/*
* systable_beginscan_ordered --- set up for ordered catalog scan
*
* These routines have essentially the same API as systable_beginscan etc,
* except that they guarantee to return multiple matching tuples in
* index order. Also, for largely historical reasons, the index to use
* is opened and locked by the caller, not here.
*
* Currently we do not support non-index-based scans here. (In principle
* we could do a heapscan and sort, but the uses are in places that
* probably don't need to still work with corrupted catalog indexes.)
* For the moment, therefore, these functions are merely the thinnest of
* wrappers around index_beginscan/index_getnext. The main reason for their
* existence is to centralize possible future support of lossy operators
* in catalog scans.
*/
SysScanDesc
systable_beginscan_ordered(Relation heapRelation,
Relation indexRelation,
Snapshot snapshot,
int nkeys, ScanKey key)
{
SysScanDesc sysscan;
int i;
/* REINDEX can probably be a hard error here ... */
if (ReindexIsProcessingIndex(RelationGetRelid(indexRelation)))
elog(ERROR, "cannot do ordered scan on index \"%s\", because it is being reindexed",
RelationGetRelationName(indexRelation));
/* ... but we only throw a warning about violating IgnoreSystemIndexes */
if (IgnoreSystemIndexes)
elog(WARNING, "using index \"%s\" despite IgnoreSystemIndexes",
RelationGetRelationName(indexRelation));
sysscan = (SysScanDesc) palloc(sizeof(SysScanDescData));
sysscan->heap_rel = heapRelation;
sysscan->irel = indexRelation;
if (snapshot == NULL)
{
Oid relid = RelationGetRelid(heapRelation);
snapshot = RegisterSnapshot(GetCatalogSnapshot(relid));
sysscan->snapshot = snapshot;
}
else
{
/* Caller is responsible for any snapshot. */
sysscan->snapshot = NULL;
}
/* Change attribute numbers to be index column numbers. */
for (i = 0; i < nkeys; i++)
{
int j;
for (j = 0; j < indexRelation->rd_index->indnatts; j++)
{
if (key[i].sk_attno == indexRelation->rd_index->indkey.values[j])
{
key[i].sk_attno = j + 1;
break;
}
}
if (j == indexRelation->rd_index->indnatts)
elog(ERROR, "column is not in index");
}
sysscan->iscan = index_beginscan(heapRelation, indexRelation,
snapshot, nkeys, 0);
index_rescan(sysscan->iscan, key, nkeys, NULL, 0);
sysscan->scan = NULL;
return sysscan;
}
/*
* Create partitions and return an OID of the partition that contain value
*/
Oid
create_partitions(Oid relid, Datum value, Oid value_type, bool *crashed)
{
int ret;
RangeEntry *ranges;
Datum vals[2];
Oid oids[] = {OIDOID, value_type};
bool nulls[] = {false, false};
char *sql;
bool found;
int pos;
PartRelationInfo *prel;
RangeRelation *rangerel;
FmgrInfo cmp_func;
char *schema;
*crashed = false;
schema = get_extension_schema();
prel = get_pathman_relation_info(relid, NULL);
rangerel = get_pathman_range_relation(relid, NULL);
ranges = dsm_array_get_pointer(&rangerel->ranges);
/* Comparison function */
cmp_func = *get_cmp_func(value_type, prel->atttype);
vals[0] = ObjectIdGetDatum(relid);
vals[1] = value;
/* Perform PL procedure */
sql = psprintf("SELECT %s.append_partitions_on_demand_internal($1, $2)",
schema);
PG_TRY();
{
ret = SPI_execute_with_args(sql, 2, oids, vals, nulls, false, 0);
if (ret > 0)
{
/* Update relation info */
free_dsm_array(&rangerel->ranges);
free_dsm_array(&prel->children);
load_check_constraints(relid, GetCatalogSnapshot(relid));
}
}
PG_CATCH();
{
elog(WARNING, "Attempt to create new partitions failed");
if (crashed != NULL)
*crashed = true;
return 0;
}
PG_END_TRY();
/* Repeat binary search */
ranges = dsm_array_get_pointer(&rangerel->ranges);
pos = range_binary_search(rangerel, &cmp_func, value, &found);
if (found)
return ranges[pos].child_oid;
return 0;
}
TableScanDesc
table_beginscan_catalog(Relation relation, int nkeys, struct ScanKeyData *key)
{
Oid relid = RelationGetRelid(relation);
Snapshot snapshot = RegisterSnapshot(GetCatalogSnapshot(relid));
return relation->rd_tableam->scan_begin(relation, snapshot, nkeys, key, NULL,
true, true, true, false, false, true);
}
/*
* systable_recheck_tuple --- recheck visibility of most-recently-fetched tuple
*
* In particular, determine if this tuple would be visible to a catalog scan
* that started now. We don't handle the case of a non-MVCC scan snapshot,
* because no caller needs that yet.
