/*
* Write a TRUNCATE xlog record
*/
static void
WriteTruncateXlogRec(int pageno)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(int));
(void) XLogInsert(RM_COMMIT_TS_ID, COMMIT_TS_TRUNCATE);
}
/*
* Write XLOG record describing a delete of leaf index tuples marked as DEAD
* during new tuple insertion. One may think that this case is already covered
* by gistXLogUpdate(). But deletion of index tuples might conflict with
* standby queries and needs special handling.
*/
XLogRecPtr
gistXLogDelete(Buffer buffer, OffsetNumber *todelete, int ntodelete,
RelFileNode hnode)
{
gistxlogDelete xlrec;
XLogRecPtr recptr;
xlrec.hnode = hnode;
xlrec.ntodelete = ntodelete;
XLogBeginInsert();
XLogRegisterData((char *) &xlrec, SizeOfGistxlogDelete);
/*
* We need the target-offsets array whether or not we store the whole
* buffer, to allow us to find the latestRemovedXid on a standby server.
*/
XLogRegisterData((char *) todelete, ntodelete * sizeof(OffsetNumber));
XLogRegisterBuffer(0, buffer, REGBUF_STANDARD);
recptr = XLogInsert(RM_GIST_ID, XLOG_GIST_DELETE);
return recptr;
}
/*
* Write XLOG record describing a page update. The update can include any
* number of deletions and/or insertions of tuples on a single index page.
*
* If this update inserts a downlink for a split page, also record that
* the F_FOLLOW_RIGHT flag on the child page is cleared and NSN set.
*
* Note that both the todelete array and the tuples are marked as belonging
* to the target buffer; they need not be stored in XLOG if XLogInsert decides
* to log the whole buffer contents instead.
*/
XLogRecPtr
gistXLogUpdate(Buffer buffer,
OffsetNumber *todelete, int ntodelete,
IndexTuple *itup, int ituplen,
Buffer leftchildbuf)
{
gistxlogPageUpdate xlrec;
int i;
XLogRecPtr recptr;
xlrec.ntodelete = ntodelete;
xlrec.ntoinsert = ituplen;
XLogBeginInsert();
XLogRegisterData((char *) &xlrec, sizeof(gistxlogPageUpdate));
XLogRegisterBuffer(0, buffer, REGBUF_STANDARD);
XLogRegisterBufData(0, (char *) todelete, sizeof(OffsetNumber) * ntodelete);
/* new tuples */
for (i = 0; i < ituplen; i++)
XLogRegisterBufData(0, (char *) (itup[i]), IndexTupleSize(itup[i]));
/*
* Include a full page image of the child buf. (only necessary if a
* checkpoint happened since the child page was split)
*/
if (BufferIsValid(leftchildbuf))
XLogRegisterBuffer(1, leftchildbuf, REGBUF_STANDARD);
recptr = XLogInsert(RM_GIST_ID, XLOG_GIST_PAGE_UPDATE);
return recptr;
}
/*
* Remove the visibility map fork for a relation. If there turn out to be
* any bugs in the visibility map code that require rebuilding the VM, this
* provides users with a way to do it that is cleaner than shutting down the
* server and removing files by hand.
*
* This is a cut-down version of RelationTruncate.
*/
Datum
pg_truncate_visibility_map(PG_FUNCTION_ARGS)
{
Oid relid = PG_GETARG_OID(0);
Relation rel;
rel = relation_open(relid, AccessExclusiveLock);
if (rel->rd_rel->relkind != RELKIND_RELATION &&
rel->rd_rel->relkind != RELKIND_MATVIEW &&
rel->rd_rel->relkind != RELKIND_TOASTVALUE)
ereport(ERROR,
(errcode(ERRCODE_WRONG_OBJECT_TYPE),
errmsg("\"%s\" is not a table, materialized view, or TOAST table",
RelationGetRelationName(rel))));
RelationOpenSmgr(rel);
rel->rd_smgr->smgr_vm_nblocks = InvalidBlockNumber;
visibilitymap_truncate(rel, 0);
if (RelationNeedsWAL(rel))
{
xl_smgr_truncate xlrec;
xlrec.blkno = 0;
xlrec.rnode = rel->rd_node;
xlrec.flags = SMGR_TRUNCATE_VM;
XLogBeginInsert();
XLogRegisterData((char *) &xlrec, sizeof(xlrec));
XLogInsert(RM_SMGR_ID, XLOG_SMGR_TRUNCATE | XLR_SPECIAL_REL_UPDATE);
}
/*
* Release the lock right away, not at commit time.
*
* It would be a problem to release the lock prior to commit if this
* truncate operation sends any transactional invalidation messages. Other
* backends would potentially be able to lock the relation without
* processing them in the window of time between when we release the lock
* here and when we sent the messages at our eventual commit. However,
* we're currently only sending a non-transactional smgr invalidation,
* which will have been posted to shared memory immediately from within
* visibilitymap_truncate. Therefore, there should be no race here.
*
* The reason why it's desirable to release the lock early here is because
* of the possibility that someone will need to use this to blow away many
* visibility map forks at once. If we can't release the lock until
* commit time, the transaction doing this will accumulate
* AccessExclusiveLocks on all of those relations at the same time, which
* is undesirable. However, if this turns out to be unsafe we may have no
* choice...
*/
relation_close(rel, AccessExclusiveLock);
/* Nothing to return. */
PG_RETURN_VOID();
}
/*
* Write a ZEROPAGE xlog record
*/
static void
WriteZeroPageXlogRec(int pageno)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(int));
(void) XLogInsert(RM_COMMIT_TS_ID, COMMIT_TS_ZEROPAGE);
}
/*
* Write logical decoding message into XLog.
*/
XLogRecPtr
LogLogicalMessage(const char *prefix, const char *message, size_t size,
bool transactional)
{
xl_logical_message xlrec;
/*
* Force xid to be allocated if we're emitting a transactional message.
*/
if (transactional)
{
Assert(IsTransactionState());
GetCurrentTransactionId();
}
xlrec.dbId = MyDatabaseId;
xlrec.transactional = transactional;
xlrec.prefix_size = strlen(prefix) + 1;
xlrec.message_size = size;
XLogBeginInsert();
XLogRegisterData((char *) &xlrec, SizeOfLogicalMessage);
XLogRegisterData((char *) prefix, xlrec.prefix_size);
XLogRegisterData((char *) message, size);
/* allow origin filtering */
XLogSetRecordFlags(XLOG_INCLUDE_ORIGIN);
return XLogInsert(RM_LOGICALMSG_ID, XLOG_LOGICAL_MESSAGE);
}
/*
* Write a WAL record containing a full image of a page. Caller is responsible
* for writing the page to disk after calling this routine.
*
* Note: If you're using this function, you should be building pages in private
* memory and writing them directly to smgr. If you're using buffers, call
* log_newpage_buffer instead.
*
* If the page follows the standard page layout, with a PageHeader and unused
* space between pd_lower and pd_upper, set 'page_std' to TRUE. That allows
* the unused space to be left out from the WAL record, making it smaller.
*/
XLogRecPtr
log_newpage(RelFileNode *rnode, ForkNumber forkNum, BlockNumber blkno,
Page page, bool page_std)
{
int flags;
XLogRecPtr recptr;
flags = REGBUF_FORCE_IMAGE;
if (page_std)
flags |= REGBUF_STANDARD;
XLogBeginInsert();
XLogRegisterBlock(0, rnode, forkNum, blkno, page, flags);
recptr = XLogInsert(RM_XLOG_ID, XLOG_FPI);
/*
* The page may be uninitialized. If so, we can't set the LSN because that
* would corrupt the page.
*/
if (!PageIsNew(page))
{
PageSetLSN(page, recptr);
}
return recptr;
}
/*
* Write a TRUNCATE xlog record
*/
static void
WriteTruncateXlogRec(int pageno, TransactionId oldestXid)
{
xl_commit_ts_truncate xlrec;
xlrec.pageno = pageno;
xlrec.oldestXid = oldestXid;
XLogBeginInsert();
XLogRegisterData((char *) (&xlrec), SizeOfCommitTsTruncate);
(void) XLogInsert(RM_COMMIT_TS_ID, COMMIT_TS_TRUNCATE);
}
/*
* Write WAL record of a page split.
*/
XLogRecPtr
gistXLogSplit(bool page_is_leaf,
SplitedPageLayout *dist,
BlockNumber origrlink, GistNSN orignsn,
Buffer leftchildbuf, bool markfollowright)
{
gistxlogPageSplit xlrec;
SplitedPageLayout *ptr;
int npage = 0;
XLogRecPtr recptr;
int i;
for (ptr = dist; ptr; ptr = ptr->next)
npage++;
xlrec.origrlink = origrlink;
xlrec.orignsn = orignsn;
xlrec.origleaf = page_is_leaf;
xlrec.npage = (uint16) npage;
xlrec.markfollowright = markfollowright;
XLogBeginInsert();
/*
* Include a full page image of the child buf. (only necessary if a
* checkpoint happened since the child page was split)
*/
if (BufferIsValid(leftchildbuf))
XLogRegisterBuffer(0, leftchildbuf, REGBUF_STANDARD);
/*
* NOTE: We register a lot of data. The caller must've called
* XLogEnsureRecordSpace() to prepare for that. We cannot do it here,
* because we're already in a critical section. If you change the number
* of buffer or data registrations here, make sure you modify the
* XLogEnsureRecordSpace() calls accordingly!
*/
XLogRegisterData((char *) &xlrec, sizeof(gistxlogPageSplit));
i = 1;
for (ptr = dist; ptr; ptr = ptr->next)
{
XLogRegisterBuffer(i, ptr->buffer, REGBUF_WILL_INIT);
XLogRegisterBufData(i, (char *) &(ptr->block.num), sizeof(int));
XLogRegisterBufData(i, (char *) ptr->list, ptr->lenlist);
i++;
}
recptr = XLogInsert(RM_GIST_ID, XLOG_GIST_PAGE_SPLIT);
return recptr;
}
/*
* Perform XLogInsert of an XLOG_SMGR_CREATE record to WAL.
*/
void
log_smgrcreate(RelFileNode *rnode, ForkNumber forkNum)
{
xl_smgr_create xlrec;
/*
* Make an XLOG entry reporting the file creation.
*/
xlrec.rnode = *rnode;
xlrec.forkNum = forkNum;
XLogBeginInsert();
XLogRegisterData((char *) &xlrec, sizeof(xlrec));
XLogInsert(RM_SMGR_ID, XLOG_SMGR_CREATE | XLR_SPECIAL_REL_UPDATE);
}
/*
* Place tuple on page and fills WAL record
*
* If the tuple doesn't fit, returns false without modifying the page.
*
* On insertion to an internal node, in addition to inserting the given item,
* the downlink of the existing item at 'off' is updated to point to
* 'updateblkno'.
*
* On INSERTED, registers the buffer as buffer ID 0, with data.
* On SPLIT, returns rdata that represents the split pages in *prdata.
