise, if two backends simultaneously try to post
* a pending entry into the main index, one will succeed and one will do
* nothing. We try to notice when someone else is a little bit ahead of
* us in the process, but that's just to avoid wasting cycles. Only the
* action of removing a page from the pending list really needs exclusive
* lock.
*
* vac_delay indicates that ginInsertCleanup is called from vacuum process,
* so call vacuum_delay_point() periodically.
* If stats isn't null, we count deleted pending pages into the counts.
*/
void
ginInsertCleanup(GinState *ginstate,
bool vac_delay, IndexBulkDeleteResult *stats)
{
Relation index = ginstate->index;
Buffer metabuffer,
buffer;
Page metapage,
page;
GinMetaPageData *metadata;
MemoryContext opCtx,
oldCtx;
BuildAccumulator accum;
KeyArray datums;
BlockNumber blkno;
metabuffer = ReadBuffer(index, GIN_METAPAGE_BLKNO);
LockBuffer(metabuffer, GIN_SHARE);
metapage = BufferGetPage(metabuffer);
metadata = GinPageGetMeta(metapage);
if (metadata->head == InvalidBlockNumber)
{
/* Nothing to do */
UnlockReleaseBuffer(metabuffer);
return;
}
/*
* Read and lock head of pending list
*/
blkno = metadata->head;
buffer = ReadBuffer(index, blkno);
LockBuffer(buffer, GIN_SHARE);
page = BufferGetPage(buffer);
LockBuffer(metabuffer, GIN_UNLOCK);
/*
* Initialize. All temporary space will be in opCtx
*/
opCtx = AllocSetContextCreate(CurrentMemoryContext,
"GIN insert cleanup temporary context",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
oldCtx = MemoryContextSwitchTo(opCtx);
initKeyArray(&datums, 128);
ginInitBA(&accum);
accum.ginstate = ginstate;
/*
* At the top of this loop, we have pin and lock on the current page of
* the pending list. However, we'll release that before exiting the loop.
* Note we also have pin but not lock on the metapage.
*/
for (;;)
{
if (GinPageIsDeleted(page))
{
/* another cleanup process is running concurrently */
UnlockReleaseBuffer(buffer);
break;
}
/*
* read page's datums into accum
*/
processPendingPage(&accum, &datums, page, FirstOffsetNumber);
vacuum_delay_point();
/*
* Is it time to flush memory to disk? Flush if we are at the end of
* the pending list, or if we have a full row and memory is getting
* full.
*
* XXX using up maintenance_work_mem here is probably unreasonably
* much, since vacuum might already be using that much.
*/
if (GinPageGetOpaque(page)->rightlink == InvalidBlockNumber ||
(GinPageHasFullRow(page) &&
(accum.allocatedMemory >= maintenance_work_mem * 1024L)))
{
ItemPointerData *list;
uint32 nlist;
Datum key;
GinNullCategory category;
OffsetNumber maxoff,
attnum;
/*
* Unlock current page to increase performance. Changes of page
* will be checked later by comparing maxoff after completion of
* memory flush.
*/
maxoff = PageGetMaxOffsetNumber(page);
LockBuffer(buffer, GIN_UNLOCK);
/*
* Moving collected data into regular structure can take
* significant amount of time - so, run it without locking pending
* list.
*/
ginBeginBAScan(&accum);
while ((list = ginGetBAEntry(&accum,
&attnum, &key, &category, &nlist)) != NULL)
{
ginEntryInsert(ginstate, attnum, key, category,
list, nlist, NULL);
vacuum_delay_point();
}
/*
* Lock the whole list to remove pages
*/
LockBuffer(metabuffer, GIN_EXCLUSIVE);
LockBuffer(buffer, GIN_SHARE);
if (GinPageIsDeleted(page))
{
/* another cleanup process is running concurrently */
UnlockReleaseBuffer(buffer);
LockBuffer(metabuffer, GIN_UNLOCK);
break;
}
/*
* While we left the page unlocked, more stuff might have gotten
* added to it. If so, process those entries immediately. There
* shouldn't be very many, so we don't worry about the fact that
* we're doing this with exclusive lock. Insertion algorithm
* guarantees that inserted row(s) will not continue on next page.
* NOTE: intentionally no vacuum_delay_point in this loop.
*/
if (PageGetMaxOffsetNumber(page) != maxoff)
{
ginInitBA(&accum);
processPendingPage(&accum, &datums, page, maxoff + 1);
ginBeginBAScan(&accum);
while ((list = ginGetBAEntry(&accum,
&attnum, &key, &category, &nlist)) != NULL)
ginEntryInsert(ginstate, attnum, key, category,
list, nlist, NULL);
}
/*
* Remember next page - it will become the new list head
*/
blkno = GinPageGetOpaque(page)->rightlink;
UnlockReleaseBuffer(buffer); /* shiftList will do exclusive
* locking */
/*
* remove read pages from pending list, at this point all
* content of read pages is in regular structure
*/
if (shiftList(index, metabuffer, blkno, stats))
{
/* another cleanup process is running concurrently */
LockBuffer(metabuffer, GIN_UNLOCK);
break;
}
Assert(blkno == metadata->head);
LockBuffer(metabuffer, GIN_UNLOCK);
/*
* if we removed the whole pending list just exit
*/
if (blkno == InvalidBlockNumber)
break;
/*
* release memory used so far and reinit state
*/
MemoryContextReset(opCtx);
initKeyArray(&datums, datums.maxvalues);
ginInitBA(&accum);
}
else
{
blkno = GinPageGetOpaque(page)->rightlink;
UnlockReleaseBuffer(buffer);
}
/*
* Read next page in pending list
*/
vacuum_delay_point();
buffer = ReadBuffer(index, blkno);
LockBuffer(buffer, GIN_SHARE);
page = BufferGetPage(buffer);
}
ReleaseBuffer(metabuffer);
/* Clean up temporary space */
MemoryContextSwitchTo(oldCtx);
MemoryContextDelete(opCtx);
}
Datum
ginbuild(PG_FUNCTION_ARGS)
{
Relation heap = (Relation) PG_GETARG_POINTER(0);
Relation index = (Relation) PG_GETARG_POINTER(1);
IndexInfo *indexInfo = (IndexInfo *) PG_GETARG_POINTER(2);
IndexBuildResult *result;
double reltuples;
GinBuildState buildstate;
Buffer RootBuffer,
MetaBuffer;
ItemPointerData *list;
Datum entry;
uint32 nlist;
MemoryContext oldCtx;
OffsetNumber attnum;
if (RelationGetNumberOfBlocks(index) != 0)
elog(ERROR, "index \"%s\" already contains data",
RelationGetRelationName(index));
initGinState(&buildstate.ginstate, index);
/* initialize the meta page */
MetaBuffer = GinNewBuffer(index);
/* initialize the root page */
RootBuffer = GinNewBuffer(index);
START_CRIT_SECTION();
GinInitMetabuffer(MetaBuffer);
MarkBufferDirty(MetaBuffer);
GinInitBuffer(RootBuffer, GIN_LEAF);
MarkBufferDirty(RootBuffer);
if (!index->rd_istemp)
{
XLogRecPtr recptr;
XLogRecData rdata;
Page page;
rdata.buffer = InvalidBuffer;
rdata.data = (char *) &(index->rd_node);
rdata.len = sizeof(RelFileNode);
rdata.next = NULL;
recptr = XLogInsert(RM_GIN_ID, XLOG_GIN_CREATE_INDEX, &rdata);
page = BufferGetPage(RootBuffer);
PageSetLSN(page, recptr);
PageSetTLI(page, ThisTimeLineID);
page = BufferGetPage(MetaBuffer);
PageSetLSN(page, recptr);
PageSetTLI(page, ThisTimeLineID);
}
UnlockReleaseBuffer(MetaBuffer);
UnlockReleaseBuffer(RootBuffer);
END_CRIT_SECTION();
/* build the index */
buildstate.indtuples = 0;
/*
* create a temporary memory context that is reset once for each tuple
* inserted into the index
*/
buildstate.tmpCtx = AllocSetContextCreate(CurrentMemoryContext,
"Gin build temporary context",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
buildstate.funcCtx = AllocSetContextCreate(buildstate.tmpCtx,
"Gin build temporary context for user-defined function",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
buildstate.accum.ginstate = &buildstate.ginstate;
ginInitBA(&buildstate.accum);
/*
* Do the heap scan. We disallow sync scan here because dataPlaceToPage
* prefers to receive tuples in TID order.
*/
reltuples = IndexBuildHeapScan(heap, index, indexInfo, false,
ginBuildCallback, (void *) &buildstate);
/* dump remaining entries to the index */
oldCtx = MemoryContextSwitchTo(buildstate.tmpCtx);
ginBeginBAScan(&buildstate.accum);
while ((list = ginGetEntry(&buildstate.accum, &attnum, &entry, &nlist)) != NULL)
{
/* there could be many entries, so be willing to abort here */
CHECK_FOR_INTERRUPTS();
ginEntryInsert(index, &buildstate.ginstate, attnum, entry, list, nlist, TRUE);
}
MemoryContextSwitchTo(oldCtx);
MemoryContextDelete(buildstate.tmpCtx);
/*
* Return statistics
*/
result = (IndexBuildResult *) palloc(sizeof(IndexBuildResult));
result->heap_tuples = reltuples;
result->index_tuples = buildstate.indtuples;
PG_RETURN_POINTER(result);
}
Datum Type_invokeSRF(Type self, jclass cls, jmethodID method, jvalue* args, PG_FUNCTION_ARGS)
{
bool hasRow;
CallContextData* ctxData;
FuncCallContext* context;
MemoryContext currCtx;
/* stuff done only on the first call of the function
*/
if(SRF_IS_FIRSTCALL())
{
jobject tmp;
/* create a function context for cross-call persistence
*/
context = SRF_FIRSTCALL_INIT();
currCtx = MemoryContextSwitchTo(context->multi_call_memory_ctx);
/* Call the declared Java function. It returns an instance that can produce
* the rows.
*/
tmp = Type_getSRFProducer(self, cls, method, args);
if(tmp == 0)
{
Invocation_assertDisconnect();
MemoryContextSwitchTo(currCtx);
fcinfo->isnull = true;
SRF_RETURN_DONE(context);
}
ctxData = (CallContextData*)palloc(sizeof(CallContextData));
context->user_fctx = ctxData;
ctxData->elemType = self;
ctxData->rowProducer = JNI_newGlobalRef(tmp);
JNI_deleteLocalRef(tmp);
/* Some row producers will need a writable result set in order
* to produce the row. If one is needed, it's created here.
*/
tmp = Type_getSRFCollector(self, fcinfo);
if(tmp == 0)
ctxData->rowCollector = 0;
else
{
ctxData->rowCollector = JNI_newGlobalRef(tmp);
JNI_deleteLocalRef(tmp);
}
ctxData->trusted = currentInvocation->trusted;
ctxData->hasConnected = currentInvocation->hasConnected;
ctxData->invocation = currentInvocation->invocation;
if(ctxData->hasConnected)
ctxData->spiContext = CurrentMemoryContext;
else
ctxData->spiContext = 0;
ctxData->rowContext = AllocSetContextCreate(context->multi_call_memory_ctx,
"PL/Java row context",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
/* Register callback to be called when the function ends
*/
RegisterExprContextCallback(((ReturnSetInfo*)fcinfo->resultinfo)->econtext, _endOfSetCB, PointerGetDatum(ctxData));
MemoryContextSwitchTo(currCtx);
}
context = SRF_PERCALL_SETUP();
ctxData = (CallContextData*)context->user_fctx;
MemoryContextReset(ctxData->rowContext);
currCtx = MemoryContextSwitchTo(ctxData->rowContext);
currentInvocation->hasConnected = ctxData->hasConnected;
currentInvocation->invocation = ctxData->invocation;
hasRow = Type_hasNextSRF(self, ctxData->rowProducer, ctxData->rowCollector, (jint)context->call_cntr);
ctxData->hasConnected = currentInvocation->hasConnected;
ctxData->invocation = currentInvocation->invocation;
currentInvocation->hasConnected = false;
currentInvocation->invocation = 0;
if(hasRow)
{
Datum result = Type_nextSRF(self, ctxData->rowProducer, ctxData->rowCollector);
MemoryContextSwitchTo(currCtx);
SRF_RETURN_NEXT(context, result);
}
MemoryContextSwitchTo(currCtx);
/* Unregister this callback and call it manually. We do this because
* otherwise it will be called when the backend is in progress of
* cleaning up Portals. If we close cursors (i.e. drop portals) in
* the close, then that mechanism fails since attempts are made to
* delete portals more then once.
*/
UnregisterExprContextCallback(
((ReturnSetInfo*)fcinfo->resultinfo)->econtext,
_endOfSetCB,
PointerGetDatum(ctxData));
_closeIteration(ctxData);
/* This is the end of the set.
*/
SRF_RETURN_DONE(context);
}
/*
* Main entry point for walwriter process
*
* This is invoked from BootstrapMain, which has already created the basic
* execution environment, but not enabled signals yet.
*/
void
WalWriterMain(void)
{
sigjmp_buf local_sigjmp_buf;
MemoryContext walwriter_context;
/*
* If possible, make this process a group leader, so that the postmaster
* can signal any child processes too. (walwriter probably never has any
* child processes, but for consistency we make all postmaster child
* processes do this.)
*/
#ifdef HAVE_SETSID
if (setsid() < 0)
elog(FATAL, "setsid() failed: %m");
#endif
/*
* Properly accept or ignore signals the postmaster might send us
*
* We have no particular use for SIGINT at the moment, but seems
* reasonable to treat like SIGTERM.
*/
pqsignal(SIGHUP, WalSigHupHandler); /* set flag to read config file */
pqsignal(SIGINT, WalShutdownHandler); /* request shutdown */
pqsignal(SIGTERM, WalShutdownHandler); /* request shutdown */
pqsignal(SIGQUIT, wal_quickdie); /* hard crash time */
pqsignal(SIGALRM, SIG_IGN);
pqsignal(SIGPIPE, SIG_IGN);
pqsignal(SIGUSR1, SIG_IGN); /* reserve for ProcSignal */
pqsignal(SIGUSR2, SIG_IGN); /* not used */
/*
* Reset some signals that are accepted by postmaster but not here
*/
pqsignal(SIGCHLD, SIG_DFL);
pqsignal(SIGTTIN, SIG_DFL);
pqsignal(SIGTTOU, SIG_DFL);
pqsignal(SIGCONT, SIG_DFL);
pqsignal(SIGWINCH, SIG_DFL);
/* We allow SIGQUIT (quickdie) at all times */
sigdelset(&BlockSig, SIGQUIT);
/*
* Create a resource owner to keep track of our resources (not clear that
* we need this, but may as well have one).
*/
CurrentResourceOwner = ResourceOwnerCreate(NULL, "Wal Writer");
/*
* Create a memory context that we will do all our work in. We do this so
* that we can reset the context during error recovery and thereby avoid
* possible memory leaks. Formerly this code just ran in
* TopMemoryContext, but resetting that would be a really bad idea.
*/
walwriter_context = AllocSetContextCreate(TopMemoryContext,
"Wal Writer",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
MemoryContextSwitchTo(walwriter_context);
/*
* If an exception is encountered, processing resumes here.
*
* This code is heavily based on bgwriter.c, q.v.
*/
if (sigsetjmp(local_sigjmp_buf, 1) != 0)
{
/* Since not using PG_TRY, must reset error stack by hand */
error_context_stack = NULL;
/* Prevent interrupts while cleaning up */
HOLD_INTERRUPTS();
/* Report the error to the server log */
EmitErrorReport();
/*
* These operations are really just a minimal subset of
* AbortTransaction(). We don't have very many resources to worry
* about in walwriter, but we do have LWLocks, and perhaps buffers?
*/
LWLockReleaseAll();
AbortBufferIO();
UnlockBuffers();
/* buffer pins are released here: */
ResourceOwnerRelease(CurrentResourceOwner,
RESOURCE_RELEASE_BEFORE_LOCKS,
false, true);
/* we needn't bother with the other ResourceOwnerRelease phases */
AtEOXact_Buffers(false);
AtEOXact_Files();
AtEOXact_HashTables(false);
/*
* Now return to normal top-level context and clear ErrorContext for
* next time.