*
* This is useful to test whether an object was deleted while we waited to
* acquire lock on it.
*
* Note: we don't actually *need* the tuple to be passed in, but it's a
* good crosscheck that the caller is interested in the right tuple.
*/
bool
systable_recheck_tuple(SysScanDesc sysscan, HeapTuple tup)
{
Snapshot freshsnap;
bool result;
/*
* Trust that LockBuffer() and HeapTupleSatisfiesMVCC() do not themselves
* acquire snapshots, so we need not register the snapshot. Those
* facilities are too low-level to have any business scanning tables.
*/
freshsnap = GetCatalogSnapshot(RelationGetRelid(sysscan->heap_rel));
if (sysscan->irel)
{
IndexScanDesc scan = sysscan->iscan;
Assert(IsMVCCSnapshot(scan->xs_snapshot));
Assert(tup == &scan->xs_ctup);
Assert(BufferIsValid(scan->xs_cbuf));
/* must hold a buffer lock to call HeapTupleSatisfiesVisibility */
LockBuffer(scan->xs_cbuf, BUFFER_LOCK_SHARE);
result = HeapTupleSatisfiesVisibility(tup, freshsnap, scan->xs_cbuf);
LockBuffer(scan->xs_cbuf, BUFFER_LOCK_UNLOCK);
}
else
{
HeapScanDesc scan = sysscan->scan;
Assert(IsMVCCSnapshot(scan->rs_snapshot));
Assert(tup == &scan->rs_ctup);
Assert(BufferIsValid(scan->rs_cbuf));
/* must hold a buffer lock to call HeapTupleSatisfiesVisibility */
LockBuffer(scan->rs_cbuf, BUFFER_LOCK_SHARE);
result = HeapTupleSatisfiesVisibility(tup, freshsnap, scan->rs_cbuf);
LockBuffer(scan->rs_cbuf, BUFFER_LOCK_UNLOCK);
}
return result;
}
/*
* Load GUC settings from pg_db_role_setting.
*
* We try specific settings for the database/role combination, as well as
* general for this database and for this user.
*/
static void
process_settings(Oid databaseid, Oid roleid)
{
Relation relsetting;
Snapshot snapshot;
if (!IsUnderPostmaster)
return;
relsetting = heap_open(DbRoleSettingRelationId, AccessShareLock);
/* read all the settings under the same snapsot for efficiency */
snapshot = RegisterSnapshot(GetCatalogSnapshot(DbRoleSettingRelationId));
/* Later settings are ignored if set earlier. */
ApplySetting(snapshot, databaseid, roleid, relsetting, PGC_S_DATABASE_USER);
ApplySetting(snapshot, InvalidOid, roleid, relsetting, PGC_S_USER);
ApplySetting(snapshot, databaseid, InvalidOid, relsetting, PGC_S_DATABASE);
ApplySetting(snapshot, InvalidOid, InvalidOid, relsetting, PGC_S_GLOBAL);
UnregisterSnapshot(snapshot);
heap_close(relsetting, AccessShareLock);
}
/*
* systable_beginscan --- set up for heap-or-index scan
*
* rel: catalog to scan, already opened and suitably locked
* indexId: OID of index to conditionally use
* indexOK: if false, forces a heap scan (see notes below)
* snapshot: time qual to use (NULL for a recent catalog snapshot)
* nkeys, key: scan keys
*
* The attribute numbers in the scan key should be set for the heap case.