*/
static GinPlaceToPageRC
entryPlaceToPage(GinBtree btree, Buffer buf, GinBtreeStack *stack,
void *insertPayload, BlockNumber updateblkno,
Page *newlpage, Page *newrpage)
{
GinBtreeEntryInsertData *insertData = insertPayload;
Page page = BufferGetPage(buf, NULL, NULL, BGP_NO_SNAPSHOT_TEST);
OffsetNumber off = stack->off;
OffsetNumber placed;
/* this must be static so it can be returned to caller. */
static ginxlogInsertEntry data;
/* quick exit if it doesn't fit */
if (!entryIsEnoughSpace(btree, buf, off, insertData))
{
entrySplitPage(btree, buf, stack, insertPayload, updateblkno,
newlpage, newrpage);
return SPLIT;
}
START_CRIT_SECTION();
entryPreparePage(btree, page, off, insertData, updateblkno);
placed = PageAddItem(page,
(Item) insertData->entry,
IndexTupleSize(insertData->entry),
off, false, false);
if (placed != off)
elog(ERROR, "failed to add item to index page in \"%s\"",
RelationGetRelationName(btree->index));
if (RelationNeedsWAL(btree->index))
{
data.isDelete = insertData->isDelete;
data.offset = off;
XLogBeginInsert();
XLogRegisterBuffer(0, buf, REGBUF_STANDARD);
XLogRegisterBufData(0, (char *) &data,
offsetof(ginxlogInsertEntry, tuple));
XLogRegisterBufData(0, (char *) insertData->entry,
IndexTupleSize(insertData->entry));
}
return INSERTED;
}
/*
* log_split_page() -- Log the split operation
*
* We log the split operation when the new page in new bucket gets full,
* so we log the entire page.
*
* 'buf' must be locked by the caller which is also responsible for unlocking
* it.
*/
static void
log_split_page(Relation rel, Buffer buf)
{
if (RelationNeedsWAL(rel))
{
XLogRecPtr recptr;
XLogBeginInsert();
XLogRegisterBuffer(0, buf, REGBUF_FORCE_IMAGE | REGBUF_STANDARD);
recptr = XLogInsert(RM_HASH_ID, XLOG_HASH_SPLIT_PAGE);
PageSetLSN(BufferGetPage(buf), recptr);
}
}
/*
* Write a SETTS xlog record
*/
static void
WriteSetTimestampXlogRec(TransactionId mainxid, int nsubxids,
TransactionId *subxids, TimestampTz timestamp,
RepOriginId nodeid)
{
xl_commit_ts_set record;
record.timestamp = timestamp;
record.nodeid = nodeid;
record.mainxid = mainxid;
XLogBeginInsert();
XLogRegisterData((char *) &record,
offsetof(xl_commit_ts_set, mainxid) +
sizeof(TransactionId));
XLogRegisterData((char *) subxids, nsubxids * sizeof(TransactionId));
XLogInsert(RM_COMMIT_TS_ID, COMMIT_TS_SETTS);
}
/*
* Write the given statistics to the index's metapage
*
* Note: nPendingPages and ginVersion are *not* copied over
*/
void
ginUpdateStats(Relation index, const GinStatsData *stats)
{
Buffer metabuffer;
Page metapage;
GinMetaPageData *metadata;
metabuffer = ReadBuffer(index, GIN_METAPAGE_BLKNO);
LockBuffer(metabuffer, GIN_EXCLUSIVE);
metapage = BufferGetPage(metabuffer);
metadata = GinPageGetMeta(metapage);
START_CRIT_SECTION();
metadata->nTotalPages = stats->nTotalPages;
metadata->nEntryPages = stats->nEntryPages;
metadata->nDataPages = stats->nDataPages;
metadata->nEntries = stats->nEntries;
MarkBufferDirty(metabuffer);
if (RelationNeedsWAL(index))
{
XLogRecPtr recptr;
ginxlogUpdateMeta data;
data.node = index->rd_node;
data.ntuples = 0;
data.newRightlink = data.prevTail = InvalidBlockNumber;
memcpy(&data.metadata, metadata, sizeof(GinMetaPageData));
XLogBeginInsert();
XLogRegisterData((char *) &data, sizeof(ginxlogUpdateMeta));
XLogRegisterBuffer(0, metabuffer, REGBUF_WILL_INIT);
recptr = XLogInsert(RM_GIN_ID, XLOG_GIN_UPDATE_META_PAGE);
PageSetLSN(metapage, recptr);
}
UnlockReleaseBuffer(metabuffer);
END_CRIT_SECTION();
}
/*
* Create a WAL record for vacuuming entry tree leaf page.
*/
static void
xlogVacuumPage(Relation index, Buffer buffer)
{
Page page = BufferGetPage(buffer);
XLogRecPtr recptr;
/* This is only used for entry tree leaf pages. */
Assert(!GinPageIsData(page));
Assert(GinPageIsLeaf(page));
if (!RelationNeedsWAL(index))
return;
/*
* Always create a full image, we don't track the changes on the page at
* any more fine-grained level. This could obviously be improved...
*/
XLogBeginInsert();
XLogRegisterBuffer(0, buffer, REGBUF_FORCE_IMAGE | REGBUF_STANDARD);
recptr = XLogInsert(RM_GIN_ID, XLOG_GIN_VACUUM_PAGE);
PageSetLSN(page, recptr);
}
/*
* Create a table space
*
* Only superusers can create a tablespace. This seems a reasonable restriction
* since we're determining the system layout and, anyway, we probably have
* root if we're doing this kind of activity
*/
Oid
CreateTableSpace(CreateTableSpaceStmt *stmt)
{
#ifdef HAVE_SYMLINK
Relation rel;
Datum values[Natts_pg_tablespace];
bool nulls[Natts_pg_tablespace];
HeapTuple tuple;
Oid tablespaceoid;
char *location;
Oid ownerId;
Datum newOptions;
/* Must be super user */
if (!superuser())
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
errmsg("permission denied to create tablespace \"%s\"",
stmt->tablespacename),
errhint("Must be superuser to create a tablespace.")));
/* However, the eventual owner of the tablespace need not be */
if (stmt->owner)
ownerId = get_rolespec_oid(stmt->owner, false);
else
ownerId = GetUserId();
/* Unix-ify the offered path, and strip any trailing slashes */
location = pstrdup(stmt->location);
canonicalize_path(location);
/* disallow quotes, else CREATE DATABASE would be at risk */
if (strchr(location, '\''))
ereport(ERROR,
(errcode(ERRCODE_INVALID_NAME),
errmsg("tablespace location cannot contain single quotes")));
/*
* Allowing relative paths seems risky
*
* this also helps us ensure that location is not empty or whitespace
*/
if (!is_absolute_path(location))
ereport(ERROR,
(errcode(ERRCODE_INVALID_OBJECT_DEFINITION),
errmsg("tablespace location must be an absolute path")));
/*
* Check that location isn't too long. Remember that we're going to append
* 'PG_XXX/<dboid>/<relid>_<fork>.<nnn>'. FYI, we never actually
* reference the whole path here, but mkdir() uses the first two parts.
*/
if (strlen(location) + 1 + strlen(TABLESPACE_VERSION_DIRECTORY) + 1 +
OIDCHARS + 1 + OIDCHARS + 1 + FORKNAMECHARS + 1 + OIDCHARS > MAXPGPATH)
ereport(ERROR,
(errcode(ERRCODE_INVALID_OBJECT_DEFINITION),
errmsg("tablespace location \"%s\" is too long",
location)));
/* Warn if the tablespace is in the data directory. */
if (path_is_prefix_of_path(DataDir, location))
ereport(WARNING,
(errcode(ERRCODE_INVALID_OBJECT_DEFINITION),
errmsg("tablespace location should not be inside the data directory")));
/*
* Disallow creation of tablespaces named "pg_xxx"; we reserve this
* namespace for system purposes.
*/
if (!allowSystemTableMods && IsReservedName(stmt->tablespacename))
ereport(ERROR,
(errcode(ERRCODE_RESERVED_NAME),
errmsg("unacceptable tablespace name \"%s\"",
stmt->tablespacename),
errdetail("The prefix \"pg_\" is reserved for system tablespaces.")));
/*
* Check that there is no other tablespace by this name. (The unique
* index would catch this anyway, but might as well give a friendlier
* message.)
*/
if (OidIsValid(get_tablespace_oid(stmt->tablespacename, true)))
ereport(ERROR,
(errcode(ERRCODE_DUPLICATE_OBJECT),
errmsg("tablespace \"%s\" already exists",
stmt->tablespacename)));
/*
* Insert tuple into pg_tablespace. The purpose of doing this first is to
* lock the proposed tablename against other would-be creators. The
* insertion will roll back if we find problems below.
*/
rel = heap_open(TableSpaceRelationId, RowExclusiveLock);
MemSet(nulls, false, sizeof(nulls));
values[Anum_pg_tablespace_spcname - 1] =
DirectFunctionCall1(namein, CStringGetDatum(stmt->tablespacename));
values[Anum_pg_tablespace_spcowner - 1] =
ObjectIdGetDatum(ownerId);
nulls[Anum_pg_tablespace_spcacl - 1] = true;
/* Generate new proposed spcoptions (text array) */
newOptions = transformRelOptions((Datum) 0,
stmt->options,
NULL, NULL, false, false);
(void) tablespace_reloptions(newOptions, true);
if (newOptions != (Datum) 0)
values[Anum_pg_tablespace_spcoptions - 1] = newOptions;
else
nulls[Anum_pg_tablespace_spcoptions - 1] = true;
tuple = heap_form_tuple(rel->rd_att, values, nulls);
tablespaceoid = simple_heap_insert(rel, tuple);
CatalogUpdateIndexes(rel, tuple);
heap_freetuple(tuple);
/* Record dependency on owner */
recordDependencyOnOwner(TableSpaceRelationId, tablespaceoid, ownerId);
/* Post creation hook for new tablespace */
InvokeObjectPostCreateHook(TableSpaceRelationId, tablespaceoid, 0);
create_tablespace_directories(location, tablespaceoid);
/* Record the filesystem change in XLOG */
{
xl_tblspc_create_rec xlrec;
xlrec.ts_id = tablespaceoid;
XLogBeginInsert();
XLogRegisterData((char *) &xlrec,
offsetof(xl_tblspc_create_rec, ts_path));
XLogRegisterData((char *) location, strlen(location) + 1);
(void) XLogInsert(RM_TBLSPC_ID, XLOG_TBLSPC_CREATE);
}
/*
* Force synchronous commit, to minimize the window between creating the
* symlink on-disk and marking the transaction committed. It's not great
* that there is any window at all, but definitely we don't want to make
* it larger than necessary.
*/
ForceSyncCommit();
pfree(location);
/* We keep the lock on pg_tablespace until commit */
heap_close(rel, NoLock);
return tablespaceoid;
#else /* !HAVE_SYMLINK */
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("tablespaces are not supported on this platform")));
return InvalidOid; /* keep compiler quiet */
#endif /* HAVE_SYMLINK */
}
/*
* Write out a new shared or local map file with the given contents.
*
* The magic number and CRC are automatically updated in *newmap. On
* success, we copy the data to the appropriate permanent static variable.
*
* If write_wal is TRUE then an appropriate WAL message is emitted.
