/*
* XLogReadBufferExtended
* Read a page during XLOG replay
*
* This is functionally comparable to ReadBufferExtended. There's some
* differences in the behavior wrt. the "mode" argument:
*
* In RBM_NORMAL mode, if the page doesn't exist, or contains all-zeroes, we
* return InvalidBuffer. In this case the caller should silently skip the
* update on this page. (In this situation, we expect that the page was later
* dropped or truncated. If we don't see evidence of that later in the WAL
* sequence, we'll complain at the end of WAL replay.)
*
* In RBM_ZERO and RBM_ZERO_ON_ERROR modes, if the page doesn't exist, the
* relation is extended with all-zeroes pages up to the given block number.
*
* In RBM_NORMAL_NO_LOG mode, we return InvalidBuffer if the page doesn't
* exist, and we don't check for all-zeroes. Thus, no log entry is made
* to imply that the page should be dropped or truncated later.
*/
Buffer
XLogReadBufferExtended(RelFileNode rnode, ForkNumber forknum,
BlockNumber blkno, ReadBufferMode mode)
{
BlockNumber lastblock;
Buffer buffer;
SMgrRelation smgr;
Assert(blkno != P_NEW);
/* Open the relation at smgr level */
smgr = smgropen(rnode, InvalidBackendId);
/*
* Create the target file if it doesn't already exist. This lets us cope
* if the replay sequence contains writes to a relation that is later
* deleted. (The original coding of this routine would instead suppress
* the writes, but that seems like it risks losing valuable data if the
* filesystem loses an inode during a crash. Better to write the data
* until we are actually told to delete the file.)
*/
smgrcreate(smgr, forknum, true);
lastblock = smgrnblocks(smgr, forknum);
if (blkno < lastblock)
{
/* page exists in file */
buffer = ReadBufferWithoutRelcache(rnode, forknum, blkno,
mode, NULL);
}
else
{
/* hm, page doesn't exist in file */
if (mode == RBM_NORMAL)
{
log_invalid_page(rnode, forknum, blkno, false);
return InvalidBuffer;
}
if (mode == RBM_NORMAL_NO_LOG)
return InvalidBuffer;
/* OK to extend the file */
/* we do this in recovery only - no rel-extension lock needed */
Assert(InRecovery);
buffer = InvalidBuffer;
do
{
if (buffer != InvalidBuffer)
ReleaseBuffer(buffer);
buffer = ReadBufferWithoutRelcache(rnode, forknum,
P_NEW, mode, NULL);
}
while (BufferGetBlockNumber(buffer) < blkno);
/* Handle the corner case that P_NEW returns non-consecutive pages */
if (BufferGetBlockNumber(buffer) != blkno)
{
ReleaseBuffer(buffer);
buffer = ReadBufferWithoutRelcache(rnode, forknum, blkno,
mode, NULL);
}
}
if (mode == RBM_NORMAL)
{
/* check that page has been initialized */
Page page = (Page) BufferGetPage(buffer);
/*
* We assume that PageIsNew is safe without a lock. During recovery,
* there should be no other backends that could modify the buffer at
* the same time.
*/
if (PageIsNew(page))
{
ReleaseBuffer(buffer);
log_invalid_page(rnode, forknum, blkno, true);
return InvalidBuffer;
}
}
return buffer;
}
/* ------------------------------------------------
* bt_metap()
*
* Get a btree meta-page information
*
* Usage: SELECT * FROM bt_metap('t1_pkey')
* ------------------------------------------------
*/
Datum
bt_metap(PG_FUNCTION_ARGS)
{
text *relname = PG_GETARG_TEXT_P(0);
Buffer buffer;
Relation rel;
RangeVar *relrv;
Datum result;
if (!superuser())
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
(errmsg("must be superuser to use pgstattuple functions"))));
relrv = makeRangeVarFromNameList(textToQualifiedNameList(relname));
rel = relation_openrv(relrv, AccessShareLock);
if (!IS_INDEX(rel) || !IS_BTREE(rel))
elog(ERROR, "bt_metap() can be used only on b-tree index.");
/*
* Reject attempts to read non-local temporary relations; we would
* be likely to get wrong data since we have no visibility into the
* owning session's local buffers.