* If we choose to index, we reset them to 1..n to reference the index
* columns. Note this means there must be one scankey qualification per
* index column! This is checked by the Asserts in the normal, index-using
* case, but won't be checked if the heapscan path is taken.
*
* The routine checks the normal cases for whether an indexscan is safe,
* but caller can make additional checks and pass indexOK=false if needed.
* In standard case indexOK can simply be constant TRUE.
*/
SysScanDesc
systable_beginscan(Relation heapRelation,
Oid indexId,
bool indexOK,
Snapshot snapshot,
int nkeys, ScanKey key)
{
SysScanDesc sysscan;
Relation irel;
if (indexOK &&
!IgnoreSystemIndexes &&
!ReindexIsProcessingIndex(indexId))
irel = index_open(indexId, AccessShareLock);
else
irel = NULL;
sysscan = (SysScanDesc) palloc(sizeof(SysScanDescData));
sysscan->heap_rel = heapRelation;
sysscan->irel = irel;
if (snapshot == NULL)
{
Oid relid = RelationGetRelid(heapRelation);
snapshot = RegisterSnapshot(GetCatalogSnapshot(relid));
sysscan->snapshot = snapshot;
}
else
{
/* Caller is responsible for any snapshot. */
sysscan->snapshot = NULL;
}
if (irel)
{
int i;
/* Change attribute numbers to be index column numbers. */
for (i = 0; i < nkeys; i++)
{
int j;
for (j = 0; j < irel->rd_index->indnatts; j++)
{
if (key[i].sk_attno == irel->rd_index->indkey.values[j])
{
key[i].sk_attno = j + 1;
break;
}
}
if (j == irel->rd_index->indnatts)
elog(ERROR, "column is not in index");
}
sysscan->iscan = index_beginscan(heapRelation, irel,
snapshot, nkeys, 0);
index_rescan(sysscan->iscan, key, nkeys, NULL, 0);
sysscan->scan = NULL;
}
else
{
/*
* We disallow synchronized scans when forced to use a heapscan on a
* catalog. In most cases the desired rows are near the front, so
* that the unpredictable start point of a syncscan is a serious
* disadvantage; and there are no compensating advantages, because
* it's unlikely that such scans will occur in parallel.
*/
sysscan->scan = heap_beginscan_strat(heapRelation, snapshot,
nkeys, key,
true, false);
sysscan->iscan = NULL;
}
return sysscan;
}
/*
* Performs a compaction of an append-only relation in column-orientation.
*
* In non-utility mode, all compaction segment files should be
* marked as in-use/in-compaction in the appendonlywriter.c code. If
* set, the insert_segno should also be marked as in-use.
* When the insert segno is negative, only truncate to eof operations
* can be executed.
*
* The caller is required to hold either an AccessExclusiveLock (vacuum full)
* or a ShareLock on the relation.