*/
MemoryContextSwitchTo(walwriter_context);
FlushErrorState();
/* Flush any leaked data in the top-level context */
MemoryContextResetAndDeleteChildren(walwriter_context);
/* Now we can allow interrupts again */
RESUME_INTERRUPTS();
/*
* Sleep at least 1 second after any error. A write error is likely
* to be repeated, and we don't want to be filling the error logs as
* fast as we can.
*/
pg_usleep(1000000L);
/*
* Close all open files after any error. This is helpful on Windows,
* where holding deleted files open causes various strange errors.
* It's not clear we need it elsewhere, but shouldn't hurt.
*/
smgrcloseall();
}
/* We can now handle ereport(ERROR) */
PG_exception_stack = &local_sigjmp_buf;
/*
* Unblock signals (they were blocked when the postmaster forked us)
*/
PG_SETMASK(&UnBlockSig);
/*
* Loop forever
*/
for (;;)
{
long udelay;
/*
* Emergency bailout if postmaster has died. This is to avoid the
* necessity for manual cleanup of all postmaster children.
*/
if (!PostmasterIsAlive(true))
exit(1);
/*
* Process any requests or signals received recently.
*/
if (got_SIGHUP)
{
got_SIGHUP = false;
ProcessConfigFile(PGC_SIGHUP);
}
if (shutdown_requested)
{
/* Normal exit from the walwriter is here */
proc_exit(0); /* done */
}
/*
* Do what we're here for...
*/
XLogBackgroundFlush();
/*
* Delay until time to do something more, but fall out of delay
* reasonably quickly if signaled.
*/
udelay = WalWriterDelay * 1000L;
while (udelay > 999999L)
{
if (got_SIGHUP || shutdown_requested)
break;
pg_usleep(1000000L);
udelay -= 1000000L;
}
if (!(got_SIGHUP || shutdown_requested))
pg_usleep(udelay);
}
}
/*
* Move tuples from pending pages into regular GIN structure.
*
* On first glance it looks completely not crash-safe. But if we crash
* after posting entries to the main index and before removing them from the
* pending list, it's okay because when we redo the posting later on, nothing
* bad will happen.
*
* fill_fsm indicates that ginInsertCleanup should add deleted pages
* to FSM otherwise caller is responsible to put deleted pages into
* FSM.
*
* If stats isn't null, we count deleted pending pages into the counts.
*/
void
ginInsertCleanup(GinState *ginstate, bool full_clean,
bool fill_fsm, IndexBulkDeleteResult *stats)
{
Relation index = ginstate->index;
Buffer metabuffer,
buffer;
Page metapage,
page;
GinMetaPageData *metadata;
MemoryContext opCtx,
oldCtx;
BuildAccumulator accum;
KeyArray datums;
BlockNumber blkno,
blknoFinish;
bool cleanupFinish = false;
bool fsm_vac = false;
Size workMemory;
bool inVacuum = (stats == NULL);
/*
* We would like to prevent concurrent cleanup process. For that we will
* lock metapage in exclusive mode using LockPage() call. Nobody other
* will use that lock for metapage, so we keep possibility of concurrent
* insertion into pending list
*/
if (inVacuum)
{
/*
* We are called from [auto]vacuum/analyze or gin_clean_pending_list()
* and we would like to wait concurrent cleanup to finish.
*/
LockPage(index, GIN_METAPAGE_BLKNO, ExclusiveLock);
workMemory =
(IsAutoVacuumWorkerProcess() && autovacuum_work_mem != -1) ?
autovacuum_work_mem : maintenance_work_mem;
}
else
{
/*
* We are called from regular insert and if we see concurrent cleanup
* just exit in hope that concurrent process will clean up pending
* list.
*/
if (!ConditionalLockPage(index, GIN_METAPAGE_BLKNO, ExclusiveLock))
return;
workMemory = work_mem;
}
metabuffer = ReadBuffer(index, GIN_METAPAGE_BLKNO);
LockBuffer(metabuffer, GIN_SHARE);
metapage = BufferGetPage(metabuffer);
metadata = GinPageGetMeta(metapage);
if (metadata->head == InvalidBlockNumber)
{
/* Nothing to do */
UnlockReleaseBuffer(metabuffer);
UnlockPage(index, GIN_METAPAGE_BLKNO, ExclusiveLock);
return;
}
/*
* Remember a tail page to prevent infinite cleanup if other backends add
* new tuples faster than we can cleanup.
*/
blknoFinish = metadata->tail;
/*
* Read and lock head of pending list
*/
blkno = metadata->head;
buffer = ReadBuffer(index, blkno);
LockBuffer(buffer, GIN_SHARE);
page = BufferGetPage(buffer);
LockBuffer(metabuffer, GIN_UNLOCK);
/*
* Initialize. All temporary space will be in opCtx
*/
opCtx = AllocSetContextCreate(CurrentMemoryContext,
"GIN insert cleanup temporary context",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
oldCtx = MemoryContextSwitchTo(opCtx);
initKeyArray(&datums, 128);
ginInitBA(&accum);
accum.ginstate = ginstate;
/*
* At the top of this loop, we have pin and lock on the current page of
* the pending list. However, we'll release that before exiting the loop.
* Note we also have pin but not lock on the metapage.
*/
for (;;)
{
Assert(!GinPageIsDeleted(page));
/*
* Are we walk through the page which as we remember was a tail when
* we start our cleanup? But if caller asks us to clean up whole
* pending list then ignore old tail, we will work until list becomes
* empty.
*/
if (blkno == blknoFinish && full_clean == false)
cleanupFinish = true;
/*
* read page's datums into accum
*/
processPendingPage(&accum, &datums, page, FirstOffsetNumber);
vacuum_delay_point();
/*
* Is it time to flush memory to disk? Flush if we are at the end of
* the pending list, or if we have a full row and memory is getting
* full.
*/
if (GinPageGetOpaque(page)->rightlink == InvalidBlockNumber ||
(GinPageHasFullRow(page) &&
(accum.allocatedMemory >= workMemory * 1024L)))
{
ItemPointerData *list;
uint32 nlist;
Datum key;
GinNullCategory category;
OffsetNumber maxoff,
attnum;
/*
* Unlock current page to increase performance. Changes of page
* will be checked later by comparing maxoff after completion of
* memory flush.
*/
maxoff = PageGetMaxOffsetNumber(page);
LockBuffer(buffer, GIN_UNLOCK);
/*
* Moving collected data into regular structure can take
* significant amount of time - so, run it without locking pending
* list.
*/
ginBeginBAScan(&accum);
while ((list = ginGetBAEntry(&accum,
&attnum, &key, &category, &nlist)) != NULL)
{
ginEntryInsert(ginstate, attnum, key, category,
list, nlist, NULL);
vacuum_delay_point();
}
/*
* Lock the whole list to remove pages
*/
LockBuffer(metabuffer, GIN_EXCLUSIVE);
LockBuffer(buffer, GIN_SHARE);
Assert(!GinPageIsDeleted(page));
/*
* While we left the page unlocked, more stuff might have gotten
* added to it. If so, process those entries immediately. There
* shouldn't be very many, so we don't worry about the fact that
* we're doing this with exclusive lock. Insertion algorithm
* guarantees that inserted row(s) will not continue on next page.
* NOTE: intentionally no vacuum_delay_point in this loop.
*/
if (PageGetMaxOffsetNumber(page) != maxoff)
{
ginInitBA(&accum);
processPendingPage(&accum, &datums, page, maxoff + 1);
ginBeginBAScan(&accum);
while ((list = ginGetBAEntry(&accum,
&attnum, &key, &category, &nlist)) != NULL)
ginEntryInsert(ginstate, attnum, key, category,
list, nlist, NULL);
}
/*
* Remember next page - it will become the new list head
*/
blkno = GinPageGetOpaque(page)->rightlink;
UnlockReleaseBuffer(buffer); /* shiftList will do exclusive
* locking */
/*
* remove read pages from pending list, at this point all content
* of read pages is in regular structure
*/
shiftList(index, metabuffer, blkno, fill_fsm, stats);
/* At this point, some pending pages have been freed up */
fsm_vac = true;
Assert(blkno == metadata->head);
LockBuffer(metabuffer, GIN_UNLOCK);
/*
* if we removed the whole pending list or we cleanup tail (which
* we remembered on start our cleanup process) then just exit
*/
if (blkno == InvalidBlockNumber || cleanupFinish)
break;
/*
* release memory used so far and reinit state
*/
MemoryContextReset(opCtx);
initKeyArray(&datums, datums.maxvalues);
ginInitBA(&accum);
}
else
{
blkno = GinPageGetOpaque(page)->rightlink;
UnlockReleaseBuffer(buffer);
}
/*
* Read next page in pending list
*/
vacuum_delay_point();
buffer = ReadBuffer(index, blkno);
LockBuffer(buffer, GIN_SHARE);
page = BufferGetPage(buffer);
}
UnlockPage(index, GIN_METAPAGE_BLKNO, ExclusiveLock);
ReleaseBuffer(metabuffer);
/*
* As pending list pages can have a high churn rate, it is desirable to
* recycle them immediately to the FreeSpace Map when ordinary backends
* clean the list.
*/
if (fsm_vac && fill_fsm)
IndexFreeSpaceMapVacuum(index);
/* Clean up temporary space */
MemoryContextSwitchTo(oldCtx);
MemoryContextDelete(opCtx);
}
/*
* A tuple in the heap is being inserted. To keep a brin index up to date,
* we need to obtain the relevant index tuple and compare its stored values
* with those of the new tuple. If the tuple values are not consistent with
* the summary tuple, we need to update the index tuple.
*
* If the range is not currently summarized (i.e. the revmap returns NULL for
* it), there's nothing to do.
*/
bool
brininsert(Relation idxRel, Datum *values, bool *nulls,
ItemPointer heaptid, Relation heapRel,
IndexUniqueCheck checkUnique)
{
BlockNumber pagesPerRange;
BrinDesc *bdesc = NULL;
BrinRevmap *revmap;
Buffer buf = InvalidBuffer;
MemoryContext tupcxt = NULL;
MemoryContext oldcxt = NULL;
revmap = brinRevmapInitialize(idxRel, &pagesPerRange, NULL);
for (;;)
{
bool need_insert = false;
OffsetNumber off;
BrinTuple *brtup;
BrinMemTuple *dtup;
BlockNumber heapBlk;
int keyno;
CHECK_FOR_INTERRUPTS();
heapBlk = ItemPointerGetBlockNumber(heaptid);
/* normalize the block number to be the first block in the range */
heapBlk = (heapBlk / pagesPerRange) * pagesPerRange;
brtup = brinGetTupleForHeapBlock(revmap, heapBlk, &buf, &off, NULL,
BUFFER_LOCK_SHARE, NULL);
/* if range is unsummarized, there's nothing to do */
if (!brtup)
break;
/* First time through? */
if (bdesc == NULL)
{
bdesc = brin_build_desc(idxRel);
tupcxt = AllocSetContextCreate(CurrentMemoryContext,
"brininsert cxt",
ALLOCSET_DEFAULT_SIZES);
oldcxt = MemoryContextSwitchTo(tupcxt);
}
dtup = brin_deform_tuple(bdesc, brtup);
/*
* Compare the key values of the new tuple to the stored index values;
* our deformed tuple will get updated if the new tuple doesn't fit
* the original range (note this means we can't break out of the loop
* early). Make a note of whether this happens, so that we know to
* insert the modified tuple later.
*/
for (keyno = 0; keyno < bdesc->bd_tupdesc->natts; keyno++)
{
Datum result;
BrinValues *bval;
FmgrInfo *addValue;
bval = &dtup->bt_columns[keyno];
addValue = index_getprocinfo(idxRel, keyno + 1,
BRIN_PROCNUM_ADDVALUE);
result = FunctionCall4Coll(addValue,
idxRel->rd_indcollation[keyno],
PointerGetDatum(bdesc),
PointerGetDatum(bval),
values[keyno],
nulls[keyno]);
/* if that returned true, we need to insert the updated tuple */
need_insert |= DatumGetBool(result);
}
if (!need_insert)
{
/*
* The tuple is consistent with the new values, so there's nothing
* to do.
*/
LockBuffer(buf, BUFFER_LOCK_UNLOCK);
}
else
{
Page page = BufferGetPage(buf);
ItemId lp = PageGetItemId(page, off);
Size origsz;
BrinTuple *origtup;
Size newsz;
BrinTuple *newtup;
bool samepage;
/*
* Make a copy of the old tuple, so that we can compare it after
* re-acquiring the lock.
*/
origsz = ItemIdGetLength(lp);
origtup = brin_copy_tuple(brtup, origsz);
/*
* Before releasing the lock, check if we can attempt a same-page
* update. Another process could insert a tuple concurrently in
* the same page though, so downstream we must be prepared to cope
* if this turns out to not be possible after all.
*/
newtup = brin_form_tuple(bdesc, heapBlk, dtup, &newsz);
samepage = brin_can_do_samepage_update(buf, origsz, newsz);
LockBuffer(buf, BUFFER_LOCK_UNLOCK);
/*
* Try to update the tuple. If this doesn't work for whatever
* reason, we need to restart from the top; the revmap might be
* pointing at a different tuple for this block now, so we need to
* recompute to ensure both our new heap tuple and the other
* inserter's are covered by the combined tuple. It might be that
* we don't need to update at all.
*/
if (!brin_doupdate(idxRel, pagesPerRange, revmap, heapBlk,
buf, off, origtup, origsz, newtup, newsz,
samepage))
{
/* no luck; start over */
MemoryContextResetAndDeleteChildren(tupcxt);
continue;
}
}
/* success! */
break;
}
brinRevmapTerminate(revmap);
if (BufferIsValid(buf))
ReleaseBuffer(buf);
if (bdesc != NULL)
{
brin_free_desc(bdesc);
MemoryContextSwitchTo(oldcxt);
MemoryContextDelete(tupcxt);
}
return false;
}
/*
* main entry pont of pcp worker child process
*/
void
pcp_worker_main(int port)
{
sigjmp_buf local_sigjmp_buf;
MemoryContext PCPMemoryContext;
int authenticated = 0;
char salt[4];
int random_salt = 0;
struct timeval uptime;
char tos;
int rsize;
char *buf = NULL;
ereport(DEBUG1,
(errmsg("I am PCP worker child with pid:%d",getpid())));
/* Identify myself via ps */
init_ps_display("", "", "", "");
gettimeofday(&uptime, NULL);
srandom((unsigned int) (getpid() ^ uptime.tv_usec));
/* set up signal handlers */
signal(SIGTERM, die);
signal(SIGINT, die);
signal(SIGQUIT, die);
signal(SIGCHLD, SIG_DFL);
signal(SIGUSR2, wakeup_handler_child);
signal(SIGUSR1, SIG_IGN);
signal(SIGHUP, SIG_IGN);
signal(SIGPIPE, SIG_IGN);
signal(SIGALRM, SIG_IGN);
/* Create per loop iteration memory context */
PCPMemoryContext = AllocSetContextCreate(TopMemoryContext,
"PCP_worker_main_loop",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
MemoryContextSwitchTo(TopMemoryContext);
/*
* install the call back for preparation of pcp worker child exit
*/
on_system_exit(pcp_worker_will_go_down, (Datum)NULL);
/* Initialize my backend status */
pool_initialize_private_backend_status();
/* Initialize process context */
pool_init_process_context();
pcp_frontend = pcp_open(port);
unset_nonblock(pcp_frontend->fd);
if (sigsetjmp(local_sigjmp_buf, 1) != 0)
{
error_context_stack = NULL;
EmitErrorReport();
MemoryContextSwitchTo(TopMemoryContext);
FlushErrorState();
}
/* We can now handle ereport(ERROR) */
PG_exception_stack = &local_sigjmp_buf;
for(;;)
{
MemoryContextSwitchTo(PCPMemoryContext);
MemoryContextResetAndDeleteChildren(PCPMemoryContext);
errno = 0;
/* read a PCP packet */
do_pcp_read(pcp_frontend, &tos, 1);
do_pcp_read(pcp_frontend, &rsize, sizeof(int));
rsize = ntohl(rsize);
if ((rsize - sizeof(int)) > 0)
{
buf = (char *)palloc(rsize - sizeof(int));
do_pcp_read(pcp_frontend, buf, rsize - sizeof(int));
}
ereport(DEBUG1,
(errmsg("received PCP packet"),
errdetail("PCP packet type of service '%c'", tos)));
if (tos == 'R') /* authentication */
{
set_ps_display("PCP: processing authentication", false);
process_authentication(pcp_frontend, buf,salt, &random_salt);
authenticated = 1;
continue;
}
if (tos == 'M') /* md5 salt */
{
set_ps_display("PCP: processing authentication", false);
send_md5salt(pcp_frontend, salt);
random_salt = 1;
continue;
}
/* is this connection authenticated? if not disconnect immediately*/
if (!authenticated)
ereport(FATAL,
(errmsg("authentication failed for new PCP connection"),
errdetail("connection not authorized")));
/* process a request */
pcp_process_command(tos, buf, rsize);
}
exit(0);
}
/*
* btvacuumscan --- scan the index for VACUUMing purposes
*
* This combines the functions of looking for leaf tuples that are deletable
* according to the vacuum callback, looking for empty pages that can be
* deleted, and looking for old deleted pages that can be recycled. Both
* btbulkdelete and btvacuumcleanup invoke this (the latter only if no
* btbulkdelete call occurred).