* (It will be false for bootstrap and WAL replay cases.)
*
* If send_sinval is TRUE then a SI invalidation message is sent.
* (This should be true except in bootstrap case.)
*
* If preserve_files is TRUE then the storage manager is warned not to
* delete the files listed in the map.
*
* Because this may be called during WAL replay when MyDatabaseId,
* DatabasePath, etc aren't valid, we require the caller to pass in suitable
* values. The caller is also responsible for being sure no concurrent
* map update could be happening.
*/
static void
write_relmap_file(bool shared, RelMapFile *newmap,
bool write_wal, bool send_sinval, bool preserve_files,
Oid dbid, Oid tsid, const char *dbpath)
{
int fd;
RelMapFile *realmap;
char mapfilename[MAXPGPATH];
/*
* Fill in the overhead fields and update CRC.
*/
newmap->magic = RELMAPPER_FILEMAGIC;
if (newmap->num_mappings < 0 || newmap->num_mappings > MAX_MAPPINGS)
elog(ERROR, "attempt to write bogus relation mapping");
INIT_CRC32C(newmap->crc);
COMP_CRC32C(newmap->crc, (char *) newmap, offsetof(RelMapFile, crc));
FIN_CRC32C(newmap->crc);
/*
* Open the target file. We prefer to do this before entering the
* critical section, so that an open() failure need not force PANIC.
*/
if (shared)
{
snprintf(mapfilename, sizeof(mapfilename), "global/%s",
RELMAPPER_FILENAME);
realmap = &shared_map;
}
else
{
snprintf(mapfilename, sizeof(mapfilename), "%s/%s",
dbpath, RELMAPPER_FILENAME);
realmap = &local_map;
}
fd = OpenTransientFile(mapfilename,
O_WRONLY | O_CREAT | PG_BINARY,
S_IRUSR | S_IWUSR);
if (fd < 0)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not open relation mapping file \"%s\": %m",
mapfilename)));
if (write_wal)
{
xl_relmap_update xlrec;
XLogRecPtr lsn;
/* now errors are fatal ... */
START_CRIT_SECTION();
xlrec.dbid = dbid;
xlrec.tsid = tsid;
xlrec.nbytes = sizeof(RelMapFile);
XLogBeginInsert();
XLogRegisterData((char *) (&xlrec), MinSizeOfRelmapUpdate);
XLogRegisterData((char *) newmap, sizeof(RelMapFile));
lsn = XLogInsert(RM_RELMAP_ID, XLOG_RELMAP_UPDATE);
/* As always, WAL must hit the disk before the data update does */
XLogFlush(lsn);
}
errno = 0;
if (write(fd, newmap, sizeof(RelMapFile)) != sizeof(RelMapFile))
{
/* if write didn't set errno, assume problem is no disk space */
if (errno == 0)
errno = ENOSPC;
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not write to relation mapping file \"%s\": %m",
mapfilename)));
}
/*
* We choose to fsync the data to disk before considering the task done.
* It would be possible to relax this if it turns out to be a performance
* issue, but it would complicate checkpointing --- see notes for
* CheckPointRelationMap.
*/
if (pg_fsync(fd) != 0)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not fsync relation mapping file \"%s\": %m",
mapfilename)));
if (CloseTransientFile(fd))
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not close relation mapping file \"%s\": %m",
mapfilename)));
/*
* Now that the file is safely on disk, send sinval message to let other
* backends know to re-read it. We must do this inside the critical
* section: if for some reason we fail to send the message, we have to
* force a database-wide PANIC. Otherwise other backends might continue
* execution with stale mapping information, which would be catastrophic
* as soon as others began to use the now-committed data.
*/
if (send_sinval)
CacheInvalidateRelmap(dbid);
/*
* Make sure that the files listed in the map are not deleted if the outer
* transaction aborts. This had better be within the critical section
* too: it's not likely to fail, but if it did, we'd arrive at transaction
* abort with the files still vulnerable. PANICing will leave things in a
* good state on-disk.
*
* Note: we're cheating a little bit here by assuming that mapped files
* are either in pg_global or the database's default tablespace.
*/
if (preserve_files)
{
int32 i;
for (i = 0; i < newmap->num_mappings; i++)
{
RelFileNode rnode;
rnode.spcNode = tsid;
rnode.dbNode = dbid;
rnode.relNode = newmap->mappings[i].mapfilenode;
RelationPreserveStorage(rnode, false);
}
}
/* Success, update permanent copy */
memcpy(realmap, newmap, sizeof(RelMapFile));
/* Critical section done */
if (write_wal)
END_CRIT_SECTION();
}
/*
* Vacuum a regular (non-root) leaf page
*
* We must delete tuples that are targeted for deletion by the VACUUM,
* but not move any tuples that are referenced by outside links; we assume
* those are the ones that are heads of chains.
*
* If we find a REDIRECT that was made by a concurrently-running transaction,
* we must add its target TID to pendingList. (We don't try to visit the
* target immediately, first because we don't want VACUUM locking more than
* one buffer at a time, and second because the duplicate-filtering logic
* in spgAddPendingTID is useful to ensure we can't get caught in an infinite
* loop in the face of continuous concurrent insertions.)
*
* If forPending is true, we are examining the page as a consequence of
* chasing a redirect link, not as part of the normal sequential scan.
* We still vacuum the page normally, but we don't increment the stats
* about live tuples; else we'd double-count those tuples, since the page
* has been or will be visited in the sequential scan as well.
*/
static void
vacuumLeafPage(spgBulkDeleteState *bds, Relation index, Buffer buffer,
bool forPending)
{
Page page = BufferGetPage(buffer);
spgxlogVacuumLeaf xlrec;
OffsetNumber toDead[MaxIndexTuplesPerPage];
OffsetNumber toPlaceholder[MaxIndexTuplesPerPage];
OffsetNumber moveSrc[MaxIndexTuplesPerPage];
OffsetNumber moveDest[MaxIndexTuplesPerPage];
OffsetNumber chainSrc[MaxIndexTuplesPerPage];
OffsetNumber chainDest[MaxIndexTuplesPerPage];
OffsetNumber predecessor[MaxIndexTuplesPerPage + 1];
bool deletable[MaxIndexTuplesPerPage + 1];
int nDeletable;
OffsetNumber i,
max = PageGetMaxOffsetNumber(page);
memset(predecessor, 0, sizeof(predecessor));
memset(deletable, 0, sizeof(deletable));
nDeletable = 0;
/* Scan page, identify tuples to delete, accumulate stats */
for (i = FirstOffsetNumber; i <= max; i++)
{
SpGistLeafTuple lt;
lt = (SpGistLeafTuple) PageGetItem(page,
PageGetItemId(page, i));
if (lt->tupstate == SPGIST_LIVE)
{
Assert(ItemPointerIsValid(<->heapPtr));
if (bds->callback(<->heapPtr, bds->callback_state))
{
bds->stats->tuples_removed += 1;
deletable[i] = true;
nDeletable++;
}
else
{
if (!forPending)
bds->stats->num_index_tuples += 1;
}
/* Form predecessor map, too */
if (lt->nextOffset != InvalidOffsetNumber)
{
/* paranoia about corrupted chain links */
if (lt->nextOffset < FirstOffsetNumber ||
lt->nextOffset > max ||
predecessor[lt->nextOffset] != InvalidOffsetNumber)
elog(ERROR, "inconsistent tuple chain links in page %u of index \"%s\"",
BufferGetBlockNumber(buffer),
RelationGetRelationName(index));
predecessor[lt->nextOffset] = i;
}
}
else if (lt->tupstate == SPGIST_REDIRECT)
{
SpGistDeadTuple dt = (SpGistDeadTuple) lt;
Assert(dt->nextOffset == InvalidOffsetNumber);
Assert(ItemPointerIsValid(&dt->pointer));
/*
* Add target TID to pending list if the redirection could have
* happened since VACUUM started.
*
* Note: we could make a tighter test by seeing if the xid is
* "running" according to the active snapshot; but snapmgr.c doesn't
* currently export a suitable API, and it's not entirely clear
* that a tighter test is worth the cycles anyway.
*/
if (TransactionIdFollowsOrEquals(dt->xid, bds->myXmin))
spgAddPendingTID(bds, &dt->pointer);
}
else
{
Assert(lt->nextOffset == InvalidOffsetNumber);
}
}
if (nDeletable == 0)
return; /* nothing more to do */
/*----------
* Figure out exactly what we have to do. We do this separately from
* actually modifying the page, mainly so that we have a representation
* that can be dumped into WAL and then the replay code can do exactly
* the same thing. The output of this step consists of six arrays
* describing four kinds of operations, to be performed in this order:
*
* toDead[]: tuple numbers to be replaced with DEAD tuples
* toPlaceholder[]: tuple numbers to be replaced with PLACEHOLDER tuples
* moveSrc[]: tuple numbers that need to be relocated to another offset
* (replacing the tuple there) and then replaced with PLACEHOLDER tuples
* moveDest[]: new locations for moveSrc tuples
* chainSrc[]: tuple numbers whose chain links (nextOffset) need updates
* chainDest[]: new values of nextOffset for chainSrc members
*
* It's easiest to figure out what we have to do by processing tuple
* chains, so we iterate over all the tuples (not just the deletable
* ones!) to identify chain heads, then chase down each chain and make
* work item entries for deletable tuples within the chain.
*----------
*/
xlrec.nDead = xlrec.nPlaceholder = xlrec.nMove = xlrec.nChain = 0;
for (i = FirstOffsetNumber; i <= max; i++)
{
SpGistLeafTuple head;
bool interveningDeletable;
OffsetNumber prevLive;
OffsetNumber j;
head = (SpGistLeafTuple) PageGetItem(page,
PageGetItemId(page, i));
if (head->tupstate != SPGIST_LIVE)
continue; /* can't be a chain member */
if (predecessor[i] != 0)
continue; /* not a chain head */
/* initialize ... */
interveningDeletable = false;
prevLive = deletable[i] ? InvalidOffsetNumber : i;
/* scan down the chain ... */
j = head->nextOffset;
while (j != InvalidOffsetNumber)
{
SpGistLeafTuple lt;
lt = (SpGistLeafTuple) PageGetItem(page,
PageGetItemId(page, j));
if (lt->tupstate != SPGIST_LIVE)
{
/* all tuples in chain should be live */
elog(ERROR, "unexpected SPGiST tuple state: %d",
lt->tupstate);
}
if (deletable[j])
{
/* This tuple should be replaced by a placeholder */
toPlaceholder[xlrec.nPlaceholder] = j;
xlrec.nPlaceholder++;
/* previous live tuple's chain link will need an update */
interveningDeletable = true;
}
else if (prevLive == InvalidOffsetNumber)
{
/*
* This is the first live tuple in the chain. It has to move
* to the head position.
*/
moveSrc[xlrec.nMove] = j;
moveDest[xlrec.nMove] = i;
xlrec.nMove++;
/* Chain updates will be applied after the move */
prevLive = i;
interveningDeletable = false;
}
else
{
/*
* Second or later live tuple. Arrange to re-chain it to the
* previous live one, if there was a gap.