*/
void
AOCSCompact(Relation aorel,
List *compaction_segno,
int insert_segno,
bool isFull)
{
const char *relname;
int total_segfiles;
AOCSFileSegInfo **segfile_array;
AOCSInsertDesc insertDesc = NULL;
int i,
segno;
LockAcquireResult acquireResult;
AOCSFileSegInfo *fsinfo;
Snapshot appendOnlyMetaDataSnapshot = RegisterSnapshot(GetCatalogSnapshot(InvalidOid));
Assert(RelationIsAoCols(aorel));
Assert(Gp_role == GP_ROLE_EXECUTE || Gp_role == GP_ROLE_UTILITY);
Assert(insert_segno >= 0);
relname = RelationGetRelationName(aorel);
elogif(Debug_appendonly_print_compaction, LOG,
"Compact AO relation %s", relname);
/* Get information about all the file segments we need to scan */
segfile_array = GetAllAOCSFileSegInfo(aorel, appendOnlyMetaDataSnapshot, &total_segfiles);
if (insert_segno >= 0)
{
insertDesc = aocs_insert_init(aorel, insert_segno, false);
}
for (i = 0; i < total_segfiles; i++)
{
segno = segfile_array[i]->segno;
if (!list_member_int(compaction_segno, segno))
{
continue;
}
if (segno == insert_segno)
{
/* We cannot compact the segment file we are inserting to. */
continue;
}
/*
* Try to get the transaction write-lock for the Append-Only segment
* file.
*
* NOTE: This is a transaction scope lock that must be held until
* commit / abort.
*/
acquireResult = LockRelationAppendOnlySegmentFile(
&aorel->rd_node,
segfile_array[i]->segno,
AccessExclusiveLock,
/* dontWait */ true);
if (acquireResult == LOCKACQUIRE_NOT_AVAIL)
{
elog(DEBUG5, "compaction skips AOCS segfile %d, "
"relation %s", segfile_array[i]->segno, relname);
continue;
}
/* Re-fetch under the write lock to get latest committed eof. */
fsinfo = GetAOCSFileSegInfo(aorel, appendOnlyMetaDataSnapshot, segno);
/*
* This should not occur since this segfile info was found by the
* "all" method, but better to catch for trouble shooting (possibly
* index corruption?)
*/
if (fsinfo == NULL)
elog(ERROR, "file seginfo for AOCS relation %s %u/%u/%u (segno=%u) is missing",
relname,
aorel->rd_node.spcNode,
aorel->rd_node.dbNode,
aorel->rd_node.relNode,
segno);
if (AppendOnlyCompaction_ShouldCompact(aorel,
fsinfo->segno, fsinfo->total_tupcount, isFull,
appendOnlyMetaDataSnapshot))
{
AOCSSegmentFileFullCompaction(aorel, insertDesc, fsinfo,
appendOnlyMetaDataSnapshot);
}
pfree(fsinfo);
}
if (insertDesc != NULL)
aocs_insert_finish(insertDesc);
if (segfile_array)
{
FreeAllAOCSSegFileInfo(segfile_array, total_segfiles);
pfree(segfile_array);
}
UnregisterSnapshot(appendOnlyMetaDataSnapshot);
}
/*
* Performs a compaction of an append-only AOCS relation.
*
* In non-utility mode, all compaction segment files should be
* marked as in-use/in-compaction in the appendonlywriter.c code.
*
*/
void
AOCSDrop(Relation aorel,
List *compaction_segno)
{
const char *relname;
int total_segfiles;
AOCSFileSegInfo **segfile_array;
int i,
segno;
LockAcquireResult acquireResult;
AOCSFileSegInfo *fsinfo;
Snapshot appendOnlyMetaDataSnapshot = RegisterSnapshot(GetCatalogSnapshot(InvalidOid));
Assert(Gp_role == GP_ROLE_EXECUTE || Gp_role == GP_ROLE_UTILITY);
Assert(RelationIsAoCols(aorel));
relname = RelationGetRelationName(aorel);
elogif(Debug_appendonly_print_compaction, LOG,
"Drop AOCS relation %s", relname);
/* Get information about all the file segments we need to scan */
segfile_array = GetAllAOCSFileSegInfo(aorel,
appendOnlyMetaDataSnapshot, &total_segfiles);
for (i = 0; i < total_segfiles; i++)
{
segno = segfile_array[i]->segno;
if (!list_member_int(compaction_segno, segno))
{
continue;
}
/*
* Try to get the transaction write-lock for the Append-Only segment
* file.