*/
if (isOtherTempNamespace(RelationGetNamespace(rel)))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("cannot access temporary tables of other sessions")));
buffer = ReadBuffer(rel, 0);
{
BTMetaPageData *metad;
TupleDesc tupleDesc;
int j;
char *values[BTMETAP_NCOLUMNS];
HeapTuple tuple;
Page page = BufferGetPage(buffer);
metad = BTPageGetMeta(page);
tupleDesc = RelationNameGetTupleDesc(BTMETAP_TYPE);
j = 0;
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", metad->btm_magic);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", metad->btm_version);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", metad->btm_root);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", metad->btm_level);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", metad->btm_fastroot);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", metad->btm_fastlevel);
tuple = BuildTupleFromCStrings(TupleDescGetAttInMetadata(tupleDesc),
values);
result = TupleGetDatum(TupleDescGetSlot(tupleDesc), tuple);
}
ReleaseBuffer(buffer);
relation_close(rel, AccessShareLock);
PG_RETURN_DATUM(result);
}
/*
* Get a buffer of the type and parity specified by flags, having at least
* as much free space as indicated by needSpace. We use the lastUsedPages
* cache to assign the same buffer previously requested when possible.
* The returned buffer is already pinned and exclusive-locked.
*
* *isNew is set true if the page was initialized here, false if it was
* already valid.
*/
Buffer
SpGistGetBuffer(Relation index, int flags, int needSpace, bool *isNew)
{
SpGistCache *cache = spgGetCache(index);
SpGistLastUsedPage *lup;
/* Bail out if even an empty page wouldn't meet the demand */
if (needSpace > SPGIST_PAGE_CAPACITY)
elog(ERROR, "desired SPGiST tuple size is too big");
/*
* If possible, increase the space request to include relation's
* fillfactor. This ensures that when we add unrelated tuples to a page,
* we try to keep 100-fillfactor% available for adding tuples that are
* related to the ones already on it. But fillfactor mustn't cause an
* error for requests that would otherwise be legal.
*/
needSpace += RelationGetTargetPageFreeSpace(index,
SPGIST_DEFAULT_FILLFACTOR);
needSpace = Min(needSpace, SPGIST_PAGE_CAPACITY);
/* Get the cache entry for this flags setting */
lup = GET_LUP(cache, flags);
/* If we have nothing cached, just turn it over to allocNewBuffer */
if (lup->blkno == InvalidBlockNumber)
{
*isNew = true;
return allocNewBuffer(index, flags);
}
/* fixed pages should never be in cache */
Assert(!SpGistBlockIsFixed(lup->blkno));
/* If cached freeSpace isn't enough, don't bother looking at the page */
if (lup->freeSpace >= needSpace)
{
Buffer buffer;
Page page;
buffer = ReadBuffer(index, lup->blkno);
if (!ConditionalLockBuffer(buffer))
{
/*
* buffer is locked by another process, so return a new buffer
*/
ReleaseBuffer(buffer);
*isNew = true;
return allocNewBuffer(index, flags);
}
page = BufferGetPage(buffer);
if (PageIsNew(page) || SpGistPageIsDeleted(page) || PageIsEmpty(page))
{
/* OK to initialize the page */
uint16 pageflags = 0;
if (GBUF_REQ_LEAF(flags))
pageflags |= SPGIST_LEAF;
if (GBUF_REQ_NULLS(flags))
pageflags |= SPGIST_NULLS;
SpGistInitBuffer(buffer, pageflags);
lup->freeSpace = PageGetExactFreeSpace(page) - needSpace;
*isNew = true;
return buffer;
}
/*
* Check that page is of right type and has enough space. We must
* recheck this since our cache isn't necessarily up to date.