*/
if (interveningDeletable)
{
chainSrc[xlrec.nChain] = prevLive;
chainDest[xlrec.nChain] = j;
xlrec.nChain++;
}
prevLive = j;
interveningDeletable = false;
}
j = lt->nextOffset;
}
if (prevLive == InvalidOffsetNumber)
{
/* The chain is entirely removable, so we need a DEAD tuple */
toDead[xlrec.nDead] = i;
xlrec.nDead++;
}
else if (interveningDeletable)
{
/* One or more deletions at end of chain, so close it off */
chainSrc[xlrec.nChain] = prevLive;
chainDest[xlrec.nChain] = InvalidOffsetNumber;
xlrec.nChain++;
}
}
/* sanity check ... */
if (nDeletable != xlrec.nDead + xlrec.nPlaceholder + xlrec.nMove)
elog(ERROR, "inconsistent counts of deletable tuples");
/* Do the updates */
START_CRIT_SECTION();
spgPageIndexMultiDelete(&bds->spgstate, page,
toDead, xlrec.nDead,
SPGIST_DEAD, SPGIST_DEAD,
InvalidBlockNumber, InvalidOffsetNumber);
spgPageIndexMultiDelete(&bds->spgstate, page,
toPlaceholder, xlrec.nPlaceholder,
SPGIST_PLACEHOLDER, SPGIST_PLACEHOLDER,
InvalidBlockNumber, InvalidOffsetNumber);
/*
* We implement the move step by swapping the line pointers of the source
* and target tuples, then replacing the newly-source tuples with
* placeholders. This is perhaps unduly friendly with the page data
* representation, but it's fast and doesn't risk page overflow when a
* tuple to be relocated is large.
*/
for (i = 0; i < xlrec.nMove; i++)
{
ItemId idSrc = PageGetItemId(page, moveSrc[i]);
ItemId idDest = PageGetItemId(page, moveDest[i]);
ItemIdData tmp;
tmp = *idSrc;
*idSrc = *idDest;
*idDest = tmp;
}
spgPageIndexMultiDelete(&bds->spgstate, page,
moveSrc, xlrec.nMove,
SPGIST_PLACEHOLDER, SPGIST_PLACEHOLDER,
InvalidBlockNumber, InvalidOffsetNumber);
for (i = 0; i < xlrec.nChain; i++)
{
SpGistLeafTuple lt;
lt = (SpGistLeafTuple) PageGetItem(page,
PageGetItemId(page, chainSrc[i]));
Assert(lt->tupstate == SPGIST_LIVE);
lt->nextOffset = chainDest[i];
}
MarkBufferDirty(buffer);
if (RelationNeedsWAL(index))
{
XLogRecPtr recptr;
XLogBeginInsert();
STORE_STATE(&bds->spgstate, xlrec.stateSrc);
XLogRegisterData((char *) &xlrec, SizeOfSpgxlogVacuumLeaf);
/* sizeof(xlrec) should be a multiple of sizeof(OffsetNumber) */
XLogRegisterData((char *) toDead, sizeof(OffsetNumber) * xlrec.nDead);
XLogRegisterData((char *) toPlaceholder, sizeof(OffsetNumber) * xlrec.nPlaceholder);
XLogRegisterData((char *) moveSrc, sizeof(OffsetNumber) * xlrec.nMove);
XLogRegisterData((char *) moveDest, sizeof(OffsetNumber) * xlrec.nMove);
XLogRegisterData((char *) chainSrc, sizeof(OffsetNumber) * xlrec.nChain);
XLogRegisterData((char *) chainDest, sizeof(OffsetNumber) * xlrec.nChain);
XLogRegisterBuffer(0, buffer, REGBUF_STANDARD);
recptr = XLogInsert(RM_SPGIST_ID, XLOG_SPGIST_VACUUM_LEAF);
PageSetLSN(page, recptr);
}
END_CRIT_SECTION();
}
/*
* Vacuum a root page when it is also a leaf
*
* On the root, we just delete any dead leaf tuples; no fancy business
*/
static void
vacuumLeafRoot(spgBulkDeleteState *bds, Relation index, Buffer buffer)
{
Page page = BufferGetPage(buffer);
spgxlogVacuumRoot xlrec;
OffsetNumber toDelete[MaxIndexTuplesPerPage];
OffsetNumber i,
max = PageGetMaxOffsetNumber(page);
xlrec.nDelete = 0;
/* Scan page, identify tuples to delete, accumulate stats */
for (i = FirstOffsetNumber; i <= max; i++)
{
SpGistLeafTuple lt;
lt = (SpGistLeafTuple) PageGetItem(page,
PageGetItemId(page, i));
if (lt->tupstate == SPGIST_LIVE)
{
Assert(ItemPointerIsValid(<->heapPtr));
if (bds->callback(<->heapPtr, bds->callback_state))
{
bds->stats->tuples_removed += 1;
toDelete[xlrec.nDelete] = i;
xlrec.nDelete++;
}
else
{
bds->stats->num_index_tuples += 1;
}
}
else
{
/* all tuples on root should be live */
elog(ERROR, "unexpected SPGiST tuple state: %d",
lt->tupstate);
}
}
if (xlrec.nDelete == 0)
return; /* nothing more to do */
/* Do the update */
START_CRIT_SECTION();
/* The tuple numbers are in order, so we can use PageIndexMultiDelete */
PageIndexMultiDelete(page, toDelete, xlrec.nDelete);
MarkBufferDirty(buffer);
if (RelationNeedsWAL(index))
{
XLogRecPtr recptr;
XLogBeginInsert();
/* Prepare WAL record */
STORE_STATE(&bds->spgstate, xlrec.stateSrc);
XLogRegisterData((char *) &xlrec, SizeOfSpgxlogVacuumRoot);
/* sizeof(xlrec) should be a multiple of sizeof(OffsetNumber) */
XLogRegisterData((char *) toDelete,
sizeof(OffsetNumber) * xlrec.nDelete);
XLogRegisterBuffer(0, buffer, REGBUF_STANDARD);
recptr = XLogInsert(RM_SPGIST_ID, XLOG_SPGIST_VACUUM_ROOT);
PageSetLSN(page, recptr);
}
END_CRIT_SECTION();
}
/*
* Helper function to perform deletion of index entries from a bucket.
*
* This function expects that the caller has acquired a cleanup lock on the
* primary bucket page, and will return with a write lock again held on the
* primary bucket page. The lock won't necessarily be held continuously,
* though, because we'll release it when visiting overflow pages.
*
* It would be very bad if this function cleaned a page while some other
* backend was in the midst of scanning it, because hashgettuple assumes
* that the next valid TID will be greater than or equal to the current
* valid TID. There can't be any concurrent scans in progress when we first
* enter this function because of the cleanup lock we hold on the primary
* bucket page, but as soon as we release that lock, there might be. We
* handle that by conspiring to prevent those scans from passing our cleanup
* scan. To do that, we lock the next page in the bucket chain before
* releasing the lock on the previous page. (This type of lock chaining is
* not ideal, so we might want to look for a better solution at some point.)
*
* We need to retain a pin on the primary bucket to ensure that no concurrent
* split can start.
*/
void
hashbucketcleanup(Relation rel, Bucket cur_bucket, Buffer bucket_buf,
BlockNumber bucket_blkno, BufferAccessStrategy bstrategy,
uint32 maxbucket, uint32 highmask, uint32 lowmask,
double *tuples_removed, double *num_index_tuples,
bool split_cleanup,
IndexBulkDeleteCallback callback, void *callback_state)
{
BlockNumber blkno;
Buffer buf;
Bucket new_bucket PG_USED_FOR_ASSERTS_ONLY = InvalidBucket;
bool bucket_dirty = false;
blkno = bucket_blkno;
buf = bucket_buf;
if (split_cleanup)
new_bucket = _hash_get_newbucket_from_oldbucket(rel, cur_bucket,
lowmask, maxbucket);
/* Scan each page in bucket */
for (;;)
{
HashPageOpaque opaque;
OffsetNumber offno;
OffsetNumber maxoffno;
Buffer next_buf;
Page page;
OffsetNumber deletable[MaxOffsetNumber];
int ndeletable = 0;
bool retain_pin = false;
bool clear_dead_marking = false;
vacuum_delay_point();
page = BufferGetPage(buf);
opaque = (HashPageOpaque) PageGetSpecialPointer(page);
/* Scan each tuple in page */
maxoffno = PageGetMaxOffsetNumber(page);
for (offno = FirstOffsetNumber;
offno <= maxoffno;
offno = OffsetNumberNext(offno))
{
ItemPointer htup;
IndexTuple itup;
Bucket bucket;
bool kill_tuple = false;
itup = (IndexTuple) PageGetItem(page,
PageGetItemId(page, offno));
htup = &(itup->t_tid);
/*
* To remove the dead tuples, we strictly want to rely on results
* of callback function. refer btvacuumpage for detailed reason.
*/
if (callback && callback(htup, callback_state))
{
kill_tuple = true;
if (tuples_removed)
*tuples_removed += 1;
}
else if (split_cleanup)
{
/* delete the tuples that are moved by split. */
bucket = _hash_hashkey2bucket(_hash_get_indextuple_hashkey(itup),
maxbucket,
highmask,
lowmask);
/* mark the item for deletion */
if (bucket != cur_bucket)
{
/*
* We expect tuples to either belong to current bucket or
* new_bucket. This is ensured because we don't allow
* further splits from bucket that contains garbage. See
* comments in _hash_expandtable.
*/
Assert(bucket == new_bucket);
kill_tuple = true;
}
}
if (kill_tuple)
{
/* mark the item for deletion */
deletable[ndeletable++] = offno;
}
else
{
/* we're keeping it, so count it */
if (num_index_tuples)
*num_index_tuples += 1;
}
}
/* retain the pin on primary bucket page till end of bucket scan */
if (blkno == bucket_blkno)
retain_pin = true;
else
retain_pin = false;
blkno = opaque->hasho_nextblkno;
/*
* Apply deletions, advance to next page and write page if needed.
*/
if (ndeletable > 0)
{
/* No ereport(ERROR) until changes are logged */
START_CRIT_SECTION();
PageIndexMultiDelete(page, deletable, ndeletable);
bucket_dirty = true;
/*
* Let us mark the page as clean if vacuum removes the DEAD tuples
* from an index page. We do this by clearing
* LH_PAGE_HAS_DEAD_TUPLES flag.
*/
if (tuples_removed && *tuples_removed > 0 &&
H_HAS_DEAD_TUPLES(opaque))
{
opaque->hasho_flag &= ~LH_PAGE_HAS_DEAD_TUPLES;
clear_dead_marking = true;
}
MarkBufferDirty(buf);
/* XLOG stuff */
if (RelationNeedsWAL(rel))
{
xl_hash_delete xlrec;
XLogRecPtr recptr;
xlrec.clear_dead_marking = clear_dead_marking;
xlrec.is_primary_bucket_page = (buf == bucket_buf) ? true : false;
XLogBeginInsert();
XLogRegisterData((char *) &xlrec, SizeOfHashDelete);
/*
* bucket buffer needs to be registered to ensure that we can
* acquire a cleanup lock on it during replay.