*
* NOTE: This is a transaction scope lock that must be held until
* commit / abort.
*/
acquireResult = LockRelationAppendOnlySegmentFile(
&aorel->rd_node,
segfile_array[i]->segno,
AccessExclusiveLock,
/* dontWait */ true);
if (acquireResult == LOCKACQUIRE_NOT_AVAIL)
{
elog(DEBUG5, "drop skips AOCS segfile %d, "
"relation %s", segfile_array[i]->segno, relname);
continue;
}
/* Re-fetch under the write lock to get latest committed eof. */
fsinfo = GetAOCSFileSegInfo(aorel, appendOnlyMetaDataSnapshot, segno);
if (fsinfo->state == AOSEG_STATE_AWAITING_DROP)
{
Assert(HasLockForSegmentFileDrop(aorel));
AOCSCompaction_DropSegmentFile(aorel, segno);
ClearAOCSFileSegInfo(aorel, segno, AOSEG_STATE_DEFAULT);
}
pfree(fsinfo);
}
if (segfile_array)
{
FreeAllAOCSSegFileInfo(segfile_array, total_segfiles);
pfree(segfile_array);
}
UnregisterSnapshot(appendOnlyMetaDataSnapshot);
}
/*
* Truncates each segment file to the AOCS relation to its EOF.
* If we cannot get a lock on the segment file (because e.g. a concurrent insert)
* the segment file is skipped.
*/
void
AOCSTruncateToEOF(Relation aorel)
{
const char *relname;
int total_segfiles;
AOCSFileSegInfo **segfile_array;
int i,
segno;
LockAcquireResult acquireResult;
AOCSFileSegInfo *fsinfo;
Snapshot appendOnlyMetaDataSnapshot = RegisterSnapshot(GetCatalogSnapshot(InvalidOid));
Assert(RelationIsAoCols(aorel));
relname = RelationGetRelationName(aorel);
elogif(Debug_appendonly_print_compaction, LOG,
"Compact AO relation %s", relname);
/* Get information about all the file segments we need to scan */
segfile_array = GetAllAOCSFileSegInfo(aorel, appendOnlyMetaDataSnapshot, &total_segfiles);
for (i = 0; i < total_segfiles; i++)
{
segno = segfile_array[i]->segno;
/*
* Try to get the transaction write-lock for the Append-Only segment
* file.
*
* NOTE: This is a transaction scope lock that must be held until
* commit / abort.
*/
acquireResult = LockRelationAppendOnlySegmentFile(
&aorel->rd_node,
segfile_array[i]->segno,
AccessExclusiveLock,
/* dontWait */ true);
if (acquireResult == LOCKACQUIRE_NOT_AVAIL)
{
elog(DEBUG5, "truncate skips AO segfile %d, "
"relation %s", segfile_array[i]->segno, relname);
continue;
}
/* Re-fetch under the write lock to get latest committed eof. */
fsinfo = GetAOCSFileSegInfo(aorel, appendOnlyMetaDataSnapshot, segno);
/*
* This should not occur since this segfile info was found by the
* "all" method, but better to catch for trouble shooting (possibly
* index corruption?)
*/
if (fsinfo == NULL)
elog(ERROR, "file seginfo for AOCS relation %s %u/%u/%u (segno=%u) is missing",
relname,
aorel->rd_node.spcNode,
aorel->rd_node.dbNode,
aorel->rd_node.relNode,
segno);
AOCSSegmentFileTruncateToEOF(aorel, fsinfo);
pfree(fsinfo);
}
if (segfile_array)
{
FreeAllAOCSSegFileInfo(segfile_array, total_segfiles);
pfree(segfile_array);
}
UnregisterSnapshot(appendOnlyMetaDataSnapshot);
}