*/
if ((GBUF_REQ_LEAF(flags) ? SpGistPageIsLeaf(page) : !SpGistPageIsLeaf(page)) &&
(GBUF_REQ_NULLS(flags) ? SpGistPageStoresNulls(page) : !SpGistPageStoresNulls(page)))
{
int freeSpace = PageGetExactFreeSpace(page);
if (freeSpace >= needSpace)
{
/* Success, update freespace info and return the buffer */
lup->freeSpace = freeSpace - needSpace;
*isNew = false;
return buffer;
}
}
/*
* fallback to allocation of new buffer
*/
UnlockReleaseBuffer(buffer);
}
/* No success with cache, so return a new buffer */
*isNew = true;
return allocNewBuffer(index, flags);
}
static void PersistentStore_DiagnoseDumpTable(
PersistentStoreData *storeData,
PersistentStoreSharedData *storeSharedData)
{
if (disable_persistent_diagnostic_dump)
{
return;
}
MIRROREDLOCK_BUFMGR_DECLARE;
PersistentStoreScan storeScan;
ItemPointerData persistentTid;
int64 persistentSerialNum;
Datum *values;
BlockNumber lastDisplayedBlockNum;
bool displayedOne;
BlockNumber currentBlockNum;
elog(LOG,
"Diagnostic dump of persistent table ('%s'): maximum in-use serial number " INT64_FORMAT ", maximum free order number " INT64_FORMAT ", free TID %s, maximum known TID %s",
storeData->tableName,
storeSharedData->maxInUseSerialNum,
storeSharedData->maxFreeOrderNum,
ItemPointerToString(&storeSharedData->freeTid),
ItemPointerToString2(&storeSharedData->maxTid));
values = (Datum*)palloc(storeData->numAttributes * sizeof(Datum));
PersistentStore_BeginScan(
storeData,
storeSharedData,
&storeScan);
lastDisplayedBlockNum = 0;
displayedOne = false;
while (PersistentStore_GetNext(
&storeScan,
values,
&persistentTid,
&persistentSerialNum))
{
/*
* Use the BlockIdGetBlockNumber routine because ItemPointerGetBlockNumber
* asserts for valid TID.
*/
currentBlockNum = BlockIdGetBlockNumber(&persistentTid.ip_blkid);
if (!displayedOne || currentBlockNum != lastDisplayedBlockNum)
{
Buffer buffer;
PageHeader page;
XLogRecPtr lsn;
/*
* Fetch the block and display the LSN.
*/
// -------- MirroredLock ----------
MIRROREDLOCK_BUFMGR_LOCK;
buffer = ReadBuffer(
storeScan.persistentRel,
currentBlockNum);
page = (PageHeader) BufferGetPage(buffer);
lsn = PageGetLSN(page);
ReleaseBuffer(buffer);
MIRROREDLOCK_BUFMGR_UNLOCK;
// -------- MirroredLock ----------
elog(LOG, "Diagnostic LSN %s of page %u",
XLogLocationToString(&lsn),
currentBlockNum);
lastDisplayedBlockNum = currentBlockNum;
displayedOne = true;
}
/*
* Display the persistent tuple.
*/
(*storeData->printTupleCallback)(
LOG,
"DIAGNOSE",
&persistentTid,
values);
}
PersistentStore_EndScan(&storeScan);
pfree(values);
}
/* -----------------------------------------------
* bt_page()
*
* Usage: SELECT * FROM bt_page('t1_pkey', 0);
* -----------------------------------------------
*/
Datum
bt_page_stats(PG_FUNCTION_ARGS)
{
text *relname = PG_GETARG_TEXT_P(0);
uint32 blkno = PG_GETARG_UINT32(1);
Buffer buffer;
Relation rel;
RangeVar *relrv;
Datum result;
if (!superuser())
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
(errmsg("must be superuser to use pgstattuple functions"))));
relrv = makeRangeVarFromNameList(textToQualifiedNameList(relname));
rel = relation_openrv(relrv, AccessShareLock);
if (!IS_INDEX(rel) || !IS_BTREE(rel))
elog(ERROR, "bt_page_stats() can be used only on b-tree index.");
/*
* Reject attempts to read non-local temporary relations; we would
* be likely to get wrong data since we have no visibility into the
* owning session's local buffers.