*
* The caller is responsible for initially allocating/zeroing a stats struct
* and for obtaining a vacuum cycle ID if necessary.
*/
static void
btvacuumscan(IndexVacuumInfo *info, IndexBulkDeleteResult *stats,
IndexBulkDeleteCallback callback, void *callback_state,
BTCycleId cycleid)
{
Relation rel = info->index;
BTVacState vstate;
BlockNumber num_pages;
BlockNumber blkno;
bool needLock;
/*
* Reset counts that will be incremented during the scan; needed in case
* of multiple scans during a single VACUUM command
*/
stats->estimated_count = false;
stats->num_index_tuples = 0;
stats->pages_deleted = 0;
/* Set up info to pass down to btvacuumpage */
vstate.info = info;
vstate.stats = stats;
vstate.callback = callback;
vstate.callback_state = callback_state;
vstate.cycleid = cycleid;
vstate.lastBlockVacuumed = BTREE_METAPAGE; /* Initialise at first block */
vstate.lastBlockLocked = BTREE_METAPAGE;
vstate.totFreePages = 0;
/* Create a temporary memory context to run _bt_pagedel in */
vstate.pagedelcontext = AllocSetContextCreate(CurrentMemoryContext,
"_bt_pagedel",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
/*
* The outer loop iterates over all index pages except the metapage, in
* physical order (we hope the kernel will cooperate in providing
* read-ahead for speed). It is critical that we visit all leaf pages,
* including ones added after we start the scan, else we might fail to
* delete some deletable tuples. Hence, we must repeatedly check the
* relation length. We must acquire the relation-extension lock while
* doing so to avoid a race condition: if someone else is extending the
* relation, there is a window where bufmgr/smgr have created a new___
* all-zero page but it hasn't yet been write-locked by _bt_getbuf(). If
* we manage to scan such a page here, we'll improperly assume it can be
* recycled. Taking the lock synchronizes things enough to prevent a
* problem: either num_pages won't include the new___ page, or _bt_getbuf
* already has write lock on the buffer and it will be fully initialized
* before we can examine it. (See also vacuumlazy.c, which has the same
* issue.) Also, we need not worry if a page is added immediately after
* we look; the page splitting code already has write-lock on the left
* page before it adds a right page, so we must already have processed any
* tuples due to be moved into such a page.
*
* We can skip locking for new___ or temp relations, however, since no one
* else could be accessing them.
*/
needLock = !RELATION_IS_LOCAL(rel);
blkno = BTREE_METAPAGE + 1;
for (;;)
{
/* Get the current relation length */
if (needLock)
LockRelationForExtension(rel, ExclusiveLock);
num_pages = RelationGetNumberOfBlocks(rel);
if (needLock)
UnlockRelationForExtension(rel, ExclusiveLock);
/* Quit if we've scanned the whole relation */
if (blkno >= num_pages)
break;
/* Iterate over pages, then loop back to recheck length */
for (; blkno < num_pages; blkno++)
{
btvacuumpage(&vstate, blkno, blkno);
}
}
/*
* If the WAL is replayed in hot standby, the replay process needs to get
* cleanup locks on all index leaf pages, just as we've been doing here.
* However, we won't issue any WAL records about pages that have no items
* to be deleted. For pages between pages we've vacuumed, the replay code
* will take locks under the direction of the lastBlockVacuumed fields in
* the XLOG_BTREE_VACUUM WAL records. To cover pages after the last one
* we vacuum, we need to issue a dummy XLOG_BTREE_VACUUM WAL record
* against the last leaf page in the index, if that one wasn't vacuumed.
*/
if (XLogStandbyInfoActive() &&
vstate.lastBlockVacuumed < vstate.lastBlockLocked)
{
Buffer buf;
/*
* The page should be valid, but we can't use _bt_getbuf() because we
* want to use a nondefault buffer access strategy. Since we aren't
* going to delete any items, getting cleanup lock again is probably
* overkill, but for consistency do that anyway.
*/
buf = ReadBufferExtended(rel, MAIN_FORKNUM, vstate.lastBlockLocked,
RBM_NORMAL, info->strategy);
LockBufferForCleanup(buf);
_bt_checkpage(rel, buf);
_bt_delitems_vacuum(rel, buf, NULL, 0, vstate.lastBlockVacuumed);
_bt_relbuf(rel, buf);
}
MemoryContextDelete(vstate.pagedelcontext);
/* update statistics */
stats->num_pages = num_pages;
stats->pages_free = vstate.totFreePages;
}
/*
* geqo_eval
*
* Returns cost of a query tree as an individual of the population.
*/
Cost
geqo_eval(PlannerInfo *root, Gene *tour, int num_gene)
{
MemoryContext mycontext;
MemoryContext oldcxt;
RelOptInfo *joinrel;
Path *best_path;
Cost fitness;
int savelength;
struct HTAB *savehash;
/*
* Create a private memory context that will hold all temp storage
* allocated inside gimme_tree().
*
* Since geqo_eval() will be called many times, we can't afford to let all
* that memory go unreclaimed until end of statement. Note we make the
* temp context a child of the planner's normal context, so that it will
* be freed even if we abort via ereport(ERROR).
*/
mycontext = AllocSetContextCreate(CurrentMemoryContext,
"GEQO",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
oldcxt = MemoryContextSwitchTo(mycontext);
/*
* gimme_tree will add entries to root->join_rel_list, which may or may
* not already contain some entries. The newly added entries will be
* recycled by the MemoryContextDelete below, so we must ensure that the
* list is restored to its former state before exiting. We can do this by
* truncating the list to its original length. NOTE this assumes that any
* added entries are appended at the end!
*
* We also must take care not to mess up the outer join_rel_hash, if there
* is one. We can do this by just temporarily setting the link to NULL.
* (If we are dealing with enough join rels, which we very likely are, a
* new hash table will get built and used locally.)
*
* join_rel_level[] shouldn't be in use, so just Assert it isn't.
*/
savelength = list_length(root->join_rel_list);
savehash = root->join_rel_hash;
Assert(root->join_rel_level == NULL);
root->join_rel_hash = NULL;
/* construct the best path for the given combination of relations */
joinrel = gimme_tree(root, tour, num_gene);
best_path = joinrel->cheapest_total_path;
/*
* compute fitness
*
* XXX geqo does not currently support optimization for partial result
* retrieval, nor do we take any cognizance of possible use of
* parameterized paths --- how to fix?
*/
fitness = best_path->total_cost;
/*
* Restore join_rel_list to its former state, and put back original
* hashtable if any.
*/
root->join_rel_list = list_truncate(root->join_rel_list,
savelength);
root->join_rel_hash = savehash;
/* release all the memory acquired within gimme_tree */
MemoryContextSwitchTo(oldcxt);
MemoryContextDelete(mycontext);
return fitness;
}
/* ----------------------------------------------------------------
* ExecHashTableCreate
*
* create an empty hashtable data structure for hashjoin.
* ----------------------------------------------------------------
*/
HashJoinTable
ExecHashTableCreate(HashState *hashState, HashJoinState *hjstate, List *hashOperators, uint64 operatorMemKB)
{
HashJoinTable hashtable;
Plan *outerNode;
int nbuckets;
int nbatch;
int nkeys;
int i;
ListCell *ho;
MemoryContext oldcxt;
START_MEMORY_ACCOUNT(hashState->ps.plan->memoryAccount);
{
Hash *node = (Hash *) hashState->ps.plan;
/*
* Get information about the size of the relation to be hashed (it's the
* "outer" subtree of this node, but the inner relation of the hashjoin).
* Compute the appropriate size of the hash table.
*/
outerNode = outerPlan(node);
/*
* Initialize the hash table control block.
*
* The hashtable control block is just palloc'd from the executor's
* per-query memory context.
*/
hashtable = (HashJoinTable)palloc0(sizeof(HashJoinTableData));
hashtable->buckets = NULL;
hashtable->bloom = NULL;
hashtable->curbatch = 0;
hashtable->growEnabled = true;
hashtable->totalTuples = 0;
hashtable->batches = NULL;
hashtable->work_set = NULL;
hashtable->state_file = NULL;
hashtable->spaceAllowed = operatorMemKB * 1024L;
hashtable->stats = NULL;
hashtable->eagerlyReleased = false;
hashtable->hjstate = hjstate;
/*
* Create temporary memory contexts in which to keep the hashtable working
* storage. See notes in executor/hashjoin.h.
*/
hashtable->hashCxt = AllocSetContextCreate(CurrentMemoryContext,
"HashTableContext",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
hashtable->batchCxt = AllocSetContextCreate(hashtable->hashCxt,
"HashBatchContext",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
/* CDB */ /* track temp buf file allocations in separate context */
hashtable->bfCxt = AllocSetContextCreate(CurrentMemoryContext,
"hbbfcxt",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
ExecChooseHashTableSize(outerNode->plan_rows, outerNode->plan_width,
&hashtable->nbuckets, &hashtable->nbatch, operatorMemKB);
nbuckets = hashtable->nbuckets;
nbatch = hashtable->nbatch;
hashtable->nbatch_original = nbatch;
hashtable->nbatch_outstart = nbatch;
#ifdef HJDEBUG
elog(LOG, "HJ: nbatch = %d, nbuckets = %d\n", nbatch, nbuckets);
#endif
/*
* Get info about the hash functions to be used for each hash key.
* Also remember whether the join operators are strict.
*/
nkeys = list_length(hashOperators);
hashtable->outer_hashfunctions =
(FmgrInfo *) palloc(nkeys * sizeof(FmgrInfo));
hashtable->inner_hashfunctions =
(FmgrInfo *) palloc(nkeys * sizeof(FmgrInfo));
hashtable->hashStrict = (bool *) palloc(nkeys * sizeof(bool));
i = 0;
foreach(ho, hashOperators)
{
Oid hashop = lfirst_oid(ho);
Oid left_hashfn;
Oid right_hashfn;
if (!get_op_hash_functions(hashop, &left_hashfn, &right_hashfn))
elog(ERROR, "could not find hash function for hash operator %u",
hashop);
fmgr_info(left_hashfn, &hashtable->outer_hashfunctions[i]);
fmgr_info(right_hashfn, &hashtable->inner_hashfunctions[i]);
hashtable->hashStrict[i] = op_strict(hashop);
i++;
}
/*
* Allocate data that will live for the life of the hashjoin
*/
oldcxt = MemoryContextSwitchTo(hashtable->hashCxt);
#ifdef HJDEBUG
{
/* Memory needed to allocate hashtable->batches, which consists of nbatch pointers */
int md_batch_size = (nbatch * sizeof(hashtable->batches[0])) / (1024 * 1024);
/* Memory needed to allocate hashtable->batches entries, which consist of nbatch HashJoinBatchData structures */
int md_batch_data_size = (nbatch * sizeof(HashJoinBatchData)) / (1024 * 1024);
/* Memory needed to allocate hashtable->buckets, which consists of nbuckets HashJoinTuple structures*/
int md_buckets_size = (nbuckets * sizeof(HashJoinTuple)) / (1024 * 1024);
/* Memory needed to allocate hashtable->bloom, which consists of nbuckets int64 values */
int md_bloom_size = (nbuckets * sizeof(uint64)) / (1024 * 1024);
/* Total memory needed for the hashtable metadata */
int md_tot = md_batch_size + md_batch_data_size + md_buckets_size + md_bloom_size;
elog(LOG, "About to allocate HashTable. HT_MEMORY=%dMB Memory needed for metadata: MDBATCH_ARR=%dMB, MDBATCH_DATA=%dMB, MDBUCKETS_ARR=%dMB, MDBLOOM_ARR=%dMB, TOTAL=%dMB",
(int) (hashtable->spaceAllowed / (1024 * 1024)),
md_batch_size, md_batch_data_size, md_buckets_size, md_bloom_size, md_tot);
elog(LOG, "sizeof(hashtable->batches[0])=%d, sizeof(HashJoinBatchData)=%d, sizeof(HashJoinTuple)=%d, sizeof(uint64)=%d",
(int) sizeof(hashtable->batches[0]), (int) sizeof(HashJoinBatchData),
(int) sizeof(HashJoinTuple), (int) sizeof(uint64));
}
#endif
/* array of BatchData ptrs */
hashtable->batches =
(HashJoinBatchData **)palloc(nbatch * sizeof(hashtable->batches[0]));
/* one BatchData entry per initial batch */
for (i = 0; i < nbatch; i++)
hashtable->batches[i] =
(HashJoinBatchData *)palloc0(sizeof(HashJoinBatchData));
/*
* Prepare context for the first-scan space allocations; allocate the
* hashbucket array therein, and set each bucket "empty".
*/
MemoryContextSwitchTo(hashtable->batchCxt);
hashtable->buckets = (HashJoinTuple *)
palloc0(nbuckets * sizeof(HashJoinTuple));
if(gp_hashjoin_bloomfilter!=0)
hashtable->bloom = (uint64*) palloc0(nbuckets * sizeof(uint64));
MemoryContextSwitchTo(oldcxt);
}
/*
* Build an SP-GiST index.
*/
Datum
spgbuild(PG_FUNCTION_ARGS)
{
Relation heap = (Relation) PG_GETARG_POINTER(0);
Relation index = (Relation) PG_GETARG_POINTER(1);
IndexInfo *indexInfo = (IndexInfo *) PG_GETARG_POINTER(2);
IndexBuildResult *result;
double reltuples;
SpGistBuildState buildstate;
Buffer metabuffer,
rootbuffer;
if (RelationGetNumberOfBlocks(index) != 0)
elog(ERROR, "index \"%s\" already contains data",
RelationGetRelationName(index));
/*
* Initialize the meta page and root page
*/
metabuffer = SpGistNewBuffer(index);
rootbuffer = SpGistNewBuffer(index);
Assert(BufferGetBlockNumber(metabuffer) == SPGIST_METAPAGE_BLKNO);
Assert(BufferGetBlockNumber(rootbuffer) == SPGIST_HEAD_BLKNO);
START_CRIT_SECTION();
SpGistInitMetapage(BufferGetPage(metabuffer));
MarkBufferDirty(metabuffer);
SpGistInitBuffer(rootbuffer, SPGIST_LEAF);
MarkBufferDirty(rootbuffer);
if (RelationNeedsWAL(index))
{
XLogRecPtr recptr;
XLogRecData rdata;
/* WAL data is just the relfilenode */
rdata.data = (char *) &(index->rd_node);
rdata.len = sizeof(RelFileNode);
rdata.buffer = InvalidBuffer;
rdata.next = NULL;
recptr = XLogInsert(RM_SPGIST_ID, XLOG_SPGIST_CREATE_INDEX, &rdata);
PageSetLSN(BufferGetPage(metabuffer), recptr);
PageSetTLI(BufferGetPage(metabuffer), ThisTimeLineID);
PageSetLSN(BufferGetPage(rootbuffer), recptr);
PageSetTLI(BufferGetPage(rootbuffer), ThisTimeLineID);
}
END_CRIT_SECTION();
UnlockReleaseBuffer(metabuffer);
UnlockReleaseBuffer(rootbuffer);
/*
* Now insert all the heap data into the index
*/
initSpGistState(&buildstate.spgstate, index);
buildstate.spgstate.isBuild = true;
buildstate.tmpCtx = AllocSetContextCreate(CurrentMemoryContext,
"SP-GiST build temporary context",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
reltuples = IndexBuildHeapScan(heap, index, indexInfo, true,
spgistBuildCallback, (void *) &buildstate);
MemoryContextDelete(buildstate.tmpCtx);
SpGistUpdateMetaPage(index);
result = (IndexBuildResult *) palloc0(sizeof(IndexBuildResult));
result->heap_tuples = result->index_tuples = reltuples;
PG_RETURN_POINTER(result);
}
/*
* CopyIntoCStoreTable handles a "COPY cstore_table FROM" statement. This
* function uses the COPY command's functions to read and parse rows from
* the data source specified in the COPY statement. The function then writes
* each row to the file specified in the cstore foreign table options. Finally,
* the function returns the number of copied rows.