*/
if (!xlrec.is_primary_bucket_page)
XLogRegisterBuffer(0, bucket_buf, REGBUF_STANDARD | REGBUF_NO_IMAGE);
XLogRegisterBuffer(1, buf, REGBUF_STANDARD);
XLogRegisterBufData(1, (char *) deletable,
ndeletable * sizeof(OffsetNumber));
recptr = XLogInsert(RM_HASH_ID, XLOG_HASH_DELETE);
PageSetLSN(BufferGetPage(buf), recptr);
}
END_CRIT_SECTION();
}
/* bail out if there are no more pages to scan. */
if (!BlockNumberIsValid(blkno))
break;
next_buf = _hash_getbuf_with_strategy(rel, blkno, HASH_WRITE,
LH_OVERFLOW_PAGE,
bstrategy);
/*
* release the lock on previous page after acquiring the lock on next
* page
*/
if (retain_pin)
LockBuffer(buf, BUFFER_LOCK_UNLOCK);
else
_hash_relbuf(rel, buf);
buf = next_buf;
}
/*
* lock the bucket page to clear the garbage flag and squeeze the bucket.
* if the current buffer is same as bucket buffer, then we already have
* lock on bucket page.
*/
if (buf != bucket_buf)
{
_hash_relbuf(rel, buf);
LockBuffer(bucket_buf, BUFFER_LOCK_EXCLUSIVE);
}
/*
* Clear the garbage flag from bucket after deleting the tuples that are
* moved by split. We purposefully clear the flag before squeeze bucket,
* so that after restart, vacuum shouldn't again try to delete the moved
* by split tuples.
*/
if (split_cleanup)
{
HashPageOpaque bucket_opaque;
Page page;
page = BufferGetPage(bucket_buf);
bucket_opaque = (HashPageOpaque) PageGetSpecialPointer(page);
/* No ereport(ERROR) until changes are logged */
START_CRIT_SECTION();
bucket_opaque->hasho_flag &= ~LH_BUCKET_NEEDS_SPLIT_CLEANUP;
MarkBufferDirty(bucket_buf);
/* XLOG stuff */
if (RelationNeedsWAL(rel))
{
XLogRecPtr recptr;
XLogBeginInsert();
XLogRegisterBuffer(0, bucket_buf, REGBUF_STANDARD);
recptr = XLogInsert(RM_HASH_ID, XLOG_HASH_SPLIT_CLEANUP);
PageSetLSN(page, recptr);
}
END_CRIT_SECTION();
}
/*
* If we have deleted anything, try to compact free space. For squeezing
* the bucket, we must have a cleanup lock, else it can impact the
* ordering of tuples for a scan that has started before it.
*/
if (bucket_dirty && IsBufferCleanupOK(bucket_buf))
_hash_squeezebucket(rel, cur_bucket, bucket_blkno, bucket_buf,
bstrategy);
else
LockBuffer(bucket_buf, BUFFER_LOCK_UNLOCK);
}
/*
* Delete a posting tree page.
*/
static void
ginDeletePage(GinVacuumState *gvs, BlockNumber deleteBlkno, BlockNumber leftBlkno,
BlockNumber parentBlkno, OffsetNumber myoff, bool isParentRoot)
{
Buffer dBuffer;
Buffer lBuffer;
Buffer pBuffer;
Page page,
parentPage;
BlockNumber rightlink;
/*
* Lock the pages in the same order as an insertion would, to avoid
* deadlocks: left, then right, then parent.
*/
lBuffer = ReadBufferExtended(gvs->index, MAIN_FORKNUM, leftBlkno,
RBM_NORMAL, gvs->strategy);
dBuffer = ReadBufferExtended(gvs->index, MAIN_FORKNUM, deleteBlkno,
RBM_NORMAL, gvs->strategy);
pBuffer = ReadBufferExtended(gvs->index, MAIN_FORKNUM, parentBlkno,
RBM_NORMAL, gvs->strategy);
LockBuffer(lBuffer, GIN_EXCLUSIVE);
LockBuffer(dBuffer, GIN_EXCLUSIVE);
if (!isParentRoot) /* parent is already locked by
* LockBufferForCleanup() */
LockBuffer(pBuffer, GIN_EXCLUSIVE);
START_CRIT_SECTION();
/* Unlink the page by changing left sibling's rightlink */
page = BufferGetPage(dBuffer);
rightlink = GinPageGetOpaque(page)->rightlink;
page = BufferGetPage(lBuffer);
GinPageGetOpaque(page)->rightlink = rightlink;
/* Delete downlink from parent */
parentPage = BufferGetPage(pBuffer);
#ifdef USE_ASSERT_CHECKING
do
{
PostingItem *tod = GinDataPageGetPostingItem(parentPage, myoff);
Assert(PostingItemGetBlockNumber(tod) == deleteBlkno);
} while (0);
#endif
GinPageDeletePostingItem(parentPage, myoff);
page = BufferGetPage(dBuffer);
/*
* we shouldn't change rightlink field to save workability of running
* search scan
*/
GinPageGetOpaque(page)->flags = GIN_DELETED;
MarkBufferDirty(pBuffer);
MarkBufferDirty(lBuffer);
MarkBufferDirty(dBuffer);
if (RelationNeedsWAL(gvs->index))
{
XLogRecPtr recptr;
ginxlogDeletePage data;
/*
* We can't pass REGBUF_STANDARD for the deleted page, because we
* didn't set pd_lower on pre-9.4 versions. The page might've been
* binary-upgraded from an older version, and hence not have pd_lower
* set correctly. Ditto for the left page, but removing the item from
* the parent updated its pd_lower, so we know that's OK at this
* point.
*/
XLogBeginInsert();
XLogRegisterBuffer(0, dBuffer, 0);
XLogRegisterBuffer(1, pBuffer, REGBUF_STANDARD);
XLogRegisterBuffer(2, lBuffer, 0);
data.parentOffset = myoff;
data.rightLink = GinPageGetOpaque(page)->rightlink;
XLogRegisterData((char *) &data, sizeof(ginxlogDeletePage));
recptr = XLogInsert(RM_GIN_ID, XLOG_GIN_DELETE_PAGE);
PageSetLSN(page, recptr);
PageSetLSN(parentPage, recptr);
PageSetLSN(BufferGetPage(lBuffer), recptr);
}
if (!isParentRoot)
LockBuffer(pBuffer, GIN_UNLOCK);
ReleaseBuffer(pBuffer);
UnlockReleaseBuffer(lBuffer);
UnlockReleaseBuffer(dBuffer);
END_CRIT_SECTION();
gvs->result->pages_deleted++;
}
/*
* Bulk deletion of all index entries pointing to a set of heap tuples.
* The set of target tuples is specified via a callback routine that tells
* whether any given heap tuple (identified by ItemPointer) is being deleted.
*
* This function also deletes the tuples that are moved by split to other
* bucket.
*
* Result: a palloc'd struct containing statistical info for VACUUM displays.
*/
IndexBulkDeleteResult *
hashbulkdelete(IndexVacuumInfo *info, IndexBulkDeleteResult *stats,
IndexBulkDeleteCallback callback, void *callback_state)
{
Relation rel = info->index;
double tuples_removed;
double num_index_tuples;
double orig_ntuples;
Bucket orig_maxbucket;
Bucket cur_maxbucket;
Bucket cur_bucket;
Buffer metabuf = InvalidBuffer;
HashMetaPage metap;
HashMetaPage cachedmetap;
tuples_removed = 0;
num_index_tuples = 0;
/*
* We need a copy of the metapage so that we can use its hashm_spares[]
* values to compute bucket page addresses, but a cached copy should be
* good enough. (If not, we'll detect that further down and refresh the
* cache as necessary.)
*/
cachedmetap = _hash_getcachedmetap(rel, &metabuf, false);
Assert(cachedmetap != NULL);
orig_maxbucket = cachedmetap->hashm_maxbucket;
orig_ntuples = cachedmetap->hashm_ntuples;
/* Scan the buckets that we know exist */
cur_bucket = 0;
cur_maxbucket = orig_maxbucket;
loop_top:
while (cur_bucket <= cur_maxbucket)
{
BlockNumber bucket_blkno;
BlockNumber blkno;
Buffer bucket_buf;
Buffer buf;
HashPageOpaque bucket_opaque;
Page page;
bool split_cleanup = false;
/* Get address of bucket's start page */
bucket_blkno = BUCKET_TO_BLKNO(cachedmetap, cur_bucket);
blkno = bucket_blkno;
/*
* We need to acquire a cleanup lock on the primary bucket page to out
* wait concurrent scans before deleting the dead tuples.
*/
buf = ReadBufferExtended(rel, MAIN_FORKNUM, blkno, RBM_NORMAL, info->strategy);
LockBufferForCleanup(buf);
_hash_checkpage(rel, buf, LH_BUCKET_PAGE);
page = BufferGetPage(buf);
bucket_opaque = (HashPageOpaque) PageGetSpecialPointer(page);
/*
* If the bucket contains tuples that are moved by split, then we need
* to delete such tuples. We can't delete such tuples if the split
* operation on bucket is not finished as those are needed by scans.
*/
if (!H_BUCKET_BEING_SPLIT(bucket_opaque) &&
H_NEEDS_SPLIT_CLEANUP(bucket_opaque))
{
split_cleanup = true;
/*
* This bucket might have been split since we last held a lock on
* the metapage. If so, hashm_maxbucket, hashm_highmask and
* hashm_lowmask might be old enough to cause us to fail to remove
* tuples left behind by the most recent split. To prevent that,
* now that the primary page of the target bucket has been locked
* (and thus can't be further split), check whether we need to
* update our cached metapage data.
*/
Assert(bucket_opaque->hasho_prevblkno != InvalidBlockNumber);
if (bucket_opaque->hasho_prevblkno > cachedmetap->hashm_maxbucket)
{
cachedmetap = _hash_getcachedmetap(rel, &metabuf, true);
Assert(cachedmetap != NULL);
}
}
bucket_buf = buf;
hashbucketcleanup(rel, cur_bucket, bucket_buf, blkno, info->strategy,
cachedmetap->hashm_maxbucket,
cachedmetap->hashm_highmask,
cachedmetap->hashm_lowmask, &tuples_removed,
&num_index_tuples, split_cleanup,
callback, callback_state);
_hash_dropbuf(rel, bucket_buf);
/* Advance to next bucket */
cur_bucket++;
}
if (BufferIsInvalid(metabuf))
metabuf = _hash_getbuf(rel, HASH_METAPAGE, HASH_NOLOCK, LH_META_PAGE);
/* Write-lock metapage and check for split since we started */
LockBuffer(metabuf, BUFFER_LOCK_EXCLUSIVE);
metap = HashPageGetMeta(BufferGetPage(metabuf));
if (cur_maxbucket != metap->hashm_maxbucket)
{
/* There's been a split, so process the additional bucket(s) */
LockBuffer(metabuf, BUFFER_LOCK_UNLOCK);
cachedmetap = _hash_getcachedmetap(rel, &metabuf, true);
Assert(cachedmetap != NULL);
cur_maxbucket = cachedmetap->hashm_maxbucket;
goto loop_top;
}
/* Okay, we're really done. Update tuple count in metapage. */
START_CRIT_SECTION();
if (orig_maxbucket == metap->hashm_maxbucket &&
orig_ntuples == metap->hashm_ntuples)
{
/*
* No one has split or inserted anything since start of scan, so
* believe our count as gospel.