*/
if (isOtherTempNamespace(RelationGetNamespace(rel)))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("cannot access temporary tables of other sessions")));
if (blkno == 0)
elog(ERROR, "Block 0 is a meta page.");
CHECK_RELATION_BLOCK_RANGE(rel, blkno);
buffer = ReadBuffer(rel, blkno);
{
HeapTuple tuple;
TupleDesc tupleDesc;
int j;
char *values[BTPAGESTATS_NCOLUMNS];
BTPageStat stat;
GetBTPageStatistics(blkno, buffer, &stat);
tupleDesc = RelationNameGetTupleDesc(BTPAGESTATS_TYPE);
j = 0;
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", stat.blkno);
values[j] = palloc(32);
snprintf(values[j++], 32, "%c", stat.type);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", stat.live_items);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", stat.dead_items);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", stat.avg_item_size);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", stat.page_size);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", stat.free_size);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", stat.btpo_prev);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", stat.btpo_next);
values[j] = palloc(32);
if (stat.type == 'd')
snprintf(values[j++], 32, "%d", stat.btpo.xact);
else
snprintf(values[j++], 32, "%d", stat.btpo.level);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", stat.btpo_flags);
tuple = BuildTupleFromCStrings(TupleDescGetAttInMetadata(tupleDesc),
values);
result = TupleGetDatum(TupleDescGetSlot(tupleDesc), tuple);
}
ReleaseBuffer(buffer);
relation_close(rel, AccessShareLock);
PG_RETURN_DATUM(result);
}
/*
* RelationGetBufferForTuple
*
* Returns pinned and exclusive-locked buffer of a page in given relation
* with free space >= given len.
*
* If otherBuffer is not InvalidBuffer, then it references a previously
* pinned buffer of another page in the same relation; on return, this
* buffer will also be exclusive-locked. (This case is used by heap_update;
* the otherBuffer contains the tuple being updated.)
*
* The reason for passing otherBuffer is that if two backends are doing
* concurrent heap_update operations, a deadlock could occur if they try
* to lock the same two buffers in opposite orders. To ensure that this
* can't happen, we impose the rule that buffers of a relation must be
* locked in increasing page number order. This is most conveniently done
* by having RelationGetBufferForTuple lock them both, with suitable care
* for ordering.
*
* NOTE: it is unlikely, but not quite impossible, for otherBuffer to be the
* same buffer we select for insertion of the new tuple (this could only
* happen if space is freed in that page after heap_update finds there's not
* enough there). In that case, the page will be pinned and locked only once.
*
* For the vmbuffer and vmbuffer_other arguments, we avoid deadlock by
* locking them only after locking the corresponding heap page, and taking
* no further lwlocks while they are locked.
*
* We normally use FSM to help us find free space. However,
* if HEAP_INSERT_SKIP_FSM is specified, we just append a new empty page to
* the end of the relation if the tuple won't fit on the current target page.
* This can save some cycles when we know the relation is new and doesn't
* contain useful amounts of free space.
*
* HEAP_INSERT_SKIP_FSM is also useful for non-WAL-logged additions to a
* relation, if the caller holds exclusive lock and is careful to invalidate
* relation's smgr_targblock before the first insertion --- that ensures that
* all insertions will occur into newly added pages and not be intermixed
* with tuples from other transactions. That way, a crash can't risk losing
* any committed data of other transactions. (See heap_insert's comments
* for additional constraints needed for safe usage of this behavior.)
*
* The caller can also provide a BulkInsertState object to optimize many
* insertions into the same relation. This keeps a pin on the current
* insertion target page (to save pin/unpin cycles) and also passes a
* BULKWRITE buffer selection strategy object to the buffer manager.
* Passing NULL for bistate selects the default behavior.
*
* We always try to avoid filling existing pages further than the fillfactor.
* This is OK since this routine is not consulted when updating a tuple and
* keeping it on the same page, which is the scenario fillfactor is meant
* to reserve space for.
*
* ereport(ERROR) is allowed here, so this routine *must* be called
* before any (unlogged) changes are made in buffer pool.
*/
Buffer
RelationGetBufferForTuple(Relation relation, Size len,
Buffer otherBuffer, int options,
BulkInsertState bistate,
Buffer *vmbuffer, Buffer *vmbuffer_other)
{
bool use_fsm = !(options & HEAP_INSERT_SKIP_FSM);
Buffer buffer = InvalidBuffer;
Page page;
Size pageFreeSpace,
saveFreeSpace;
BlockNumber targetBlock,
otherBlock;
bool needLock;
len = MAXALIGN(len); /* be conservative */
/* Bulk insert is not supported for updates, only inserts. */
Assert(otherBuffer == InvalidBuffer || !bistate);
/*
* If we're gonna fail for oversize tuple, do it right away
*/
if (len > MaxHeapTupleSize)
ereport(ERROR,
(errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
errmsg("row is too big: size %lu, maximum size %lu",
(unsigned long) len,
(unsigned long) MaxHeapTupleSize)));
/* Compute desired extra freespace due to fillfactor option */
saveFreeSpace = RelationGetTargetPageFreeSpace(relation,
HEAP_DEFAULT_FILLFACTOR);
if (otherBuffer != InvalidBuffer)
otherBlock = BufferGetBlockNumber(otherBuffer);
else
otherBlock = InvalidBlockNumber; /* just to keep compiler quiet */
/*
* We first try to put the tuple on the same page we last inserted a tuple
* on, as cached in the BulkInsertState or relcache entry. If that
* doesn't work, we ask the Free Space Map to locate a suitable page.