*/
static uint64
CopyIntoCStoreTable(const CopyStmt *copyStatement, const char *queryString)
{
uint64 processedRowCount = 0;
Relation relation = NULL;
Oid relationId = InvalidOid;
TupleDesc tupleDescriptor = NULL;
uint32 columnCount = 0;
CopyState copyState = NULL;
bool nextRowFound = true;
Datum *columnValues = NULL;
bool *columnNulls = NULL;
TableWriteState *writeState = NULL;
CStoreFdwOptions *cstoreFdwOptions = NULL;
MemoryContext tupleContext = NULL;
List *columnNameList = copyStatement->attlist;
if (columnNameList != NULL)
{
ereport(ERROR, (errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("copy column list is not supported")));
}
/*
* We disallow copy from file or program except to superusers. These checks
* are based on the checks in DoCopy() function of copy.c.
*/
if (copyStatement->filename != NULL && !superuser())
{
if (copyStatement->is_program)
{
ereport(ERROR, (errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
errmsg("must be superuser to COPY to or from a program"),
errhint("Anyone can COPY to stdout or from stdin. "
"psql's \\copy command also works for anyone.")));
}
else
{
ereport(ERROR, (errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
errmsg("must be superuser to COPY to or from a file"),
errhint("Anyone can COPY to stdout or from stdin. "
"psql's \\copy command also works for anyone.")));
}
}
Assert(copyStatement->relation != NULL);
/*
* Open and lock the relation. We acquire ExclusiveLock to allow concurrent
* reads, but block concurrent writes.
*/
relation = heap_openrv(copyStatement->relation, ExclusiveLock);
relationId = RelationGetRelid(relation);
/* allocate column values and nulls arrays */
tupleDescriptor = RelationGetDescr(relation);
columnCount = tupleDescriptor->natts;
columnValues = palloc0(columnCount * sizeof(Datum));
columnNulls = palloc0(columnCount * sizeof(bool));
cstoreFdwOptions = CStoreGetOptions(relationId);
/*
* We create a new memory context called tuple context, and read and write
* each row's values within this memory context. After each read and write,
* we reset the memory context. That way, we immediately release memory
* allocated for each row, and don't bloat memory usage with large input
* files.
*/
tupleContext = AllocSetContextCreate(CurrentMemoryContext,
"CStore COPY Row Memory Context",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
/* init state to read from COPY data source */
copyState = BeginCopyFrom(relation, copyStatement->filename,
copyStatement->is_program, NIL,
copyStatement->options);
/* init state to write to the cstore file */
writeState = CStoreBeginWrite(cstoreFdwOptions->filename,
cstoreFdwOptions->compressionType,
cstoreFdwOptions->stripeRowCount,
cstoreFdwOptions->blockRowCount,
tupleDescriptor);
while (nextRowFound)
{
/* read the next row in tupleContext */
MemoryContext oldContext = MemoryContextSwitchTo(tupleContext);
nextRowFound = NextCopyFrom(copyState, NULL, columnValues, columnNulls, NULL);
MemoryContextSwitchTo(oldContext);
/* write the row to the cstore file */
if (nextRowFound)
{
CStoreWriteRow(writeState, columnValues, columnNulls);
processedRowCount++;
}
MemoryContextReset(tupleContext);
}
/* end read/write sessions and close the relation */
EndCopyFrom(copyState);
CStoreEndWrite(writeState);
heap_close(relation, ExclusiveLock);
return processedRowCount;
}
/*
* Execute the index scan.
*
* This works by reading index TIDs from the revmap, and obtaining the index
* tuples pointed to by them; the summary values in the index tuples are
* compared to the scan keys. We return into the TID bitmap all the pages in
* ranges corresponding to index tuples that match the scan keys.
*
* If a TID from the revmap is read as InvalidTID, we know that range is
* unsummarized. Pages in those ranges need to be returned regardless of scan
* keys.
*/
int64
bringetbitmap(IndexScanDesc scan, TIDBitmap *tbm)
{
Relation idxRel = scan->indexRelation;
Buffer buf = InvalidBuffer;
BrinDesc *bdesc;
Oid heapOid;
Relation heapRel;
BrinOpaque *opaque;
BlockNumber nblocks;
BlockNumber heapBlk;
int totalpages = 0;
FmgrInfo *consistentFn;
MemoryContext oldcxt;
MemoryContext perRangeCxt;
opaque = (BrinOpaque *) scan->opaque;
bdesc = opaque->bo_bdesc;
pgstat_count_index_scan(idxRel);
/*
* We need to know the size of the table so that we know how long to
* iterate on the revmap.
*/
heapOid = IndexGetRelation(RelationGetRelid(idxRel), false);
heapRel = heap_open(heapOid, AccessShareLock);
nblocks = RelationGetNumberOfBlocks(heapRel);
heap_close(heapRel, AccessShareLock);
/*
* Make room for the consistent support procedures of indexed columns. We
* don't look them up here; we do that lazily the first time we see a scan
* key reference each of them. We rely on zeroing fn_oid to InvalidOid.
*/
consistentFn = palloc0(sizeof(FmgrInfo) * bdesc->bd_tupdesc->natts);
/*
* Setup and use a per-range memory context, which is reset every time we
* loop below. This avoids having to free the tuples within the loop.
*/
perRangeCxt = AllocSetContextCreate(CurrentMemoryContext,
"bringetbitmap cxt",
ALLOCSET_DEFAULT_SIZES);
oldcxt = MemoryContextSwitchTo(perRangeCxt);
/*
* Now scan the revmap. We start by querying for heap page 0,
* incrementing by the number of pages per range; this gives us a full
* view of the table.
*/
for (heapBlk = 0; heapBlk < nblocks; heapBlk += opaque->bo_pagesPerRange)
{
bool addrange;
BrinTuple *tup;
OffsetNumber off;
Size size;
CHECK_FOR_INTERRUPTS();
MemoryContextResetAndDeleteChildren(perRangeCxt);
tup = brinGetTupleForHeapBlock(opaque->bo_rmAccess, heapBlk, &buf,
&off, &size, BUFFER_LOCK_SHARE,
scan->xs_snapshot);
if (tup)
{
tup = brin_copy_tuple(tup, size);
LockBuffer(buf, BUFFER_LOCK_UNLOCK);
}
/*
* For page ranges with no indexed tuple, we must return the whole
* range; otherwise, compare it to the scan keys.
*/
if (tup == NULL)
{
addrange = true;
}
else
{
BrinMemTuple *dtup;
dtup = brin_deform_tuple(bdesc, tup);
if (dtup->bt_placeholder)
{
/*
* Placeholder tuples are always returned, regardless of the
* values stored in them.
*/
addrange = true;
}
else
{
int keyno;
/*
* Compare scan keys with summary values stored for the range.
* If scan keys are matched, the page range must be added to
* the bitmap. We initially assume the range needs to be
* added; in particular this serves the case where there are
* no keys.
*/
addrange = true;
for (keyno = 0; keyno < scan->numberOfKeys; keyno++)
{
ScanKey key = &scan->keyData[keyno];
AttrNumber keyattno = key->sk_attno;
BrinValues *bval = &dtup->bt_columns[keyattno - 1];
Datum add;
/*
* The collation of the scan key must match the collation
* used in the index column (but only if the search is not
* IS NULL/ IS NOT NULL). Otherwise we shouldn't be using
* this index ...
*/
Assert((key->sk_flags & SK_ISNULL) ||
(key->sk_collation ==
bdesc->bd_tupdesc->attrs[keyattno - 1]->attcollation));
/* First time this column? look up consistent function */
if (consistentFn[keyattno - 1].fn_oid == InvalidOid)
{
FmgrInfo *tmp;
tmp = index_getprocinfo(idxRel, keyattno,
BRIN_PROCNUM_CONSISTENT);
fmgr_info_copy(&consistentFn[keyattno - 1], tmp,
CurrentMemoryContext);
}
/*
* Check whether the scan key is consistent with the page
* range values; if so, have the pages in the range added
* to the output bitmap.
*
* When there are multiple scan keys, failure to meet the
* criteria for a single one of them is enough to discard
* the range as a whole, so break out of the loop as soon
* as a false return value is obtained.
*/
add = FunctionCall3Coll(&consistentFn[keyattno - 1],
key->sk_collation,
PointerGetDatum(bdesc),
PointerGetDatum(bval),
PointerGetDatum(key));
addrange = DatumGetBool(add);
if (!addrange)
break;
}
}
}
/* add the pages in the range to the output bitmap, if needed */
if (addrange)
{
BlockNumber pageno;
for (pageno = heapBlk;
pageno <= heapBlk + opaque->bo_pagesPerRange - 1;
pageno++)
{
MemoryContextSwitchTo(oldcxt);
tbm_add_page(tbm, pageno);
totalpages++;
MemoryContextSwitchTo(perRangeCxt);
}
}
}
MemoryContextSwitchTo(oldcxt);
MemoryContextDelete(perRangeCxt);
if (buf != InvalidBuffer)
ReleaseBuffer(buf);
/*
* XXX We have an approximation of the number of *pages* that our scan
* returns, but we don't have a precise idea of the number of heap tuples
* involved.
*/
return totalpages * 10;
}
/*
* init_MultiFuncCall
* Create an empty FuncCallContext data structure
* and do some other basic Multi-function call setup
* and error checking
*/
FuncCallContext *
init_MultiFuncCall(PG_FUNCTION_ARGS)
{
FuncCallContext *retval;
/*
* Bail if we're called in the wrong context
*/
if (fcinfo->resultinfo == NULL || !IsA(fcinfo->resultinfo, ReturnSetInfo))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("set-valued function called in context that cannot accept a set")));
if (fcinfo->flinfo->fn_extra == NULL)
{
/*
* First call
*/
ReturnSetInfo *rsi = (ReturnSetInfo *) fcinfo->resultinfo;
MemoryContext multi_call_ctx;
/*
* Create a suitably long-lived context to hold cross-call data
*/
multi_call_ctx = AllocSetContextCreate(fcinfo->flinfo->fn_mcxt,
"SRF multi-call context",
ALLOCSET_SMALL_MINSIZE,
ALLOCSET_SMALL_INITSIZE,
ALLOCSET_SMALL_MAXSIZE);
/*
* Allocate suitably long-lived space and zero it
*/
retval = (FuncCallContext *)
MemoryContextAllocZero(multi_call_ctx,
sizeof(FuncCallContext));
/*
* initialize the elements
*/
retval->call_cntr = 0;
retval->max_calls = 0;
retval->slot = NULL;
retval->user_fctx = NULL;
retval->attinmeta = NULL;
retval->tuple_desc = NULL;
retval->multi_call_memory_ctx = multi_call_ctx;
/*
* save the pointer for cross-call use
*/
fcinfo->flinfo->fn_extra = retval;
/*
* Ensure we will get shut down cleanly if the exprcontext is not run
* to completion.
*/
RegisterExprContextCallback(rsi->econtext,
shutdown_MultiFuncCall,
PointerGetDatum(fcinfo->flinfo));
}
else
{
/* second and subsequent calls */
elog(ERROR, "init_MultiFuncCall cannot be called more than once");
/* never reached, but keep compiler happy */
retval = NULL;
}
return retval;
}
/*
* Main entry point for checkpointer process
*
* This is invoked from AuxiliaryProcessMain, which has already created the
* basic execution environment, but not enabled signals yet.
*/
void
CheckpointerMain(void)
{
sigjmp_buf local_sigjmp_buf;
MemoryContext checkpointer_context;
CheckpointerShmem->checkpointer_pid = MyProcPid;
/*
* If possible, make this process a group leader, so that the postmaster
* can signal any child processes too. (checkpointer probably never has
* any child processes, but for consistency we make all postmaster child
* processes do this.)
*/
#ifdef HAVE_SETSID
if (setsid() < 0)
elog(FATAL, "setsid() failed: %m");
#endif
/*
* Properly accept or ignore signals the postmaster might send us
*
* Note: we deliberately ignore SIGTERM, because during a standard Unix
* system shutdown cycle, init will SIGTERM all processes at once. We
* want to wait for the backends to exit, whereupon the postmaster will
* tell us it's okay to shut down (via SIGUSR2).
*/
pqsignal(SIGHUP, ChkptSigHupHandler); /* set flag to read config
* file */
pqsignal(SIGINT, ReqCheckpointHandler); /* request checkpoint */
pqsignal(SIGTERM, SIG_IGN); /* ignore SIGTERM */
pqsignal(SIGQUIT, chkpt_quickdie); /* hard crash time */
pqsignal(SIGALRM, SIG_IGN);
pqsignal(SIGPIPE, SIG_IGN);
pqsignal(SIGUSR1, chkpt_sigusr1_handler);
pqsignal(SIGUSR2, ReqShutdownHandler); /* request shutdown */
/*
* Reset some signals that are accepted by postmaster but not here
*/
pqsignal(SIGCHLD, SIG_DFL);
pqsignal(SIGTTIN, SIG_DFL);
pqsignal(SIGTTOU, SIG_DFL);
pqsignal(SIGCONT, SIG_DFL);
pqsignal(SIGWINCH, SIG_DFL);
/* We allow SIGQUIT (quickdie) at all times */
sigdelset(&BlockSig, SIGQUIT);
/*
* Initialize so that first time-driven event happens at the correct time.
*/
last_checkpoint_time = last_xlog_switch_time = (pg_time_t) time(NULL);
/*
* Create a resource owner to keep track of our resources (currently only
* buffer pins).
*/
CurrentResourceOwner = ResourceOwnerCreate(NULL, "Checkpointer");
/*
* Create a memory context that we will do all our work in. We do this so
* that we can reset the context during error recovery and thereby avoid
* possible memory leaks. Formerly this code just ran in
* TopMemoryContext, but resetting that would be a really bad idea.
*/
checkpointer_context = AllocSetContextCreate(TopMemoryContext,
"Checkpointer",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
MemoryContextSwitchTo(checkpointer_context);
/*
* If an exception is encountered, processing resumes here.
*
* See notes in postgres.c about the design of this coding.
*/
if (sigsetjmp(local_sigjmp_buf, 1) != 0)
{
/* Since not using PG_TRY, must reset error stack by hand */
error_context_stack = NULL;
/* Prevent interrupts while cleaning up */
HOLD_INTERRUPTS();
/* Report the error to the server log */
EmitErrorReport();
/*
* These operations are really just a minimal subset of
* AbortTransaction(). We don't have very many resources to worry
* about in checkpointer, but we do have LWLocks, buffers, and temp
* files.
*/
LWLockReleaseAll();
AbortBufferIO();
UnlockBuffers();
/* buffer pins are released here: */
ResourceOwnerRelease(CurrentResourceOwner,
RESOURCE_RELEASE_BEFORE_LOCKS,
false, true);
/* we needn't bother with the other ResourceOwnerRelease phases */
AtEOXact_Buffers(false);
AtEOXact_SMgr();
AtEOXact_Files();
AtEOXact_HashTables(false);
/* Warn any waiting backends that the checkpoint failed. */
if (ckpt_active)
{
/* use volatile pointer to prevent code rearrangement */
volatile CheckpointerShmemStruct *cps = CheckpointerShmem;
SpinLockAcquire(&cps->ckpt_lck);
cps->ckpt_failed++;
cps->ckpt_done = cps->ckpt_started;
SpinLockRelease(&cps->ckpt_lck);
ckpt_active = false;
}
/*
* Now return to normal top-level context and clear ErrorContext for
* next time.