*/
metap->hashm_ntuples = num_index_tuples;
}
else
{
/*
* Otherwise, our count is untrustworthy since we may have
* double-scanned tuples in split buckets. Proceed by dead-reckoning.
* (Note: we still return estimated_count = false, because using this
* count is better than not updating reltuples at all.)
*/
if (metap->hashm_ntuples > tuples_removed)
metap->hashm_ntuples -= tuples_removed;
else
metap->hashm_ntuples = 0;
num_index_tuples = metap->hashm_ntuples;
}
MarkBufferDirty(metabuf);
/* XLOG stuff */
if (RelationNeedsWAL(rel))
{
xl_hash_update_meta_page xlrec;
XLogRecPtr recptr;
xlrec.ntuples = metap->hashm_ntuples;
XLogBeginInsert();
XLogRegisterData((char *) &xlrec, SizeOfHashUpdateMetaPage);
XLogRegisterBuffer(0, metabuf, REGBUF_STANDARD);
recptr = XLogInsert(RM_HASH_ID, XLOG_HASH_UPDATE_META_PAGE);
PageSetLSN(BufferGetPage(metabuf), recptr);
}
END_CRIT_SECTION();
_hash_relbuf(rel, metabuf);
/* return statistics */
if (stats == NULL)
stats = (IndexBulkDeleteResult *) palloc0(sizeof(IndexBulkDeleteResult));
stats->estimated_count = false;
stats->num_index_tuples = num_index_tuples;
stats->tuples_removed += tuples_removed;
/* hashvacuumcleanup will fill in num_pages */
return stats;
}
/*
* Clean up redirect and placeholder tuples on the given page
*
* Redirect tuples can be marked placeholder once they're old enough.
* Placeholder tuples can be removed if it won't change the offsets of
* non-placeholder ones.
*
* Unlike the routines above, this works on both leaf and inner pages.
*/
static void
vacuumRedirectAndPlaceholder(Relation index, Buffer buffer)
{
Page page = BufferGetPage(buffer);
SpGistPageOpaque opaque = SpGistPageGetOpaque(page);
OffsetNumber i,
max = PageGetMaxOffsetNumber(page),
firstPlaceholder = InvalidOffsetNumber;
bool hasNonPlaceholder = false;
bool hasUpdate = false;
OffsetNumber itemToPlaceholder[MaxIndexTuplesPerPage];
OffsetNumber itemnos[MaxIndexTuplesPerPage];
spgxlogVacuumRedirect xlrec;
xlrec.nToPlaceholder = 0;
xlrec.newestRedirectXid = InvalidTransactionId;
START_CRIT_SECTION();
/*
* Scan backwards to convert old redirection tuples to placeholder tuples,
* and identify location of last non-placeholder tuple while at it.
*/
for (i = max;
i >= FirstOffsetNumber &&
(opaque->nRedirection > 0 || !hasNonPlaceholder);
i--)
{
SpGistDeadTuple dt;
dt = (SpGistDeadTuple) PageGetItem(page, PageGetItemId(page, i));
if (dt->tupstate == SPGIST_REDIRECT &&
TransactionIdPrecedes(dt->xid, RecentGlobalXmin))
{
dt->tupstate = SPGIST_PLACEHOLDER;
Assert(opaque->nRedirection > 0);
opaque->nRedirection--;
opaque->nPlaceholder++;
/* remember newest XID among the removed redirects */
if (!TransactionIdIsValid(xlrec.newestRedirectXid) ||
TransactionIdPrecedes(xlrec.newestRedirectXid, dt->xid))
xlrec.newestRedirectXid = dt->xid;
ItemPointerSetInvalid(&dt->pointer);
itemToPlaceholder[xlrec.nToPlaceholder] = i;
xlrec.nToPlaceholder++;
hasUpdate = true;
}
if (dt->tupstate == SPGIST_PLACEHOLDER)
{
if (!hasNonPlaceholder)
firstPlaceholder = i;
}
else
{
hasNonPlaceholder = true;
}
}
/*
* Any placeholder tuples at the end of page can safely be removed. We
* can't remove ones before the last non-placeholder, though, because we
* can't alter the offset numbers of non-placeholder tuples.
*/
if (firstPlaceholder != InvalidOffsetNumber)
{
/*
* We do not store this array to rdata because it's easy to recreate.
*/
for (i = firstPlaceholder; i <= max; i++)
itemnos[i - firstPlaceholder] = i;
i = max - firstPlaceholder + 1;
Assert(opaque->nPlaceholder >= i);
opaque->nPlaceholder -= i;
/* The array is surely sorted, so can use PageIndexMultiDelete */
PageIndexMultiDelete(page, itemnos, i);
hasUpdate = true;
}
xlrec.firstPlaceholder = firstPlaceholder;
if (hasUpdate)
MarkBufferDirty(buffer);
if (hasUpdate && RelationNeedsWAL(index))
{
XLogRecPtr recptr;
XLogBeginInsert();
XLogRegisterData((char *) &xlrec, SizeOfSpgxlogVacuumRedirect);
XLogRegisterData((char *) itemToPlaceholder,
sizeof(OffsetNumber) * xlrec.nToPlaceholder);
XLogRegisterBuffer(0, buffer, REGBUF_STANDARD);
recptr = XLogInsert(RM_SPGIST_ID, XLOG_SPGIST_VACUUM_REDIRECT);
PageSetLSN(page, recptr);
}
END_CRIT_SECTION();
}
/*
* Write the index tuples contained in *collector into the index's
* pending list.
*
* Function guarantees that all these tuples will be inserted consecutively,
* preserving order
*/
void
ginHeapTupleFastInsert(GinState *ginstate, GinTupleCollector *collector)
{
Relation index = ginstate->index;
Buffer metabuffer;
Page metapage;
GinMetaPageData *metadata = NULL;
Buffer buffer = InvalidBuffer;
Page page = NULL;
ginxlogUpdateMeta data;
bool separateList = false;
bool needCleanup = false;
int cleanupSize;
bool needWal;
if (collector->ntuples == 0)
return;
needWal = RelationNeedsWAL(index);
data.node = index->rd_node;
data.ntuples = 0;
data.newRightlink = data.prevTail = InvalidBlockNumber;
metabuffer = ReadBuffer(index, GIN_METAPAGE_BLKNO);
metapage = BufferGetPage(metabuffer);
if (collector->sumsize + collector->ntuples * sizeof(ItemIdData) > GinListPageSize)
{
/*
* Total size is greater than one page => make sublist
*/
separateList = true;
}
else
{
LockBuffer(metabuffer, GIN_EXCLUSIVE);
metadata = GinPageGetMeta(metapage);
if (metadata->head == InvalidBlockNumber ||
collector->sumsize + collector->ntuples * sizeof(ItemIdData) > metadata->tailFreeSize)
{
/*
* Pending list is empty or total size is greater than freespace
* on tail page => make sublist
*
* We unlock metabuffer to keep high concurrency
*/
separateList = true;
LockBuffer(metabuffer, GIN_UNLOCK);
}
}
if (separateList)
{
/*
* We should make sublist separately and append it to the tail
*/
GinMetaPageData sublist;
memset(&sublist, 0, sizeof(GinMetaPageData));
makeSublist(index, collector->tuples, collector->ntuples, &sublist);
if (needWal)
XLogBeginInsert();
/*
* metapage was unlocked, see above
*/
LockBuffer(metabuffer, GIN_EXCLUSIVE);
metadata = GinPageGetMeta(metapage);
if (metadata->head == InvalidBlockNumber)
{
/*
* Main list is empty, so just insert sublist as main list
*/
START_CRIT_SECTION();
metadata->head = sublist.head;
metadata->tail = sublist.tail;
metadata->tailFreeSize = sublist.tailFreeSize;
metadata->nPendingPages = sublist.nPendingPages;
metadata->nPendingHeapTuples = sublist.nPendingHeapTuples;
}
else
{
/*
* Merge lists
*/
data.prevTail = metadata->tail;
data.newRightlink = sublist.head;
buffer = ReadBuffer(index, metadata->tail);
LockBuffer(buffer, GIN_EXCLUSIVE);
page = BufferGetPage(buffer);
Assert(GinPageGetOpaque(page)->rightlink == InvalidBlockNumber);
START_CRIT_SECTION();
GinPageGetOpaque(page)->rightlink = sublist.head;
MarkBufferDirty(buffer);
metadata->tail = sublist.tail;
metadata->tailFreeSize = sublist.tailFreeSize;
metadata->nPendingPages += sublist.nPendingPages;
metadata->nPendingHeapTuples += sublist.nPendingHeapTuples;
if (needWal)
XLogRegisterBuffer(1, buffer, REGBUF_STANDARD);
}
}
else
{
/*
* Insert into tail page. Metapage is already locked
*/
OffsetNumber l,
off;
int i,
tupsize;
char *ptr;
char *collectordata;
buffer = ReadBuffer(index, metadata->tail);
LockBuffer(buffer, GIN_EXCLUSIVE);
page = BufferGetPage(buffer);
off = (PageIsEmpty(page)) ? FirstOffsetNumber :
OffsetNumberNext(PageGetMaxOffsetNumber(page));
collectordata = ptr = (char *) palloc(collector->sumsize);
data.ntuples = collector->ntuples;
if (needWal)
XLogBeginInsert();
START_CRIT_SECTION();
/*
* Increase counter of heap tuples
*/
Assert(GinPageGetOpaque(page)->maxoff <= metadata->nPendingHeapTuples);
GinPageGetOpaque(page)->maxoff++;
metadata->nPendingHeapTuples++;
for (i = 0; i < collector->ntuples; i++)
{
tupsize = IndexTupleSize(collector->tuples[i]);
l = PageAddItem(page, (Item) collector->tuples[i], tupsize, off, false, false);
if (l == InvalidOffsetNumber)
elog(ERROR, "failed to add item to index page in \"%s\"",
RelationGetRelationName(index));
memcpy(ptr, collector->tuples[i], tupsize);
ptr += tupsize;
off++;
}
Assert((ptr - collectordata) <= collector->sumsize);
if (needWal)
{
XLogRegisterBuffer(1, buffer, REGBUF_STANDARD);
XLogRegisterBufData(1, collectordata, collector->sumsize);
}
metadata->tailFreeSize = PageGetExactFreeSpace(page);
MarkBufferDirty(buffer);
}
/*
* Write metabuffer, make xlog entry
*/
MarkBufferDirty(metabuffer);
if (needWal)
{
XLogRecPtr recptr;
memcpy(&data.metadata, metadata, sizeof(GinMetaPageData));
XLogRegisterBuffer(0, metabuffer, REGBUF_WILL_INIT);
XLogRegisterData((char *) &data, sizeof(ginxlogUpdateMeta));
recptr = XLogInsert(RM_GIN_ID, XLOG_GIN_UPDATE_META_PAGE);
PageSetLSN(metapage, recptr);
if (buffer != InvalidBuffer)
{
PageSetLSN(page, recptr);
}
}
if (buffer != InvalidBuffer)
UnlockReleaseBuffer(buffer);
/*
* Force pending list cleanup when it becomes too long. And,
* ginInsertCleanup could take significant amount of time, so we prefer to
* call it when it can do all the work in a single collection cycle. In
* non-vacuum mode, it shouldn't require maintenance_work_mem, so fire it
* while pending list is still small enough to fit into
* gin_pending_list_limit.