* Since the FSM's info might be out of date, we have to be prepared to
* loop around and retry multiple times. (To insure this isn't an infinite
* loop, we must update the FSM with the correct amount of free space on
* each page that proves not to be suitable.) If the FSM has no record of
* a page with enough free space, we give up and extend the relation.
*
* When use_fsm is false, we either put the tuple onto the existing target
* page or extend the relation.
*/
if (len + saveFreeSpace > MaxHeapTupleSize)
{
/* can't fit, don't bother asking FSM */
targetBlock = InvalidBlockNumber;
use_fsm = false;
}
else if (bistate && bistate->current_buf != InvalidBuffer)
targetBlock = BufferGetBlockNumber(bistate->current_buf);
else
targetBlock = RelationGetTargetBlock(relation);
if (targetBlock == InvalidBlockNumber && use_fsm)
{
/*
* We have no cached target page, so ask the FSM for an initial
* target.
*/
targetBlock = GetPageWithFreeSpace(relation, len + saveFreeSpace);
/*
* If the FSM knows nothing of the rel, try the last page before we
* give up and extend. This avoids one-tuple-per-page syndrome during
* bootstrapping or in a recently-started system.
*/
if (targetBlock == InvalidBlockNumber)
{
BlockNumber nblocks = RelationGetNumberOfBlocks(relation);
if (nblocks > 0)
targetBlock = nblocks - 1;
}
}
while (targetBlock != InvalidBlockNumber)
{
/*
* Read and exclusive-lock the target block, as well as the other
* block if one was given, taking suitable care with lock ordering and
* the possibility they are the same block.
*
* If the page-level all-visible flag is set, caller will need to
* clear both that and the corresponding visibility map bit. However,
* by the time we return, we'll have x-locked the buffer, and we don't
* want to do any I/O while in that state. So we check the bit here
* before taking the lock, and pin the page if it appears necessary.
* Checking without the lock creates a risk of getting the wrong
* answer, so we'll have to recheck after acquiring the lock.
*/
if (otherBuffer == InvalidBuffer)
{
/* easy case */
buffer = ReadBufferBI(relation, targetBlock, bistate);
if (PageIsAllVisible(BufferGetPage(buffer)))
visibilitymap_pin(relation, targetBlock, vmbuffer);
LockBuffer(buffer, BUFFER_LOCK_EXCLUSIVE);
}
else if (otherBlock == targetBlock)
{
/* also easy case */
buffer = otherBuffer;
if (PageIsAllVisible(BufferGetPage(buffer)))
visibilitymap_pin(relation, targetBlock, vmbuffer);
LockBuffer(buffer, BUFFER_LOCK_EXCLUSIVE);
}
else if (otherBlock < targetBlock)
{
/* lock other buffer first */
buffer = ReadBuffer(relation, targetBlock);
if (PageIsAllVisible(BufferGetPage(buffer)))
visibilitymap_pin(relation, targetBlock, vmbuffer);
LockBuffer(otherBuffer, BUFFER_LOCK_EXCLUSIVE);
LockBuffer(buffer, BUFFER_LOCK_EXCLUSIVE);
}
else
{
/* lock target buffer first */
buffer = ReadBuffer(relation, targetBlock);
if (PageIsAllVisible(BufferGetPage(buffer)))
visibilitymap_pin(relation, targetBlock, vmbuffer);
LockBuffer(buffer, BUFFER_LOCK_EXCLUSIVE);
LockBuffer(otherBuffer, BUFFER_LOCK_EXCLUSIVE);
}
/*
* We now have the target page (and the other buffer, if any) pinned
* and locked. However, since our initial PageIsAllVisible checks
* were performed before acquiring the lock, the results might now be
* out of date, either for the selected victim buffer, or for the
* other buffer passed by the caller. In that case, we'll need to
* give up our locks, go get the pin(s) we failed to get earlier, and
* re-lock. That's pretty painful, but hopefully shouldn't happen
* often.