*/
MemoryContextSwitchTo(checkpointer_context);
FlushErrorState();
/* Flush any leaked data in the top-level context */
MemoryContextResetAndDeleteChildren(checkpointer_context);
/* Now we can allow interrupts again */
RESUME_INTERRUPTS();
/*
* Sleep at least 1 second after any error. A write error is likely
* to be repeated, and we don't want to be filling the error logs as
* fast as we can.
*/
pg_usleep(1000000L);
/*
* Close all open files after any error. This is helpful on Windows,
* where holding deleted files open causes various strange errors.
* It's not clear we need it elsewhere, but shouldn't hurt.
*/
smgrcloseall();
}
/* We can now handle ereport(ERROR) */
PG_exception_stack = &local_sigjmp_buf;
/*
* Unblock signals (they were blocked when the postmaster forked us)
*/
PG_SETMASK(&UnBlockSig);
/*
* Ensure all shared memory values are set correctly for the config. Doing
* this here ensures no race conditions from other concurrent updaters.
*/
UpdateSharedMemoryConfig();
/*
* Advertise our latch that backends can use to wake us up while we're
* sleeping.
*/
ProcGlobal->checkpointerLatch = &MyProc->procLatch;
/*
* Loop forever
*/
for (;;)
{
bool do_checkpoint = false;
int flags = 0;
pg_time_t now;
int elapsed_secs;
int cur_timeout;
int rc;
/* Clear any already-pending wakeups */
ResetLatch(&MyProc->procLatch);
/*
* Process any requests or signals received recently.
*/
AbsorbFsyncRequests();
if (got_SIGHUP)
{
got_SIGHUP = false;
ProcessConfigFile(PGC_SIGHUP);
/*
* Checkpointer is the last process to shut down, so we ask it to
* hold the keys for a range of other tasks required most of which
* have nothing to do with checkpointing at all.
*
* For various reasons, some config values can change dynamically
* so the primary copy of them is held in shared memory to make
* sure all backends see the same value. We make Checkpointer
* responsible for updating the shared memory copy if the
* parameter setting changes because of SIGHUP.
*/
UpdateSharedMemoryConfig();
}
if (checkpoint_requested)
{
checkpoint_requested = false;
do_checkpoint = true;
BgWriterStats.m_requested_checkpoints++;
}
if (shutdown_requested)
{
/*
* From here on, elog(ERROR) should end with exit(1), not send
* control back to the sigsetjmp block above
*/
ExitOnAnyError = true;
/* Close down the database */
ShutdownXLOG(0, 0);
/* Normal exit from the checkpointer is here */
proc_exit(0); /* done */
}
/*
* Force a checkpoint if too much time has elapsed since the last one.
* Note that we count a timed checkpoint in stats only when this
* occurs without an external request, but we set the CAUSE_TIME flag
* bit even if there is also an external request.
*/
now = (pg_time_t) time(NULL);
elapsed_secs = now - last_checkpoint_time;
if (elapsed_secs >= CheckPointTimeout)
{
if (!do_checkpoint)
BgWriterStats.m_timed_checkpoints++;
do_checkpoint = true;
flags |= CHECKPOINT_CAUSE_TIME;
}
/*
* Do a checkpoint if requested.
*/
if (do_checkpoint)
{
bool ckpt_performed = false;
bool do_restartpoint;
/* use volatile pointer to prevent code rearrangement */
volatile CheckpointerShmemStruct *cps = CheckpointerShmem;
/*
* Check if we should perform a checkpoint or a restartpoint. As a
* side-effect, RecoveryInProgress() initializes TimeLineID if
* it's not set yet.
*/
do_restartpoint = RecoveryInProgress();
/*
* Atomically fetch the request flags to figure out what kind of a
* checkpoint we should perform, and increase the started-counter
* to acknowledge that we've started a new checkpoint.
*/
SpinLockAcquire(&cps->ckpt_lck);
flags |= cps->ckpt_flags;
cps->ckpt_flags = 0;
cps->ckpt_started++;
SpinLockRelease(&cps->ckpt_lck);
/*
* The end-of-recovery checkpoint is a real checkpoint that's
* performed while we're still in recovery.
*/
if (flags & CHECKPOINT_END_OF_RECOVERY)
do_restartpoint = false;
/*
* We will warn if (a) too soon since last checkpoint (whatever
* caused it) and (b) somebody set the CHECKPOINT_CAUSE_XLOG flag
* since the last checkpoint start. Note in particular that this
* implementation will not generate warnings caused by
* CheckPointTimeout < CheckPointWarning.
*/
if (!do_restartpoint &&
(flags & CHECKPOINT_CAUSE_XLOG) &&
elapsed_secs < CheckPointWarning)
ereport(LOG,
(errmsg_plural("checkpoints are occurring too frequently (%d second apart)",
"checkpoints are occurring too frequently (%d seconds apart)",
elapsed_secs,
elapsed_secs),
errhint("Consider increasing the configuration parameter \"checkpoint_segments\".")));
/*
* Initialize checkpointer-private variables used during
* checkpoint
*/
ckpt_active = true;
if (!do_restartpoint)
ckpt_start_recptr = GetInsertRecPtr();
ckpt_start_time = now;
ckpt_cached_elapsed = 0;
/*
* Do the checkpoint.
*/
if (!do_restartpoint)
{
CreateCheckPoint(flags);
ckpt_performed = true;
}
else
ckpt_performed = CreateRestartPoint(flags);
/*
* After any checkpoint, close all smgr files. This is so we
* won't hang onto smgr references to deleted files indefinitely.
*/
smgrcloseall();
/*
* Indicate checkpoint completion to any waiting backends.
*/
SpinLockAcquire(&cps->ckpt_lck);
cps->ckpt_done = cps->ckpt_started;
SpinLockRelease(&cps->ckpt_lck);
if (ckpt_performed)
{
/*
* Note we record the checkpoint start time not end time as
* last_checkpoint_time. This is so that time-driven
* checkpoints happen at a predictable spacing.
*/
last_checkpoint_time = now;
}
else
{
/*
* We were not able to perform the restartpoint (checkpoints
* throw an ERROR in case of error). Most likely because we
* have not received any new checkpoint WAL records since the
* last restartpoint. Try again in 15 s.
*/
last_checkpoint_time = now - CheckPointTimeout + 15;
}
ckpt_active = false;
}
/* Check for archive_timeout and switch xlog files if necessary. */
CheckArchiveTimeout();
/*
* Send off activity statistics to the stats collector. (The reason
* why we re-use bgwriter-related code for this is that the bgwriter
* and checkpointer used to be just one process. It's probably not
* worth the trouble to split the stats support into two independent
* stats message types.)
*/
pgstat_send_bgwriter();
/*
* Sleep until we are signaled or it's time for another checkpoint or
* xlog file switch.
*/
now = (pg_time_t) time(NULL);
elapsed_secs = now - last_checkpoint_time;
if (elapsed_secs >= CheckPointTimeout)
continue; /* no sleep for us ... */
cur_timeout = CheckPointTimeout - elapsed_secs;
if (XLogArchiveTimeout > 0 && !RecoveryInProgress())
{
elapsed_secs = now - last_xlog_switch_time;
if (elapsed_secs >= XLogArchiveTimeout)
continue; /* no sleep for us ... */
cur_timeout = Min(cur_timeout, XLogArchiveTimeout - elapsed_secs);
}
rc = WaitLatch(&MyProc->procLatch,
WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH,
cur_timeout * 1000L /* convert to ms */ );
/*
* Emergency bailout if postmaster has died. This is to avoid the
* necessity for manual cleanup of all postmaster children.
*/
if (rc & WL_POSTMASTER_DEATH)
exit(1);
}
}
int
SPI_connect(void)
{
int newdepth;
/*
* When procedure called by Executor _SPI_curid expected to be equal
* to _SPI_connected
*/
if (_SPI_curid != _SPI_connected)
return SPI_ERROR_CONNECT;
if (_SPI_stack == NULL)
{
if (_SPI_connected != -1 || _SPI_stack_depth != 0)
elog(ERROR, "SPI stack corrupted");
newdepth = 16;
_SPI_stack = (_SPI_connection *)
MemoryContextAlloc(TopTransactionContext,
newdepth * sizeof(_SPI_connection));
_SPI_stack_depth = newdepth;
}
else
{
if (_SPI_stack_depth <= 0 || _SPI_stack_depth <= _SPI_connected)
elog(ERROR, "SPI stack corrupted");
if (_SPI_stack_depth == _SPI_connected + 1)
{
newdepth = _SPI_stack_depth * 2;
_SPI_stack = (_SPI_connection *)
repalloc(_SPI_stack,
newdepth * sizeof(_SPI_connection));
_SPI_stack_depth = newdepth;
}
}
/*
* We're entering procedure where _SPI_curid == _SPI_connected - 1
*/
_SPI_connected++;
Assert(_SPI_connected >= 0 && _SPI_connected < _SPI_stack_depth);
_SPI_current = &(_SPI_stack[_SPI_connected]);
_SPI_current->processed = 0;
_SPI_current->tuptable = NULL;
_SPI_current->procCxt = NULL; /* in case we fail to create 'em */
_SPI_current->execCxt = NULL;
_SPI_current->connectSubid = GetCurrentSubTransactionId();
/*
* Create memory contexts for this procedure
*
* XXX it would be better to use PortalContext as the parent context, but
* we may not be inside a portal (consider deferred-trigger
* execution). Perhaps CurTransactionContext would do? For now it
* doesn't matter because we clean up explicitly in AtEOSubXact_SPI().
*/
_SPI_current->procCxt = AllocSetContextCreate(TopTransactionContext,
"SPI Proc",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
_SPI_current->execCxt = AllocSetContextCreate(TopTransactionContext,
"SPI Exec",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
/* ... and switch to procedure's context */
_SPI_current->savedcxt = MemoryContextSwitchTo(_SPI_current->procCxt);
return SPI_OK_CONNECT;
}
Datum
gistrescan(PG_FUNCTION_ARGS)
{
IndexScanDesc scan = (IndexScanDesc) PG_GETARG_POINTER(0);
ScanKey key = (ScanKey) PG_GETARG_POINTER(1);
ScanKey orderbys = (ScanKey) PG_GETARG_POINTER(3);
/* nkeys and norderbys arguments are ignored */
GISTScanOpaque so = (GISTScanOpaque) scan->opaque;
bool first_time;
int i;
MemoryContext oldCxt;
/* rescan an existing indexscan --- reset state */
/*
* The first time through, we create the search queue in the scanCxt.
* Subsequent times through, we create the queue in a separate queueCxt,
* which is created on the second call and reset on later calls. Thus, in
* the common case where a scan is only rescan'd once, we just put the
* queue in scanCxt and don't pay the overhead of making a second memory
* context. If we do rescan more than once, the first RBTree is just left
* for dead until end of scan; this small wastage seems worth the savings
* in the common case.
*/
if (so->queue == NULL)
{
/* first time through */
Assert(so->queueCxt == so->giststate->scanCxt);
first_time = true;
}
else if (so->queueCxt == so->giststate->scanCxt)
{
/* second time through */
so->queueCxt = AllocSetContextCreate(so->giststate->scanCxt,
"GiST queue context",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
first_time = false;
}
else
{
/* third or later time through */
MemoryContextReset(so->queueCxt);
first_time = false;
}
/* create new, empty RBTree for search queue */
oldCxt = MemoryContextSwitchTo(so->queueCxt);
so->queue = pairingheap_allocate(pairingheap_GISTSearchItem_cmp, scan);
MemoryContextSwitchTo(oldCxt);
so->firstCall = true;
/* Update scan key, if a new one is given */
if (key && scan->numberOfKeys > 0)
{
void **fn_extras = NULL;
/*
* If this isn't the first time through, preserve the fn_extra
* pointers, so that if the consistentFns are using them to cache
* data, that data is not leaked across a rescan.
*/
if (!first_time)
{
fn_extras = (void **) palloc(scan->numberOfKeys * sizeof(void *));
for (i = 0; i < scan->numberOfKeys; i++)
fn_extras[i] = scan->keyData[i].sk_func.fn_extra;
}
memmove(scan->keyData, key,
scan->numberOfKeys * sizeof(ScanKeyData));
/*
* Modify the scan key so that the Consistent method is called for all
* comparisons. The original operator is passed to the Consistent
* function in the form of its strategy number, which is available
* from the sk_strategy field, and its subtype from the sk_subtype
* field.
*
* Next, if any of keys is a NULL and that key is not marked with
* SK_SEARCHNULL/SK_SEARCHNOTNULL then nothing can be found (ie, we
* assume all indexable operators are strict).
*/
so->qual_ok = true;
for (i = 0; i < scan->numberOfKeys; i++)
{
ScanKey skey = scan->keyData + i;
fmgr_info_copy(&(skey->sk_func),
&(so->giststate->consistentFn[skey->sk_attno - 1]),
so->giststate->scanCxt);
/* Restore prior fn_extra pointers, if not first time */
if (!first_time)
skey->sk_func.fn_extra = fn_extras[i];
if (skey->sk_flags & SK_ISNULL)
{
if (!(skey->sk_flags & (SK_SEARCHNULL | SK_SEARCHNOTNULL)))
so->qual_ok = false;
}
}
if (!first_time)
pfree(fn_extras);
}
/* Update order-by key, if a new one is given */
if (orderbys && scan->numberOfOrderBys > 0)
{
void **fn_extras = NULL;
/* As above, preserve fn_extra if not first time through */
if (!first_time)
{
fn_extras = (void **) palloc(scan->numberOfOrderBys * sizeof(void *));
for (i = 0; i < scan->numberOfOrderBys; i++)
fn_extras[i] = scan->orderByData[i].sk_func.fn_extra;
}
memmove(scan->orderByData, orderbys,
scan->numberOfOrderBys * sizeof(ScanKeyData));
/*
* Modify the order-by key so that the Distance method is called for
* all comparisons. The original operator is passed to the Distance
* function in the form of its strategy number, which is available
* from the sk_strategy field, and its subtype from the sk_subtype
* field.
*/
for (i = 0; i < scan->numberOfOrderBys; i++)
{
ScanKey skey = scan->orderByData + i;
FmgrInfo *finfo = &(so->giststate->distanceFn[skey->sk_attno - 1]);
/* Check we actually have a distance function ... */
if (!OidIsValid(finfo->fn_oid))
elog(ERROR, "missing support function %d for attribute %d of index \"%s\"",
GIST_DISTANCE_PROC, skey->sk_attno,
RelationGetRelationName(scan->indexRelation));
fmgr_info_copy(&(skey->sk_func), finfo, so->giststate->scanCxt);
/* Restore prior fn_extra pointers, if not first time */
if (!first_time)
skey->sk_func.fn_extra = fn_extras[i];
}
if (!first_time)
pfree(fn_extras);
}
PG_RETURN_VOID();
}
/*
* btvacuumscan --- scan the index for VACUUMing purposes
*
* This combines the functions of looking for leaf tuples that are deletable
* according to the vacuum callback, looking for empty pages that can be
* deleted, and looking for old deleted pages that can be recycled. Both
* btbulkdelete and btvacuumcleanup invoke this (the latter only if no
* btbulkdelete call occurred).
*
* The caller is responsible for initially allocating/zeroing a stats struct
* and for obtaining a vacuum cycle ID if necessary.