*
* ginInsertCleanup() should not be called inside our CRIT_SECTION.
*/
cleanupSize = GinGetPendingListCleanupSize(index);
if (metadata->nPendingPages * GIN_PAGE_FREESIZE > cleanupSize * 1024L)
needCleanup = true;
UnlockReleaseBuffer(metabuffer);
END_CRIT_SECTION();
if (needCleanup)
ginInsertCleanup(ginstate, false, true, NULL);
}
/*
* Build a pending-list page from the given array of tuples, and write it out.
*
* Returns amount of free space left on the page.
*/
static int32
writeListPage(Relation index, Buffer buffer,
IndexTuple *tuples, int32 ntuples, BlockNumber rightlink)
{
Page page = BufferGetPage(buffer);
int32 i,
freesize,
size = 0;
OffsetNumber l,
off;
char *workspace;
char *ptr;
/* workspace could be a local array; we use palloc for alignment */
workspace = palloc(BLCKSZ);
START_CRIT_SECTION();
GinInitBuffer(buffer, GIN_LIST);
off = FirstOffsetNumber;
ptr = workspace;
for (i = 0; i < ntuples; i++)
{
int this_size = IndexTupleSize(tuples[i]);
memcpy(ptr, tuples[i], this_size);
ptr += this_size;
size += this_size;
l = PageAddItem(page, (Item) tuples[i], this_size, off, false, false);
if (l == InvalidOffsetNumber)
elog(ERROR, "failed to add item to index page in \"%s\"",
RelationGetRelationName(index));
off++;
}
Assert(size <= BLCKSZ); /* else we overran workspace */
GinPageGetOpaque(page)->rightlink = rightlink;
/*
* tail page may contain only whole row(s) or final part of row placed on
* previous pages (a "row" here meaning all the index tuples generated for
* one heap tuple)
*/
if (rightlink == InvalidBlockNumber)
{
GinPageSetFullRow(page);
GinPageGetOpaque(page)->maxoff = 1;
}
else
{
GinPageGetOpaque(page)->maxoff = 0;
}
MarkBufferDirty(buffer);
if (RelationNeedsWAL(index))
{
ginxlogInsertListPage data;
XLogRecPtr recptr;
data.rightlink = rightlink;
data.ntuples = ntuples;
XLogBeginInsert();
XLogRegisterData((char *) &data, sizeof(ginxlogInsertListPage));
XLogRegisterBuffer(0, buffer, REGBUF_WILL_INIT);
XLogRegisterBufData(0, workspace, size);
recptr = XLogInsert(RM_GIN_ID, XLOG_GIN_INSERT_LISTPAGE);
PageSetLSN(page, recptr);
}
/* get free space before releasing buffer */
freesize = PageGetExactFreeSpace(page);
UnlockReleaseBuffer(buffer);
END_CRIT_SECTION();
pfree(workspace);
return freesize;
}
/*
* Deletes pending list pages up to (not including) newHead page.
* If newHead == InvalidBlockNumber then function drops the whole list.
*
* metapage is pinned and exclusive-locked throughout this function.
*
* Returns true if another cleanup process is running concurrently
* (if so, we can just abandon our own efforts)
*/
static bool
shiftList(Relation index, Buffer metabuffer, BlockNumber newHead,
bool fill_fsm, IndexBulkDeleteResult *stats)
{
Page metapage;
GinMetaPageData *metadata;
BlockNumber blknoToDelete;
metapage = BufferGetPage(metabuffer);
metadata = GinPageGetMeta(metapage);
blknoToDelete = metadata->head;
do
{
Page page;
int i;
int64 nDeletedHeapTuples = 0;
ginxlogDeleteListPages data;
Buffer buffers[GIN_NDELETE_AT_ONCE];
BlockNumber freespace[GIN_NDELETE_AT_ONCE];
data.ndeleted = 0;
while (data.ndeleted < GIN_NDELETE_AT_ONCE && blknoToDelete != newHead)
{
freespace[data.ndeleted] = blknoToDelete;
buffers[data.ndeleted] = ReadBuffer(index, blknoToDelete);
LockBuffer(buffers[data.ndeleted], GIN_EXCLUSIVE);
page = BufferGetPage(buffers[data.ndeleted]);
data.ndeleted++;
if (GinPageIsDeleted(page))
{
/* concurrent cleanup process is detected */
for (i = 0; i < data.ndeleted; i++)
UnlockReleaseBuffer(buffers[i]);
return true;
}
nDeletedHeapTuples += GinPageGetOpaque(page)->maxoff;
blknoToDelete = GinPageGetOpaque(page)->rightlink;
}
if (stats)
stats->pages_deleted += data.ndeleted;
/*
* This operation touches an unusually large number of pages, so
* prepare the XLogInsert machinery for that before entering the
* critical section.
*/
if (RelationNeedsWAL(index))
XLogEnsureRecordSpace(data.ndeleted, 0);
START_CRIT_SECTION();
metadata->head = blknoToDelete;
Assert(metadata->nPendingPages >= data.ndeleted);
metadata->nPendingPages -= data.ndeleted;
Assert(metadata->nPendingHeapTuples >= nDeletedHeapTuples);
metadata->nPendingHeapTuples -= nDeletedHeapTuples;
if (blknoToDelete == InvalidBlockNumber)
{
metadata->tail = InvalidBlockNumber;
metadata->tailFreeSize = 0;
metadata->nPendingPages = 0;
metadata->nPendingHeapTuples = 0;
}
MarkBufferDirty(metabuffer);
for (i = 0; i < data.ndeleted; i++)
{
page = BufferGetPage(buffers[i]);
GinPageGetOpaque(page)->flags = GIN_DELETED;
MarkBufferDirty(buffers[i]);
}
if (RelationNeedsWAL(index))
{
XLogRecPtr recptr;
XLogBeginInsert();
XLogRegisterBuffer(0, metabuffer, REGBUF_WILL_INIT);
for (i = 0; i < data.ndeleted; i++)
XLogRegisterBuffer(i + 1, buffers[i], REGBUF_WILL_INIT);
memcpy(&data.metadata, metadata, sizeof(GinMetaPageData));
XLogRegisterData((char *) &data,
sizeof(ginxlogDeleteListPages));
recptr = XLogInsert(RM_GIN_ID, XLOG_GIN_DELETE_LISTPAGE);
PageSetLSN(metapage, recptr);
for (i = 0; i < data.ndeleted; i++)
{
page = BufferGetPage(buffers[i]);
PageSetLSN(page, recptr);
}
}
for (i = 0; i < data.ndeleted; i++)
UnlockReleaseBuffer(buffers[i]);
END_CRIT_SECTION();
for (i = 0; fill_fsm && i < data.ndeleted; i++)
RecordFreeIndexPage(index, freespace[i]);
} while (blknoToDelete != newHead);
return false;
}
/*
* Write a backup block if needed when we are setting a hint. Note that
* this may be called for a variety of page types, not just heaps.
*
* Callable while holding just share lock on the buffer content.
*
* We can't use the plain backup block mechanism since that relies on the
* Buffer being exclusively locked. Since some modifications (setting LSN, hint
* bits) are allowed in a sharelocked buffer that can lead to wal checksum
* failures. So instead we copy the page and insert the copied data as normal
* record data.
*
* We only need to do something if page has not yet been full page written in
* this checkpoint round. The LSN of the inserted wal record is returned if we
* had to write, InvalidXLogRecPtr otherwise.
*
* It is possible that multiple concurrent backends could attempt to write WAL
* records. In that case, multiple copies of the same block would be recorded
* in separate WAL records by different backends, though that is still OK from
* a correctness perspective.
*/
XLogRecPtr
XLogSaveBufferForHint(Buffer buffer, bool buffer_std)
{
XLogRecPtr recptr = InvalidXLogRecPtr;
XLogRecPtr lsn;
XLogRecPtr RedoRecPtr;
/*
* Ensure no checkpoint can change our view of RedoRecPtr.
*/
Assert(MyPgXact->delayChkpt);
/*
* Update RedoRecPtr so that we can make the right decision
*/
RedoRecPtr = GetRedoRecPtr();
/*
* We assume page LSN is first data on *every* page that can be passed to
* XLogInsert, whether it has the standard page layout or not. Since we're
* only holding a share-lock on the page, we must take the buffer header
* lock when we look at the LSN.
*/
lsn = BufferGetLSNAtomic(buffer);
if (lsn <= RedoRecPtr)
{
int flags;
char copied_buffer[BLCKSZ];
char *origdata = (char *) BufferGetBlock(buffer);
RelFileNode rnode;
ForkNumber forkno;
BlockNumber blkno;
/*
* Copy buffer so we don't have to worry about concurrent hint bit or
* lsn updates. We assume pd_lower/upper cannot be changed without an
* exclusive lock, so the contents bkp are not racy.
*/
if (buffer_std)
{
/* Assume we can omit data between pd_lower and pd_upper */
Page page = BufferGetPage(buffer);
uint16 lower = ((PageHeader) page)->pd_lower;
uint16 upper = ((PageHeader) page)->pd_upper;
memcpy(copied_buffer, origdata, lower);
memcpy(copied_buffer + upper, origdata + upper, BLCKSZ - upper);
}
else
memcpy(copied_buffer, origdata, BLCKSZ);
XLogBeginInsert();
flags = REGBUF_FORCE_IMAGE;
if (buffer_std)
flags |= REGBUF_STANDARD;
BufferGetTag(buffer, &rnode, &forkno, &blkno);
XLogRegisterBlock(0, &rnode, forkno, blkno, copied_buffer, flags);
recptr = XLogInsert(RM_XLOG_ID, XLOG_FPI_FOR_HINT);
}
return recptr;
}
/*
* Drop a table space
*
* Be careful to check that the tablespace is empty.