*
* Note that there's a small possibility that we didn't pin the page
* above but still have the correct page pinned anyway, either because
* we've already made a previous pass through this loop, or because
* caller passed us the right page anyway.
*
* Note also that it's possible that by the time we get the pin and
* retake the buffer locks, the visibility map bit will have been
* cleared by some other backend anyway. In that case, we'll have
* done a bit of extra work for no gain, but there's no real harm
* done.
*/
if (otherBuffer == InvalidBuffer || buffer <= otherBuffer)
GetVisibilityMapPins(relation, buffer, otherBuffer,
targetBlock, otherBlock, vmbuffer,
vmbuffer_other);
else
GetVisibilityMapPins(relation, otherBuffer, buffer,
otherBlock, targetBlock, vmbuffer_other,
vmbuffer);
/*
* Now we can check to see if there's enough free space here. If so,
* we're done.
*/
page = BufferGetPage(buffer);
pageFreeSpace = PageGetHeapFreeSpace(page);
if (len + saveFreeSpace <= pageFreeSpace)
{
/* use this page as future insert target, too */
RelationSetTargetBlock(relation, targetBlock);
return buffer;
}
/*
* Not enough space, so we must give up our page locks and pin (if
* any) and prepare to look elsewhere. We don't care which order we
* unlock the two buffers in, so this can be slightly simpler than the
* code above.
*/
LockBuffer(buffer, BUFFER_LOCK_UNLOCK);
if (otherBuffer == InvalidBuffer)
ReleaseBuffer(buffer);
else if (otherBlock != targetBlock)
{
LockBuffer(otherBuffer, BUFFER_LOCK_UNLOCK);
ReleaseBuffer(buffer);
}
/* Without FSM, always fall out of the loop and extend */
if (!use_fsm)
break;
/*
* Update FSM as to condition of this page, and ask for another page
* to try.
*/
targetBlock = RecordAndGetPageWithFreeSpace(relation,
targetBlock,
pageFreeSpace,
len + saveFreeSpace);
}
/*
* Have to extend the relation.
*
* We have to use a lock to ensure no one else is extending the rel at the
* same time, else we will both try to initialize the same new page. We
* can skip locking for new or temp relations, however, since no one else
* could be accessing them.
*/
needLock = !RELATION_IS_LOCAL(relation);
if (needLock)
LockRelationForExtension(relation, ExclusiveLock);
/*
* XXX This does an lseek - rather expensive - but at the moment it is the
* only way to accurately determine how many blocks are in a relation. Is
* it worth keeping an accurate file length in shared memory someplace,
* rather than relying on the kernel to do it for us?
*/
buffer = ReadBufferBI(relation, P_NEW, bistate);
/*
* We can be certain that locking the otherBuffer first is OK, since it
* must have a lower page number.
*/
if (otherBuffer != InvalidBuffer)
LockBuffer(otherBuffer, BUFFER_LOCK_EXCLUSIVE);
/*
* Now acquire lock on the new page.
*/
LockBuffer(buffer, BUFFER_LOCK_EXCLUSIVE);
/*
* Release the file-extension lock; it's now OK for someone else to extend
* the relation some more. Note that we cannot release this lock before
* we have buffer lock on the new page, or we risk a race condition
* against vacuumlazy.c --- see comments therein.
*/
if (needLock)
UnlockRelationForExtension(relation, ExclusiveLock);
/*
* We need to initialize the empty new page. Double-check that it really
* is empty (this should never happen, but if it does we don't want to
* risk wiping out valid data).
*/
page = BufferGetPage(buffer);
if (!PageIsNew(page))
elog(ERROR, "page %u of relation \"%s\" should be empty but is not",
BufferGetBlockNumber(buffer),
RelationGetRelationName(relation));
PageInit(page, BufferGetPageSize(buffer), 0);
if (len > PageGetHeapFreeSpace(page))
{
/* We should not get here given the test at the top */
elog(PANIC, "tuple is too big: size %lu", (unsigned long) len);
}
/*
* Remember the new page as our target for future insertions.