*/
static void
btvacuumscan(IndexVacuumInfo *info, IndexBulkDeleteResult *stats,
IndexBulkDeleteCallback callback, void *callback_state,
BTCycleId cycleid)
{
Relation rel = info->index;
BTVacState vstate;
BlockNumber num_pages;
BlockNumber blkno;
bool needLock;
/*
* Reset counts that will be incremented during the scan; needed in case
* of multiple scans during a single VACUUM command
*/
stats->estimated_count = false;
stats->num_index_tuples = 0;
stats->pages_deleted = 0;
/* Set up info to pass down to btvacuumpage */
vstate.info = info;
vstate.stats = stats;
vstate.callback = callback;
vstate.callback_state = callback_state;
vstate.cycleid = cycleid;
vstate.lastBlockVacuumed = BTREE_METAPAGE; /* Initialise at first block */
vstate.lastUsedPage = BTREE_METAPAGE;
vstate.totFreePages = 0;
/* Create a temporary memory context to run _bt_pagedel in */
vstate.pagedelcontext = AllocSetContextCreate(CurrentMemoryContext,
"_bt_pagedel",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
/*
* The outer loop iterates over all index pages except the metapage, in
* physical order (we hope the kernel will cooperate in providing
* read-ahead for speed). It is critical that we visit all leaf pages,
* including ones added after we start the scan, else we might fail to
* delete some deletable tuples. Hence, we must repeatedly check the
* relation length. We must acquire the relation-extension lock while
* doing so to avoid a race condition: if someone else is extending the
* relation, there is a window where bufmgr/smgr have created a new
* all-zero page but it hasn't yet been write-locked by _bt_getbuf(). If
* we manage to scan such a page here, we'll improperly assume it can be
* recycled. Taking the lock synchronizes things enough to prevent a
* problem: either num_pages won't include the new page, or _bt_getbuf
* already has write lock on the buffer and it will be fully initialized
* before we can examine it. (See also vacuumlazy.c, which has the same
* issue.) Also, we need not worry if a page is added immediately after
* we look; the page splitting code already has write-lock on the left
* page before it adds a right page, so we must already have processed any
* tuples due to be moved into such a page.
*
* We can skip locking for new or temp relations, however, since no one
* else could be accessing them.
*/
needLock = !RELATION_IS_LOCAL(rel);
blkno = BTREE_METAPAGE + 1;
for (;;)
{
/* Get the current relation length */
if (needLock)
LockRelationForExtension(rel, ExclusiveLock);
num_pages = RelationGetNumberOfBlocks(rel);
if (needLock)
UnlockRelationForExtension(rel, ExclusiveLock);
/* Quit if we've scanned the whole relation */
if (blkno >= num_pages)
break;
/* Iterate over pages, then loop back to recheck length */
for (; blkno < num_pages; blkno++)
{
btvacuumpage(&vstate, blkno, blkno);
}
}
/*
* InHotStandby we need to scan right up to the end of the index for
* correct locking, so we may need to write a WAL record for the final
* block in the index if it was not vacuumed. It's possible that VACUUMing
* has actually removed zeroed pages at the end of the index so we need to
* take care to issue the record for last actual block and not for the
* last block that was scanned. Ignore empty indexes.
*/
if (XLogStandbyInfoActive() &&
num_pages > 1 && vstate.lastBlockVacuumed < (num_pages - 1))
{
Buffer buf;
/*
* We can't use _bt_getbuf() here because it always applies
* _bt_checkpage(), which will barf on an all-zero page. We want to
* recycle all-zero pages, not fail. Also, we want to use a
* nondefault buffer access strategy.
*/
buf = ReadBufferExtended(rel, MAIN_FORKNUM, num_pages - 1, RBM_NORMAL,
info->strategy);
LockBufferForCleanup(buf);
_bt_delitems_vacuum(rel, buf, NULL, 0, vstate.lastBlockVacuumed);
_bt_relbuf(rel, buf);
}
MemoryContextDelete(vstate.pagedelcontext);
/* update statistics */
stats->num_pages = num_pages;
stats->pages_free = vstate.totFreePages;
}
Datum
gistrescan(PG_FUNCTION_ARGS)
{
IndexScanDesc scan = (IndexScanDesc) PG_GETARG_POINTER(0);
ScanKey key = (ScanKey) PG_GETARG_POINTER(1);
ScanKey orderbys = (ScanKey) PG_GETARG_POINTER(3);
/* nkeys and norderbys arguments are ignored */
GISTScanOpaque so = (GISTScanOpaque) scan->opaque;
bool first_time;
int i;
MemoryContext oldCxt;
/* rescan an existing indexscan --- reset state */
/*
* The first time through, we create the search queue in the scanCxt.
* Subsequent times through, we create the queue in a separate queueCxt,
* which is created on the second call and reset on later calls. Thus, in
* the common case where a scan is only rescan'd once, we just put the
* queue in scanCxt and don't pay the overhead of making a second memory
* context. If we do rescan more than once, the first RBTree is just left
* for dead until end of scan; this small wastage seems worth the savings
* in the common case.
*/
if (so->queue == NULL)
{
/* first time through */
Assert(so->queueCxt == so->giststate->scanCxt);
first_time = true;
}
else if (so->queueCxt == so->giststate->scanCxt)
{
/* second time through */
so->queueCxt = AllocSetContextCreate(so->giststate->scanCxt,
"GiST queue context",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
first_time = false;
}
else
{
/* third or later time through */
MemoryContextReset(so->queueCxt);
first_time = false;
}
/*
* If we're doing an index-only scan, on the first call, also initialize
* a tuple descriptor to represent the returned index tuples and create a
* memory context to hold them during the scan.
*/
if (scan->xs_want_itup && !scan->xs_itupdesc)
{
int natts;
int attno;
/*
* The storage type of the index can be different from the original
* datatype being indexed, so we cannot just grab the index's tuple
* descriptor. Instead, construct a descriptor with the original data
* types.
*/
natts = RelationGetNumberOfAttributes(scan->indexRelation);
so->giststate->fetchTupdesc = CreateTemplateTupleDesc(natts, false);
for (attno = 1; attno <= natts; attno++)
{
TupleDescInitEntry(so->giststate->fetchTupdesc, attno, NULL,
scan->indexRelation->rd_opcintype[attno - 1],
-1, 0);
}
scan->xs_itupdesc = so->giststate->fetchTupdesc;
so->pageDataCxt = AllocSetContextCreate(so->giststate->scanCxt,
"GiST page data context",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
}
/* create new___, empty RBTree for search queue */
oldCxt = MemoryContextSwitchTo(so->queueCxt);
so->queue = pairingheap_allocate(pairingheap_GISTSearchItem_cmp, scan);
MemoryContextSwitchTo(oldCxt);
so->firstCall = true;
/* Update scan key, if a new___ one is given */
if (key && scan->numberOfKeys > 0)
{
void **fn_extras = NULL;
/*
* If this isn't the first time through, preserve the fn_extra
* pointers, so that if the consistentFns are using them to cache
* data, that data is not leaked across a rescan.
*/
if (!first_time)
{
fn_extras = (void **) palloc(scan->numberOfKeys * sizeof(void *));
for (i = 0; i < scan->numberOfKeys; i++)
fn_extras[i] = scan->keyData[i].sk_func.fn_extra;
}
memmove(scan->keyData, key,
scan->numberOfKeys * sizeof(ScanKeyData));
/*
* Modify the scan key so that the Consistent method is called for all
* comparisons. The original operator___ is passed to the Consistent
* function in the form of its strategy number, which is available
* from the sk_strategy field, and its subtype from the sk_subtype
* field.
*
* Next, if any of keys is a NULL and that key is not marked with
* SK_SEARCHNULL/SK_SEARCHNOTNULL then nothing can be found (ie, we
* assume all indexable operators are strict).
*/
so->qual_ok = true;
for (i = 0; i < scan->numberOfKeys; i++)
{
ScanKey skey = scan->keyData + i;
fmgr_info_copy(&(skey->sk_func),
&(so->giststate->consistentFn[skey->sk_attno - 1]),
so->giststate->scanCxt);
/* Restore prior fn_extra pointers, if not first time */
if (!first_time)
skey->sk_func.fn_extra = fn_extras[i];
if (skey->sk_flags & SK_ISNULL)
{
if (!(skey->sk_flags & (SK_SEARCHNULL | SK_SEARCHNOTNULL)))
so->qual_ok = false;
}
}
if (!first_time)
pfree(fn_extras);
}
/* Update order-by key, if a new___ one is given */
if (orderbys && scan->numberOfOrderBys > 0)
{
void **fn_extras = NULL;
/* As above, preserve fn_extra if not first time through */
if (!first_time)
{
fn_extras = (void **) palloc(scan->numberOfOrderBys * sizeof(void *));
for (i = 0; i < scan->numberOfOrderBys; i++)
fn_extras[i] = scan->orderByData[i].sk_func.fn_extra;
}
memmove(scan->orderByData, orderbys,
scan->numberOfOrderBys * sizeof(ScanKeyData));
so->orderByTypes = (Oid *) palloc(scan->numberOfOrderBys * sizeof(Oid));
/*
* Modify the order-by key so that the Distance method is called for
* all comparisons. The original operator___ is passed to the Distance
* function in the form of its strategy number, which is available
* from the sk_strategy field, and its subtype from the sk_subtype
* field.
*/
for (i = 0; i < scan->numberOfOrderBys; i++)
{
ScanKey skey = scan->orderByData + i;
FmgrInfo *finfo = &(so->giststate->distanceFn[skey->sk_attno - 1]);
/* Check we actually have a distance function ... */
if (!OidIsValid(finfo->fn_oid))
elog(ERROR, "missing support function %d for attribute %d of index \"%s\"",
GIST_DISTANCE_PROC, skey->sk_attno,
RelationGetRelationName(scan->indexRelation));
fmgr_info_copy(&(skey->sk_func), finfo, so->giststate->scanCxt);
/*
* Look up the datatype returned by the original ordering operator___.
* GiST always uses a float8 for the distance function, but the
* ordering operator___ could be anything else.
*
* XXX: The distance function is only allowed to be lossy if the
* ordering operator___'s result type is float4 or float8. Otherwise
* we don't know how to return the distance to the executor. But
* we cannot check that here, as we won't know if the distance
* function is lossy until it returns *recheck = true for the
* first time.
*/
so->orderByTypes[i] = get_func_rettype(skey->sk_func.fn_oid);
/* Restore prior fn_extra pointers, if not first time */
if (!first_time)
skey->sk_func.fn_extra = fn_extras[i];
}
if (!first_time)
pfree(fn_extras);
}
PG_RETURN_VOID();
}
/*
* Main entry point for bgwriter process
*
* This is invoked from BootstrapMain, which has already created the basic
* execution environment, but not enabled signals yet.
*/
void
BackgroundWriterMain(void)
{
sigjmp_buf local_sigjmp_buf;
MemoryContext bgwriter_context;
BgWriterShmem->bgwriter_pid = MyProcPid;
am_bg_writer = true;
/*
* If possible, make this process a group leader, so that the postmaster
* can signal any child processes too. (bgwriter probably never has any
* child processes, but for consistency we make all postmaster child
* processes do this.)
*/
#ifdef HAVE_SETSID
if (setsid() < 0)
elog(FATAL, "setsid() failed: %m");
#endif
/*
* Properly accept or ignore signals the postmaster might send us
*
* Note: we deliberately ignore SIGTERM, because during a standard Unix
* system shutdown cycle, init will SIGTERM all processes at once. We
* want to wait for the backends to exit, whereupon the postmaster will
* tell us it's okay to shut down (via SIGUSR2).
*
* SIGUSR1 is presently unused; keep it spare in case someday we want this
* process to participate in sinval messaging.
*/
pqsignal(SIGHUP, BgSigHupHandler); /* set flag to read config file */
pqsignal(SIGINT, ReqCheckpointHandler); /* request checkpoint */
pqsignal(SIGTERM, SIG_IGN); /* ignore SIGTERM */
pqsignal(SIGQUIT, bg_quickdie); /* hard crash time */
pqsignal(SIGALRM, SIG_IGN);
pqsignal(SIGPIPE, SIG_IGN);
pqsignal(SIGUSR1, SIG_IGN); /* reserve for sinval */
pqsignal(SIGUSR2, ReqShutdownHandler); /* request shutdown */
/*
* Reset some signals that are accepted by postmaster but not here
*/
pqsignal(SIGCHLD, SIG_DFL);
pqsignal(SIGTTIN, SIG_DFL);
pqsignal(SIGTTOU, SIG_DFL);
pqsignal(SIGCONT, SIG_DFL);
pqsignal(SIGWINCH, SIG_DFL);
/* We allow SIGQUIT (quickdie) at all times */
#ifdef HAVE_SIGPROCMASK
sigdelset(&BlockSig, SIGQUIT);
#else
BlockSig &= ~(sigmask(SIGQUIT));
#endif
/*
* Initialize so that first time-driven event happens at the correct time.
*/
last_checkpoint_time = last_xlog_switch_time = time(NULL);
/*
* Create a resource owner to keep track of our resources (currently only
* buffer pins).
*/
CurrentResourceOwner = ResourceOwnerCreate(NULL, "Background Writer");
/*
* Create a memory context that we will do all our work in. We do this so
* that we can reset the context during error recovery and thereby avoid
* possible memory leaks. Formerly this code just ran in
* TopMemoryContext, but resetting that would be a really bad idea.
*/
bgwriter_context = AllocSetContextCreate(TopMemoryContext,
"Background Writer",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
MemoryContextSwitchTo(bgwriter_context);
/*
* If an exception is encountered, processing resumes here.
*
* See notes in postgres.c about the design of this coding.
*/
if (sigsetjmp(local_sigjmp_buf, 1) != 0)
{
/* Since not using PG_TRY, must reset error stack by hand */
error_context_stack = NULL;
/* Prevent interrupts while cleaning up */
HOLD_INTERRUPTS();
/* Report the error to the server log */
EmitErrorReport();
/*
* These operations are really just a minimal subset of
* AbortTransaction(). We don't have very many resources to worry
* about in bgwriter, but we do have LWLocks, buffers, and temp files.
*/
LWLockReleaseAll();
AbortBufferIO();
UnlockBuffers();
/* buffer pins are released here: */
ResourceOwnerRelease(CurrentResourceOwner,
RESOURCE_RELEASE_BEFORE_LOCKS,
false, true);
/* we needn't bother with the other ResourceOwnerRelease phases */
AtEOXact_Buffers(false);
AtEOXact_Files();
AtEOXact_HashTables(false);
/* Warn any waiting backends that the checkpoint failed. */
if (ckpt_active)
{
/* use volatile pointer to prevent code rearrangement */
volatile BgWriterShmemStruct *bgs = BgWriterShmem;
SpinLockAcquire(&bgs->ckpt_lck);
bgs->ckpt_failed++;
bgs->ckpt_done = bgs->ckpt_started;
SpinLockRelease(&bgs->ckpt_lck);
ckpt_active = false;
}
/*
* Now return to normal top-level context and clear ErrorContext for
* next time.
*/
MemoryContextSwitchTo(bgwriter_context);
FlushErrorState();
/* Flush any leaked data in the top-level context */
MemoryContextResetAndDeleteChildren(bgwriter_context);
/* Now we can allow interrupts again */
RESUME_INTERRUPTS();
/*
* Sleep at least 1 second after any error. A write error is likely
* to be repeated, and we don't want to be filling the error logs as
* fast as we can.
*/
pg_usleep(1000000L);
/*
* Close all open files after any error. This is helpful on Windows,
* where holding deleted files open causes various strange errors.
* It's not clear we need it elsewhere, but shouldn't hurt.
*/
smgrcloseall();
}
/* We can now handle ereport(ERROR) */
PG_exception_stack = &local_sigjmp_buf;
/*
* Unblock signals (they were blocked when the postmaster forked us)
*/
PG_SETMASK(&UnBlockSig);
/*
* Loop forever
*/
for (;;)
{
bool do_checkpoint = false;
int flags = 0;
time_t now;
int elapsed_secs;
/*
* Emergency bailout if postmaster has died. This is to avoid the
* necessity for manual cleanup of all postmaster children.
*/
if (!PostmasterIsAlive(true))
exit(1);
/*
* Process any requests or signals received recently.