*/
void
DropTableSpace(DropTableSpaceStmt *stmt)
{
#ifdef HAVE_SYMLINK
char *tablespacename = stmt->tablespacename;
HeapScanDesc scandesc;
Relation rel;
HeapTuple tuple;
ScanKeyData entry[1];
Oid tablespaceoid;
/*
* Find the target tuple
*/
rel = heap_open(TableSpaceRelationId, RowExclusiveLock);
ScanKeyInit(&entry[0],
Anum_pg_tablespace_spcname,
BTEqualStrategyNumber, F_NAMEEQ,
CStringGetDatum(tablespacename));
scandesc = heap_beginscan_catalog(rel, 1, entry);
tuple = heap_getnext(scandesc, ForwardScanDirection);
if (!HeapTupleIsValid(tuple))
{
if (!stmt->missing_ok)
{
ereport(ERROR,
(errcode(ERRCODE_UNDEFINED_OBJECT),
errmsg("tablespace \"%s\" does not exist",
tablespacename)));
}
else
{
ereport(NOTICE,
(errmsg("tablespace \"%s\" does not exist, skipping",
tablespacename)));
/* XXX I assume I need one or both of these next two calls */
heap_endscan(scandesc);
heap_close(rel, NoLock);
}
return;
}
tablespaceoid = HeapTupleGetOid(tuple);
/* Must be tablespace owner */
if (!pg_tablespace_ownercheck(tablespaceoid, GetUserId()))
aclcheck_error(ACLCHECK_NOT_OWNER, ACL_KIND_TABLESPACE,
tablespacename);
/* Disallow drop of the standard tablespaces, even by superuser */
if (tablespaceoid == GLOBALTABLESPACE_OID ||
tablespaceoid == DEFAULTTABLESPACE_OID)
aclcheck_error(ACLCHECK_NO_PRIV, ACL_KIND_TABLESPACE,
tablespacename);
/* DROP hook for the tablespace being removed */
InvokeObjectDropHook(TableSpaceRelationId, tablespaceoid, 0);
/*
* Remove the pg_tablespace tuple (this will roll back if we fail below)
*/
simple_heap_delete(rel, &tuple->t_self);
heap_endscan(scandesc);
/*
* Remove any comments or security labels on this tablespace.
*/
DeleteSharedComments(tablespaceoid, TableSpaceRelationId);
DeleteSharedSecurityLabel(tablespaceoid, TableSpaceRelationId);
/*
* Remove dependency on owner.
*/
deleteSharedDependencyRecordsFor(TableSpaceRelationId, tablespaceoid, 0);
/*
* Acquire TablespaceCreateLock to ensure that no TablespaceCreateDbspace
* is running concurrently.
*/
LWLockAcquire(TablespaceCreateLock, LW_EXCLUSIVE);
/*
* Try to remove the physical infrastructure.
*/
if (!destroy_tablespace_directories(tablespaceoid, false))
{
/*
* Not all files deleted? However, there can be lingering empty files
* in the directories, left behind by for example DROP TABLE, that
* have been scheduled for deletion at next checkpoint (see comments
* in mdunlink() for details). We could just delete them immediately,
* but we can't tell them apart from important data files that we
* mustn't delete. So instead, we force a checkpoint which will clean
* out any lingering files, and try again.
*
* XXX On Windows, an unlinked file persists in the directory listing
* until no process retains an open handle for the file. The DDL
* commands that schedule files for unlink send invalidation messages
* directing other PostgreSQL processes to close the files. DROP
* TABLESPACE should not give up on the tablespace becoming empty
* until all relevant invalidation processing is complete.
*/
RequestCheckpoint(CHECKPOINT_IMMEDIATE | CHECKPOINT_FORCE | CHECKPOINT_WAIT);
if (!destroy_tablespace_directories(tablespaceoid, false))
{
/* Still not empty, the files must be important then */
ereport(ERROR,
(errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
errmsg("tablespace \"%s\" is not empty",
tablespacename)));
}
}
/* Record the filesystem change in XLOG */
{
xl_tblspc_drop_rec xlrec;
xlrec.ts_id = tablespaceoid;
XLogBeginInsert();
XLogRegisterData((char *) &xlrec, sizeof(xl_tblspc_drop_rec));
(void) XLogInsert(RM_TBLSPC_ID, XLOG_TBLSPC_DROP);
}
/*
* Note: because we checked that the tablespace was empty, there should be
* no need to worry about flushing shared buffers or free space map
* entries for relations in the tablespace.
*/
/*
* Force synchronous commit, to minimize the window between removing the
* files on-disk and marking the transaction committed. It's not great
* that there is any window at all, but definitely we don't want to make
* it larger than necessary.
*/
ForceSyncCommit();
/*
* Allow TablespaceCreateDbspace again.
*/
LWLockRelease(TablespaceCreateLock);
/* We keep the lock on pg_tablespace until commit */
heap_close(rel, NoLock);
#else /* !HAVE_SYMLINK */
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("tablespaces are not supported on this platform")));
#endif /* HAVE_SYMLINK */
}
IndexBuildResult *
ginbuild(Relation heap, Relation index, IndexInfo *indexInfo)
{
IndexBuildResult *result;
double reltuples;
GinBuildState buildstate;
Buffer RootBuffer,
MetaBuffer;
ItemPointerData *list;
Datum key;
GinNullCategory category;
uint32 nlist;
MemoryContext oldCtx;
OffsetNumber attnum;
if (RelationGetNumberOfBlocks(index) != 0)
elog(ERROR, "index \"%s\" already contains data",
RelationGetRelationName(index));
initGinState(&buildstate.ginstate, index);
buildstate.indtuples = 0;
memset(&buildstate.buildStats, 0, sizeof(GinStatsData));
/* 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 (RelationNeedsWAL(index))
{
XLogRecPtr recptr;
Page page;
XLogBeginInsert();
XLogRegisterBuffer(0, MetaBuffer, REGBUF_WILL_INIT | REGBUF_STANDARD);
XLogRegisterBuffer(1, RootBuffer, REGBUF_WILL_INIT);
recptr = XLogInsert(RM_GIN_ID, XLOG_GIN_CREATE_INDEX);
page = BufferGetPage(RootBuffer);
PageSetLSN(page, recptr);
page = BufferGetPage(MetaBuffer);
PageSetLSN(page, recptr);
}
UnlockReleaseBuffer(MetaBuffer);
UnlockReleaseBuffer(RootBuffer);
END_CRIT_SECTION();
/* count the root as first entry page */
buildstate.buildStats.nEntryPages++;
/*
* create a temporary memory context that is used to hold data not yet
* dumped out to the index
*/
buildstate.tmpCtx = AllocSetContextCreate(CurrentMemoryContext,
"Gin build temporary context",
ALLOCSET_DEFAULT_SIZES);
/*
* create a temporary memory context that is used for calling
* ginExtractEntries(), and can be reset after each tuple
*/
buildstate.funcCtx = AllocSetContextCreate(CurrentMemoryContext,
"Gin build temporary context for user-defined function",
ALLOCSET_DEFAULT_SIZES);
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, NULL);
/* dump remaining entries to the index */
oldCtx = MemoryContextSwitchTo(buildstate.tmpCtx);
ginBeginBAScan(&buildstate.accum);
while ((list = ginGetBAEntry(&buildstate.accum,
&attnum, &key, &category, &nlist)) != NULL)
{
/* there could be many entries, so be willing to abort here */
CHECK_FOR_INTERRUPTS();
ginEntryInsert(&buildstate.ginstate, attnum, key, category,
list, nlist, &buildstate.buildStats);
}
MemoryContextSwitchTo(oldCtx);
MemoryContextDelete(buildstate.funcCtx);
MemoryContextDelete(buildstate.tmpCtx);
/*
* Update metapage stats
*/
buildstate.buildStats.nTotalPages = RelationGetNumberOfBlocks(index);
ginUpdateStats(index, &buildstate.buildStats);
/*
* Return statistics
*/
result = (IndexBuildResult *) palloc(sizeof(IndexBuildResult));
result->heap_tuples = reltuples;
result->index_tuples = buildstate.indtuples;
return result;
}
/*
* Main entry point to GiST index build. Initially calls insert over and over,
* but switches to more efficient buffering build algorithm after a certain
* number of tuples (unless buffering mode is disabled).
*/
IndexBuildResult *
gistbuild(Relation heap, Relation index, IndexInfo *indexInfo)
{
IndexBuildResult *result;
double reltuples;
GISTBuildState buildstate;
Buffer buffer;
Page page;
MemoryContext oldcxt = CurrentMemoryContext;
int fillfactor;
buildstate.indexrel = index;
if (index->rd_options)
{
/* Get buffering mode from the options string */
GiSTOptions *options = (GiSTOptions *) index->rd_options;
char *bufferingMode = (char *) options + options->bufferingModeOffset;
if (strcmp(bufferingMode, "on") == 0)
buildstate.bufferingMode = GIST_BUFFERING_STATS;
else if (strcmp(bufferingMode, "off") == 0)
buildstate.bufferingMode = GIST_BUFFERING_DISABLED;
else
buildstate.bufferingMode = GIST_BUFFERING_AUTO;
fillfactor = options->fillfactor;
}
else
{
/*
* By default, switch to buffering mode when the index grows too large
* to fit in cache.
*/
buildstate.bufferingMode = GIST_BUFFERING_AUTO;
fillfactor = GIST_DEFAULT_FILLFACTOR;
}
/* Calculate target amount of free space to leave on pages */
buildstate.freespace = BLCKSZ * (100 - fillfactor) / 100;
/*
* We expect to be called exactly once for any index relation. If that's
* not the case, big trouble's what we have.
*/
if (RelationGetNumberOfBlocks(index) != 0)
elog(ERROR, "index \"%s\" already contains data",
RelationGetRelationName(index));
/* no locking is needed */
buildstate.giststate = initGISTstate(index);
/*
* Create a temporary memory context that is reset once for each tuple
* processed. (Note: we don't bother to make this a child of the
* giststate's scanCxt, so we have to delete it separately at the end.)
*/
buildstate.giststate->tempCxt = createTempGistContext();
/* initialize the root page */
buffer = gistNewBuffer(index);
Assert(BufferGetBlockNumber(buffer) == GIST_ROOT_BLKNO);
page = BufferGetPage(buffer);
START_CRIT_SECTION();
GISTInitBuffer(buffer, F_LEAF);
MarkBufferDirty(buffer);
if (RelationNeedsWAL(index))
{
XLogRecPtr recptr;
XLogBeginInsert();
XLogRegisterBuffer(0, buffer, REGBUF_WILL_INIT);
recptr = XLogInsert(RM_GIST_ID, XLOG_GIST_CREATE_INDEX);
PageSetLSN(page, recptr);
}
else
PageSetLSN(page, gistGetFakeLSN(heap));
UnlockReleaseBuffer(buffer);
END_CRIT_SECTION();
/* build the index */
buildstate.indtuples = 0;
buildstate.indtuplesSize = 0;
/*
* Do the heap scan.
*/
reltuples = IndexBuildHeapScan(heap, index, indexInfo, true,
gistBuildCallback, (void *) &buildstate);
/*
* If buffering was used, flush out all the tuples that are still in the
* buffers.
*/
if (buildstate.bufferingMode == GIST_BUFFERING_ACTIVE)
{
elog(DEBUG1, "all tuples processed, emptying buffers");
gistEmptyAllBuffers(&buildstate);
gistFreeBuildBuffers(buildstate.gfbb);
}
/* okay, all heap tuples are indexed */
MemoryContextSwitchTo(oldcxt);
MemoryContextDelete(buildstate.giststate->tempCxt);
freeGISTstate(buildstate.giststate);
/*
* Return statistics
*/
result = (IndexBuildResult *) palloc(sizeof(IndexBuildResult));
result->heap_tuples = reltuples;
result->index_tuples = (double) buildstate.indtuples;
return result;
}