*/
AbsorbFsyncRequests();
if (got_SIGHUP)
{
got_SIGHUP = false;
ProcessConfigFile(PGC_SIGHUP);
}
if (checkpoint_requested)
{
checkpoint_requested = false;
do_checkpoint = true;
BgWriterStats.m_requested_checkpoints++;
}
if (shutdown_requested)
{
/*
* From here on, elog(ERROR) should end with exit(1), not send
* control back to the sigsetjmp block above
*/
ExitOnAnyError = true;
/* Close down the database */
ShutdownXLOG(0, 0);
DumpFreeSpaceMap(0, 0);
/* Normal exit from the bgwriter is here */
proc_exit(0); /* done */
}
/*
* Force a checkpoint if too much time has elapsed since the last one.
* Note that we count a timed checkpoint in stats only when this
* occurs without an external request, but we set the CAUSE_TIME flag
* bit even if there is also an external request.
*/
now = time(NULL);
elapsed_secs = now - last_checkpoint_time;
if (elapsed_secs >= CheckPointTimeout)
{
if (!do_checkpoint)
BgWriterStats.m_timed_checkpoints++;
do_checkpoint = true;
flags |= CHECKPOINT_CAUSE_TIME;
}
/*
* Do a checkpoint if requested, otherwise do one cycle of
* dirty-buffer writing.
*/
if (do_checkpoint)
{
/* use volatile pointer to prevent code rearrangement */
volatile BgWriterShmemStruct *bgs = BgWriterShmem;
/*
* Atomically fetch the request flags to figure out what kind of a
* checkpoint we should perform, and increase the started-counter
* to acknowledge that we've started a new checkpoint.
*/
SpinLockAcquire(&bgs->ckpt_lck);
flags |= bgs->ckpt_flags;
bgs->ckpt_flags = 0;
bgs->ckpt_started++;
SpinLockRelease(&bgs->ckpt_lck);
/*
* We will warn if (a) too soon since last checkpoint (whatever
* caused it) and (b) somebody set the CHECKPOINT_CAUSE_XLOG flag
* since the last checkpoint start. Note in particular that this
* implementation will not generate warnings caused by
* CheckPointTimeout < CheckPointWarning.
*/
if ((flags & CHECKPOINT_CAUSE_XLOG) &&
elapsed_secs < CheckPointWarning)
ereport(LOG,
(errmsg("checkpoints are occurring too frequently (%d seconds apart)",
elapsed_secs),
errhint("Consider increasing the configuration parameter \"checkpoint_segments\".")));
/*
* Initialize bgwriter-private variables used during checkpoint.
*/
ckpt_active = true;
ckpt_start_recptr = GetInsertRecPtr();
ckpt_start_time = now;
ckpt_cached_elapsed = 0;
/*
* Do the checkpoint.
*/
CreateCheckPoint(flags);
/*
* After any checkpoint, close all smgr files. This is so we
* won't hang onto smgr references to deleted files indefinitely.
*/
smgrcloseall();
/*
* Indicate checkpoint completion to any waiting backends.
*/
SpinLockAcquire(&bgs->ckpt_lck);
bgs->ckpt_done = bgs->ckpt_started;
SpinLockRelease(&bgs->ckpt_lck);
ckpt_active = false;
/*
* Note we record the checkpoint start time not end time as
* last_checkpoint_time. This is so that time-driven checkpoints
* happen at a predictable spacing.
*/
last_checkpoint_time = now;
}
else
BgBufferSync();
/* Check for archive_timeout and switch xlog files if necessary. */
CheckArchiveTimeout();
/* Nap for the configured time. */
BgWriterNap();
}
}
/*
* Insert a new item to a page.
*
* Returns true if the insertion was finished. On false, the page was split and
* the parent needs to be updated. (A root split returns true as it doesn't
* need any further action by the caller to complete.)
*
* When inserting a downlink to an internal page, 'childbuf' contains the
* child page that was split. Its GIN_INCOMPLETE_SPLIT flag will be cleared
* atomically with the insert. Also, the existing item at offset stack->off
* in the target page is updated to point to updateblkno.
*
* stack->buffer is locked on entry, and is kept locked.
* Likewise for childbuf, if given.
*/
static bool
ginPlaceToPage(GinBtree btree, GinBtreeStack *stack,
void *insertdata, BlockNumber updateblkno,
Buffer childbuf, GinStatsData *buildStats)
{
Page page = BufferGetPage(stack->buffer);
bool result;
GinPlaceToPageRC rc;
uint16 xlflags = 0;
Page childpage = NULL;
Page newlpage = NULL,
newrpage = NULL;
void *ptp_workspace = NULL;
XLogRecData payloadrdata[10];
MemoryContext tmpCxt;
MemoryContext oldCxt;
/*
* We do all the work of this function and its subfunctions in a temporary
* memory context. This avoids leakages and simplifies APIs, since some
* subfunctions allocate storage that has to survive until we've finished
* the WAL insertion.
*/
tmpCxt = AllocSetContextCreate(CurrentMemoryContext,
"ginPlaceToPage temporary context",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
oldCxt = MemoryContextSwitchTo(tmpCxt);
if (GinPageIsData(page))
xlflags |= GIN_INSERT_ISDATA;
if (GinPageIsLeaf(page))
{
xlflags |= GIN_INSERT_ISLEAF;
Assert(!BufferIsValid(childbuf));
Assert(updateblkno == InvalidBlockNumber);
}
else
{
Assert(BufferIsValid(childbuf));
Assert(updateblkno != InvalidBlockNumber);
childpage = BufferGetPage(childbuf);
}
/*
* See if the incoming tuple will fit on the page. beginPlaceToPage will
* decide if the page needs to be split, and will compute the split
* contents if so. See comments for beginPlaceToPage and execPlaceToPage
* functions for more details of the API here.
*/
rc = btree->beginPlaceToPage(btree, stack->buffer, stack,
insertdata, updateblkno,
&ptp_workspace,
&newlpage, &newrpage,
payloadrdata);
if (rc == GPTP_NO_WORK)
{
/* Nothing to do */
result = true;
}
else if (rc == GPTP_INSERT)
{
/* It will fit, perform the insertion */
START_CRIT_SECTION();
/* Perform the page update, and set up WAL data about it */
btree->execPlaceToPage(btree, stack->buffer, stack,
insertdata, updateblkno,
ptp_workspace, payloadrdata);
MarkBufferDirty(stack->buffer);
/* An insert to an internal page finishes the split of the child. */
if (BufferIsValid(childbuf))
{
GinPageGetOpaque(childpage)->flags &= ~GIN_INCOMPLETE_SPLIT;
MarkBufferDirty(childbuf);
}
if (RelationNeedsWAL(btree->index))
{
XLogRecPtr recptr;
XLogRecData rdata[3];
ginxlogInsert xlrec;
BlockIdData childblknos[2];
xlrec.node = btree->index->rd_node;
xlrec.blkno = BufferGetBlockNumber(stack->buffer);
xlrec.flags = xlflags;
rdata[0].buffer = InvalidBuffer;
rdata[0].data = (char *) &xlrec;
rdata[0].len = sizeof(ginxlogInsert);
/*
* Log information about child if this was an insertion of a
* downlink.
*/
if (BufferIsValid(childbuf))
{
rdata[0].next = &rdata[1];
BlockIdSet(&childblknos[0], BufferGetBlockNumber(childbuf));
BlockIdSet(&childblknos[1], GinPageGetOpaque(childpage)->rightlink);
rdata[1].buffer = InvalidBuffer;
rdata[1].data = (char *) childblknos;
rdata[1].len = sizeof(BlockIdData) * 2;
rdata[1].next = &rdata[2];
rdata[2].buffer = childbuf;
rdata[2].buffer_std = true;
rdata[2].data = NULL;
rdata[2].len = 0;
rdata[2].next = payloadrdata;
}
else
rdata[0].next = payloadrdata;
recptr = XLogInsert(RM_GIN_ID, XLOG_GIN_INSERT, rdata);
PageSetLSN(page, recptr);
if (BufferIsValid(childbuf))
PageSetLSN(childpage, recptr);
}
END_CRIT_SECTION();
/* Insertion is complete. */
result = true;
}
else if (rc == GPTP_SPLIT)
{
/*
* Didn't fit, need to split. The split has been computed in newlpage
* and newrpage, which are pointers to palloc'd pages, not associated
* with buffers. stack->buffer is not touched yet.
*/
Buffer rbuffer;
BlockNumber savedRightLink;
ginxlogSplit data;
Buffer lbuffer = InvalidBuffer;
Page newrootpg = NULL;
/* Get a new index page to become the right page */
rbuffer = GinNewBuffer(btree->index);
/* During index build, count the new page */
if (buildStats)
{
if (btree->isData)
buildStats->nDataPages++;
else
buildStats->nEntryPages++;
}
savedRightLink = GinPageGetOpaque(page)->rightlink;
/* Begin setting up WAL record (which we might not use) */
data.node = btree->index->rd_node;
data.rblkno = BufferGetBlockNumber(rbuffer);
data.flags = xlflags;
if (BufferIsValid(childbuf))
{
data.leftChildBlkno = BufferGetBlockNumber(childbuf);
data.rightChildBlkno = GinPageGetOpaque(childpage)->rightlink;
}
else
data.leftChildBlkno = data.rightChildBlkno = InvalidBlockNumber;
if (stack->parent == NULL)
{
/*
* splitting the root, so we need to allocate new left page and
* place pointers to left and right page on root page.
*/
lbuffer = GinNewBuffer(btree->index);
/* During index build, count the new left page */
if (buildStats)
{
if (btree->isData)
buildStats->nDataPages++;
else
buildStats->nEntryPages++;
}
/*
* root never has a right-link, so we borrow the rrlink field to
* store the root block number.
*/
data.rrlink = BufferGetBlockNumber(stack->buffer);
data.lblkno = BufferGetBlockNumber(lbuffer);
data.flags |= GIN_SPLIT_ROOT;
GinPageGetOpaque(newrpage)->rightlink = InvalidBlockNumber;
GinPageGetOpaque(newlpage)->rightlink = BufferGetBlockNumber(rbuffer);
/*
* Construct a new root page containing downlinks to the new left
* and right pages. (Do this in a temporary copy rather than
* overwriting the original page directly, since we're not in the
* critical section yet.)
*/
newrootpg = PageGetTempPage(newrpage);
GinInitPage(newrootpg, GinPageGetOpaque(newlpage)->flags & ~(GIN_LEAF | GIN_COMPRESSED), BLCKSZ);
btree->fillRoot(btree, newrootpg,
BufferGetBlockNumber(lbuffer), newlpage,
BufferGetBlockNumber(rbuffer), newrpage);
}
else
{
/* splitting a non-root page */
data.rrlink = savedRightLink;
data.lblkno = BufferGetBlockNumber(stack->buffer);
GinPageGetOpaque(newrpage)->rightlink = savedRightLink;
GinPageGetOpaque(newlpage)->flags |= GIN_INCOMPLETE_SPLIT;
GinPageGetOpaque(newlpage)->rightlink = BufferGetBlockNumber(rbuffer);
}
/*
* OK, we have the new contents of the left page in a temporary copy
* now (newlpage), and likewise for the new contents of the
* newly-allocated right block. The original page is still unchanged.
*
* If this is a root split, we also have a temporary page containing
* the new contents of the root.
*/
START_CRIT_SECTION();
MarkBufferDirty(rbuffer);
MarkBufferDirty(stack->buffer);
/*
* Restore the temporary copies over the real buffers.
*/
if (stack->parent == NULL)
{
/* Splitting the root, three pages to update */
MarkBufferDirty(lbuffer);
memcpy(page, newrootpg, BLCKSZ);
memcpy(BufferGetPage(lbuffer), newlpage, BLCKSZ);
memcpy(BufferGetPage(rbuffer), newrpage, BLCKSZ);
}
else
{
/* Normal split, only two pages to update */
memcpy(page, newlpage, BLCKSZ);
memcpy(BufferGetPage(rbuffer), newrpage, BLCKSZ);
}
/* We also clear childbuf's INCOMPLETE_SPLIT flag, if passed */
if (BufferIsValid(childbuf))
{
GinPageGetOpaque(childpage)->flags &= ~GIN_INCOMPLETE_SPLIT;
MarkBufferDirty(childbuf);
}
/* write WAL record */
if (RelationNeedsWAL(btree->index))
{
XLogRecData rdata[2];
XLogRecPtr recptr;
rdata[0].buffer = InvalidBuffer;
rdata[0].data = (char *) &data;
rdata[0].len = sizeof(ginxlogSplit);
if (BufferIsValid(childbuf))
{
rdata[0].next = &rdata[1];
rdata[1].buffer = childbuf;
rdata[1].buffer_std = true;
rdata[1].data = NULL;
rdata[1].len = 0;
rdata[1].next = payloadrdata;
}
else
rdata[0].next = payloadrdata;
recptr = XLogInsert(RM_GIN_ID, XLOG_GIN_SPLIT, rdata);
PageSetLSN(page, recptr);
PageSetLSN(BufferGetPage(rbuffer), recptr);
if (stack->parent == NULL)
PageSetLSN(BufferGetPage(lbuffer), recptr);
if (BufferIsValid(childbuf))
PageSetLSN(childpage, recptr);
}
END_CRIT_SECTION();
/*
* We can release the locks/pins on the new pages now, but keep
* stack->buffer locked. childbuf doesn't get unlocked either.
*/
UnlockReleaseBuffer(rbuffer);
if (stack->parent == NULL)
UnlockReleaseBuffer(lbuffer);
/*
* If we split the root, we're done. Otherwise the split is not
* complete until the downlink for the new page has been inserted to
* the parent.
*/
result = (stack->parent == NULL);
}
else
{
elog(ERROR, "invalid return code from GIN placeToPage method: %d", rc);
result = false; /* keep compiler quiet */
}
/* Clean up temp context */
MemoryContextSwitchTo(oldCxt);
MemoryContextDelete(tmpCxt);
return result;
}
void
initMotionLayerStructs(MotionLayerState **mlStates)
{
MemoryContext oldCtxt;
MemoryContext ml_mctx;
uint8 *pData;
if (Gp_role == GP_ROLE_UTILITY)
return;
if (Gp_interconnect_type == INTERCONNECT_TYPE_UDPIFC)
Gp_max_tuple_chunk_size = Gp_max_packet_size - sizeof(struct icpkthdr) - TUPLE_CHUNK_HEADER_SIZE;
else if (Gp_interconnect_type == INTERCONNECT_TYPE_TCP)
Gp_max_tuple_chunk_size = Gp_max_packet_size - PACKET_HEADER_SIZE - TUPLE_CHUNK_HEADER_SIZE;
/*
* Use the statically allocated chunk that is intended for sending end-of-
* stream messages so that we don't incur allocation and deallocation
* overheads.
*/
s_eos_chunk_data->p_next = NULL;
s_eos_chunk_data->inplace = NULL;
s_eos_chunk_data->chunk_length = TUPLE_CHUNK_HEADER_SIZE;
pData = s_eos_chunk_data->chunk_data;
SetChunkDataSize(pData, 0);
SetChunkType(pData, TC_END_OF_STREAM);
/*
* Create the memory-contexts that we will use within the Motion Layer.
*
* We make the Motion Layer memory-context a child of the ExecutorState
* Context, as it lives inside of the estate of a specific query and needs
* to get freed when the query is finished.
*
* The tuple-serial memory-context is a child of the Motion Layer
* memory-context
*
* NOTE: we need to be sure the caller is in ExecutorState memory context
* (estate->es_query_cxt) before calling us .
*/
ml_mctx =
AllocSetContextCreate(CurrentMemoryContext, "MotionLayerMemCtxt",
ALLOCSET_SMALL_MINSIZE,
ALLOCSET_SMALL_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE); /* use a setting bigger
* than "small" */
/*
* Switch to the Motion Layer memory context, so that we can clean things
* up easily.
*/
oldCtxt = MemoryContextSwitchTo(ml_mctx);
Assert(*mlStates == NULL);
*mlStates = palloc0(sizeof(MotionLayerState));
(*mlStates)->mnEntries = palloc0(MNE_INITIAL_COUNT * sizeof(MotionNodeEntry));
(*mlStates)->mneCount = MNE_INITIAL_COUNT;
/* Allocation is done. Go back to caller memory-context. */
MemoryContextSwitchTo(oldCtxt);
/*
* Keep our motion layer memory context in our newly created motion layer.
*/
(*mlStates)->motion_layer_mctx = ml_mctx;
}
/*
* Set the client encoding and save fmgrinfo for the conversion
* function if necessary. Returns 0 if okay, -1 if not (bad encoding
* or can't support conversion)
*/
int
SetClientEncoding(int encoding, bool doit)
{
int current_server_encoding;
Oid to_server_proc,
to_client_proc;
FmgrInfo *to_server;
FmgrInfo *to_client;
MemoryContext oldcontext;
if (!PG_VALID_FE_ENCODING(encoding))
return -1;
/* Can't do anything during startup, per notes above */
if (!backend_startup_complete)
{
if (doit)
pending_client_encoding = encoding;
return 0;
}
current_server_encoding = GetDatabaseEncoding();
/*
* Check for cases that require no conversion function.
*/
if (current_server_encoding == encoding ||
current_server_encoding == PG_SQL_ASCII ||
encoding == PG_SQL_ASCII)
{
if (doit)
{
ClientEncoding = &pg_enc2name_tbl[encoding];
ToServerConvProc = NULL;
ToClientConvProc = NULL;
if (MbProcContext)
MemoryContextReset(MbProcContext);
}
return 0;
}
/*
* If we're not inside a transaction then we can't do catalog lookups, so
* fail. After backend startup, this could only happen if we are
* re-reading postgresql.conf due to SIGHUP --- so basically this just
* constrains the ability to change client_encoding on the fly from
* postgresql.conf. Which would probably be a stupid thing to do anyway.
*/
if (!IsTransactionState())
return -1;
/*
* Look up the conversion functions.
*/
to_server_proc = FindDefaultConversionProc(encoding,
current_server_encoding);
if (!OidIsValid(to_server_proc))
return -1;
to_client_proc = FindDefaultConversionProc(current_server_encoding,
encoding);
if (!OidIsValid(to_client_proc))
return -1;
/*
* Done if not wanting to actually apply setting.
*/
if (!doit)
return 0;
/* Before loading the new fmgr info, remove the old info, if any */
ToServerConvProc = NULL;
ToClientConvProc = NULL;
if (MbProcContext != NULL)
{
MemoryContextReset(MbProcContext);
}
else
{
/*
* This is the first time through, so create the context. Make it a
* child of TopMemoryContext so that these values survive across
* transactions.
*/
MbProcContext = AllocSetContextCreate(TopMemoryContext,
"MbProcContext",
ALLOCSET_SMALL_MINSIZE,
ALLOCSET_SMALL_INITSIZE,
ALLOCSET_SMALL_MAXSIZE);
}
/* Load the fmgr info into MbProcContext */
oldcontext = MemoryContextSwitchTo(MbProcContext);
to_server = palloc(sizeof(FmgrInfo));
to_client = palloc(sizeof(FmgrInfo));
fmgr_info(to_server_proc, to_server);
fmgr_info(to_client_proc, to_client);
MemoryContextSwitchTo(oldcontext);
ClientEncoding = &pg_enc2name_tbl[encoding];
ToServerConvProc = to_server;
ToClientConvProc = to_client;
return 0;
}
IndexBulkDeleteResult *
ginbulkdelete(IndexVacuumInfo *info, IndexBulkDeleteResult *stats,
IndexBulkDeleteCallback callback, void *callback_state)
{
Relation index = info->index;
BlockNumber blkno = GIN_ROOT_BLKNO;
GinVacuumState gvs;
Buffer buffer;
BlockNumber rootOfPostingTree[BLCKSZ / (sizeof(IndexTupleData) + sizeof(ItemId))];
uint32 nRoot;
gvs.tmpCxt = AllocSetContextCreate(CurrentMemoryContext,
"Gin vacuum temporary context",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
gvs.index = index;
gvs.callback = callback;
gvs.callback_state = callback_state;
gvs.strategy = info->strategy;
initGinState(&gvs.ginstate, index);
/* first time through? */
if (stats == NULL)
{
/* Yes, so initialize stats to zeroes */
stats = (IndexBulkDeleteResult *) palloc0(sizeof(IndexBulkDeleteResult));
/* and cleanup any pending inserts */
ginInsertCleanup(&gvs.ginstate, false, stats);
}
/* we'll re-count the tuples each time */
stats->num_index_tuples = 0;
gvs.result = stats;
buffer = ReadBufferExtended(index, MAIN_FORKNUM, blkno,
RBM_NORMAL, info->strategy);
/* find leaf page */
for (;;)
{
Page page = BufferGetPage(buffer);
IndexTuple itup;
LockBuffer(buffer, GIN_SHARE);
Assert(!GinPageIsData(page));
if (GinPageIsLeaf(page))
{
LockBuffer(buffer, GIN_UNLOCK);
LockBuffer(buffer, GIN_EXCLUSIVE);
if (blkno == GIN_ROOT_BLKNO && !GinPageIsLeaf(page))
{
LockBuffer(buffer, GIN_UNLOCK);
continue; /* check it one more */
}
break;
}
Assert(PageGetMaxOffsetNumber(page) >= FirstOffsetNumber);
itup = (IndexTuple) PageGetItem(page, PageGetItemId(page, FirstOffsetNumber));
blkno = GinGetDownlink(itup);
Assert(blkno != InvalidBlockNumber);
UnlockReleaseBuffer(buffer);
buffer = ReadBufferExtended(index, MAIN_FORKNUM, blkno,
RBM_NORMAL, info->strategy);
}
/* right now we found leftmost page in entry's BTree */
for (;;)
{
Page page = BufferGetPage(buffer);
Page resPage;
uint32 i;
Assert(!GinPageIsData(page));
resPage = ginVacuumEntryPage(&gvs, buffer, rootOfPostingTree, &nRoot);
blkno = GinPageGetOpaque(page)->rightlink;
if (resPage)
{
START_CRIT_SECTION();
PageRestoreTempPage(resPage, page);
MarkBufferDirty(buffer);
xlogVacuumPage(gvs.index, buffer);
UnlockReleaseBuffer(buffer);
END_CRIT_SECTION();
}
else
{
UnlockReleaseBuffer(buffer);
}
vacuum_delay_point();
for (i = 0; i < nRoot; i++)
{
ginVacuumPostingTree(&gvs, rootOfPostingTree[i]);
vacuum_delay_point();
}
if (blkno == InvalidBlockNumber) /* rightmost page */
break;
buffer = ReadBufferExtended(index, MAIN_FORKNUM, blkno,
RBM_NORMAL, info->strategy);
LockBuffer(buffer, GIN_EXCLUSIVE);
}
MemoryContextDelete(gvs.tmpCxt);
return gvs.result;
}
/* ----------------
* CreateExecutorState
*
* Create and initialize an EState node, which is the root of
* working storage for an entire Executor invocation.
*
* Principally, this creates the per-query memory context that will be
* used to hold all working data that lives till the end of the query.
* Note that the per-query context will become a child of the caller's
* CurrentMemoryContext.
* ----------------
*/
EState *
CreateExecutorState(void)
{
EState *estate;
MemoryContext qcontext;
MemoryContext oldcontext;
/*
* Create the per-query context for this Executor run.
*/
qcontext = AllocSetContextCreate(CurrentMemoryContext,
"ExecutorState",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
/*
* Make the EState node within the per-query context. This way, we don't
* need a separate pfree() operation for it at shutdown.
*/
oldcontext = MemoryContextSwitchTo(qcontext);
estate = makeNode(EState);
/*
* Initialize all fields of the Executor State structure
*/
estate->es_direction = ForwardScanDirection;
estate->es_snapshot = SnapshotNow;
estate->es_crosscheck_snapshot = InvalidSnapshot; /* no crosscheck */
estate->es_range_table = NIL;
estate->es_plannedstmt = NULL;
estate->es_junkFilter = NULL;
estate->es_output_cid = (CommandId) 0;
estate->es_result_relations = NULL;
estate->es_num_result_relations = 0;
estate->es_result_relation_info = NULL;
estate->es_trig_target_relations = NIL;
estate->es_trig_tuple_slot = NULL;
estate->es_trig_oldtup_slot = NULL;
estate->es_trig_newtup_slot = NULL;
estate->es_param_list_info = NULL;
estate->es_param_exec_vals = NULL;
estate->es_query_cxt = qcontext;
estate->es_tupleTable = NIL;
estate->es_rowMarks = NIL;
estate->es_processed = 0;
estate->es_lastoid = InvalidOid;
estate->es_top_eflags = 0;
estate->es_instrument = 0;
estate->es_finished = false;
estate->es_exprcontexts = NIL;
estate->es_subplanstates = NIL;
estate->es_auxmodifytables = NIL;
estate->es_per_tuple_exprcontext = NULL;
estate->es_epqTuple = NULL;
estate->es_epqTupleSet = NULL;
estate->es_epqScanDone = NULL;
/*
* Return the executor state structure
*/
MemoryContextSwitchTo(oldcontext);
return estate;
}
/*
* SQL function json_array_elements
*
* get the elements from a json array
*
* a lot of this processing is similar to the json_each* functions
*/
Datum
json_array_elements(PG_FUNCTION_ARGS)
{
text *json = PG_GETARG_TEXT_P(0);
/* elements doesn't need any escaped strings, so use false here */
JsonLexContext *lex = makeJsonLexContext(json, false);
JsonSemAction *sem;
ReturnSetInfo *rsi;
MemoryContext old_cxt;
TupleDesc tupdesc;
ElementsState *state;
state = palloc0(sizeof(ElementsState));
sem = palloc0(sizeof(JsonSemAction));
rsi = (ReturnSetInfo *) fcinfo->resultinfo;
if (!rsi || !IsA(rsi, ReturnSetInfo) ||
(rsi->allowedModes & SFRM_Materialize) == 0 ||
rsi->expectedDesc == NULL)
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("set-valued function called in context that "
"cannot accept a set")));
rsi->returnMode = SFRM_Materialize;
/* it's a simple type, so don't use get_call_result_type() */
tupdesc = rsi->expectedDesc;
/* make these in a sufficiently long-lived memory context */
old_cxt = MemoryContextSwitchTo(rsi->econtext->ecxt_per_query_memory);
state->ret_tdesc = CreateTupleDescCopy(tupdesc);
BlessTupleDesc(state->ret_tdesc);
state->tuple_store =
tuplestore_begin_heap(rsi->allowedModes & SFRM_Materialize_Random,
false, work_mem);
MemoryContextSwitchTo(old_cxt);
sem->semstate = (void *) state;
sem->object_start = elements_object_start;
sem->scalar = elements_scalar;
sem->array_element_start = elements_array_element_start;
sem->array_element_end = elements_array_element_end;
state->lex = lex;
state->tmp_cxt = AllocSetContextCreate(CurrentMemoryContext,
"json_array_elements temporary cxt",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
pg_parse_json(lex, sem);
rsi->setResult = state->tuple_store;
rsi->setDesc = state->ret_tdesc;
PG_RETURN_NULL();
}
/*
* Insert new tuple to the bloom index.
*/
bool
blinsert(Relation index, Datum *values, bool *isnull,
ItemPointer ht_ctid, Relation heapRel,
IndexUniqueCheck checkUnique,
IndexInfo *indexInfo)
{
BloomState blstate;
BloomTuple *itup;
MemoryContext oldCtx;
MemoryContext insertCtx;
BloomMetaPageData *metaData;
Buffer buffer,
metaBuffer;
Page page,
metaPage;
BlockNumber blkno = InvalidBlockNumber;
OffsetNumber nStart;
GenericXLogState *state;
insertCtx = AllocSetContextCreate(CurrentMemoryContext,
"Bloom insert temporary context",
ALLOCSET_DEFAULT_SIZES);
oldCtx = MemoryContextSwitchTo(insertCtx);
initBloomState(&blstate, index);
itup = BloomFormTuple(&blstate, ht_ctid, values, isnull);
/*
* At first, try to insert new tuple to the first page in notFullPage
* array. If successful, we don't need to modify the meta page.
*/
metaBuffer = ReadBuffer(index, BLOOM_METAPAGE_BLKNO);
LockBuffer(metaBuffer, BUFFER_LOCK_SHARE);
metaData = BloomPageGetMeta(BufferGetPage(metaBuffer));
if (metaData->nEnd > metaData->nStart)
{
Page page;
blkno = metaData->notFullPage[metaData->nStart];
Assert(blkno != InvalidBlockNumber);
/* Don't hold metabuffer lock while doing insert */
LockBuffer(metaBuffer, BUFFER_LOCK_UNLOCK);
buffer = ReadBuffer(index, blkno);
LockBuffer(buffer, BUFFER_LOCK_EXCLUSIVE);
state = GenericXLogStart(index);
page = GenericXLogRegisterBuffer(state, buffer, 0);
/*
* We might have found a page that was recently deleted by VACUUM. If
* so, we can reuse it, but we must reinitialize it.
*/
if (PageIsNew(page) || BloomPageIsDeleted(page))
BloomInitPage(page, 0);
if (BloomPageAddItem(&blstate, page, itup))
{
/* Success! Apply the change, clean up, and exit */
GenericXLogFinish(state);
UnlockReleaseBuffer(buffer);
ReleaseBuffer(metaBuffer);
MemoryContextSwitchTo(oldCtx);
MemoryContextDelete(insertCtx);
return false;
}
/* Didn't fit, must try other pages */
GenericXLogAbort(state);
UnlockReleaseBuffer(buffer);
}
else
{
/* No entries in notFullPage */
LockBuffer(metaBuffer, BUFFER_LOCK_UNLOCK);
}
/*
* Try other pages in notFullPage array. We will have to change nStart in
* metapage. Thus, grab exclusive lock on metapage.
*/
LockBuffer(metaBuffer, BUFFER_LOCK_EXCLUSIVE);
/* nStart might have changed while we didn't have lock */
nStart = metaData->nStart;
/* Skip first page if we already tried it above */
if (nStart < metaData->nEnd &&
blkno == metaData->notFullPage[nStart])
nStart++;
/*
* This loop iterates for each page we try from the notFullPage array, and
* will also initialize a GenericXLogState for the fallback case of having
* to allocate a new page.
*/
for (;;)
{
state = GenericXLogStart(index);
/* get modifiable copy of metapage */
metaPage = GenericXLogRegisterBuffer(state, metaBuffer, 0);
metaData = BloomPageGetMeta(metaPage);
if (nStart >= metaData->nEnd)
break; /* no more entries in notFullPage array */
blkno = metaData->notFullPage[nStart];
Assert(blkno != InvalidBlockNumber);
buffer = ReadBuffer(index, blkno);
LockBuffer(buffer, BUFFER_LOCK_EXCLUSIVE);
page = GenericXLogRegisterBuffer(state, buffer, 0);
/* Basically same logic as above */
if (PageIsNew(page) || BloomPageIsDeleted(page))
BloomInitPage(page, 0);
if (BloomPageAddItem(&blstate, page, itup))
{
/* Success! Apply the changes, clean up, and exit */
metaData->nStart = nStart;
GenericXLogFinish(state);
UnlockReleaseBuffer(buffer);
UnlockReleaseBuffer(metaBuffer);
MemoryContextSwitchTo(oldCtx);
MemoryContextDelete(insertCtx);
return false;
}
/* Didn't fit, must try other pages */
GenericXLogAbort(state);
UnlockReleaseBuffer(buffer);
nStart++;
}
/*
* Didn't find place to insert in notFullPage array. Allocate new page.
* (XXX is it good to do this while holding ex-lock on the metapage??)
*/
buffer = BloomNewBuffer(index);
page = GenericXLogRegisterBuffer(state, buffer, GENERIC_XLOG_FULL_IMAGE);
BloomInitPage(page, 0);
if (!BloomPageAddItem(&blstate, page, itup))
{
/* We shouldn't be here since we're inserting to an empty page */
elog(ERROR, "could not add new bloom tuple to empty page");
}
/* Reset notFullPage array to contain just this new page */
metaData->nStart = 0;
metaData->nEnd = 1;
metaData->notFullPage[0] = BufferGetBlockNumber(buffer);
/* Apply the changes, clean up, and exit */
GenericXLogFinish(state);
UnlockReleaseBuffer(buffer);
UnlockReleaseBuffer(metaBuffer);
MemoryContextSwitchTo(oldCtx);
MemoryContextDelete(insertCtx);
return false;
}