*
* XXX should we enter the new page into the free space map immediately,
* or just keep it for this backend's exclusive use in the short run
* (until VACUUM sees it)? Seems to depend on whether you expect the
* current backend to make more insertions or not, which is probably a
* good bet most of the time. So for now, don't add it to FSM yet.
*/
RelationSetTargetBlock(relation, BufferGetBlockNumber(buffer));
return buffer;
}
void PersistentStore_ReplaceTuple(
PersistentStoreData *storeData,
PersistentStoreSharedData *storeSharedData,
ItemPointer persistentTid,
/* TID of the stored tuple. */
HeapTuple tuple,
Datum *newValues,
bool *replaces,
bool flushToXLog)
/* When true, the XLOG record for this change will be flushed to disk. */
{
Relation persistentRel;
bool *nulls;
HeapTuple replacementTuple = NULL;
XLogRecPtr xlogUpdateEndLoc;
#ifdef USE_ASSERT_CHECKING
if (storeSharedData == NULL ||
!PersistentStoreSharedData_EyecatcherIsValid(storeSharedData))
elog(ERROR, "Persistent store shared-memory not valid");
#endif
if (Debug_persistent_store_print)
elog(PersistentStore_DebugPrintLevel(),
"PersistentStore_ReplaceTuple: Going to replace set of columns in tuple at TID %s ('%s', shared data %p)",
ItemPointerToString(persistentTid),
storeData->tableName,
storeSharedData);
persistentRel = (*storeData->openRel)();
/*
* In order to keep the tuples the exact same size to enable direct reuse of
* free tuples, we do not use NULLs.
*/
nulls = (bool*)palloc0(storeData->numAttributes * sizeof(bool));
/*
* Modify the tuple.
*/
replacementTuple = heap_modify_tuple(tuple, persistentRel->rd_att,
newValues, nulls, replaces);
replacementTuple->t_self = *persistentTid;
frozen_heap_inplace_update(persistentRel, replacementTuple);
/*
* Return the XLOG location of the UPDATE tuple's XLOG record.
*/
xlogUpdateEndLoc = XLogLastInsertEndLoc();
heap_freetuple(replacementTuple);
pfree(nulls);
if (Debug_persistent_store_print)
{
Datum *readValues;
bool *readNulls;
HeapTupleData readTuple;
Buffer buffer;
HeapTuple readTupleCopy;
elog(PersistentStore_DebugPrintLevel(),
"PersistentStore_ReplaceTuple: Replaced set of columns in tuple at TID %s ('%s')",
ItemPointerToString(persistentTid),
storeData->tableName);
readValues = (Datum*)palloc(storeData->numAttributes * sizeof(Datum));
readNulls = (bool*)palloc(storeData->numAttributes * sizeof(bool));
readTuple.t_self = *persistentTid;
if (!heap_fetch(persistentRel, SnapshotAny,
&readTuple, &buffer, false, NULL))
{
elog(ERROR, "Failed to fetch persistent tuple at %s ('%s')",
ItemPointerToString(&readTuple.t_self),
storeData->tableName);
}
readTupleCopy = heaptuple_copy_to(&readTuple, NULL, NULL);
ReleaseBuffer(buffer);
heap_deform_tuple(readTupleCopy, persistentRel->rd_att, readValues, readNulls);
(*storeData->printTupleCallback)(
PersistentStore_DebugPrintLevel(),
"STORE REPLACED TUPLE",
persistentTid,
readValues);
heap_freetuple(readTupleCopy);
pfree(readValues);
pfree(readNulls);
}
(*storeData->closeRel)(persistentRel);
if (flushToXLog)
{
XLogFlush(xlogUpdateEndLoc);
XLogRecPtr_Zero(&nowaitXLogEndLoc);
}
else
nowaitXLogEndLoc = xlogUpdateEndLoc;
}
/*
* _hash_dropbuf() -- release an unlocked buffer.
*
* This is used to unpin a buffer on which we hold no lock.
*/
void
_hash_dropbuf(Relation rel, Buffer buf)
{
ReleaseBuffer(buf);
}
STDMETHODIMP CDecWMV9::ProcessOutput()
{
HRESULT hr = S_OK;
DWORD dwStatus = 0;
BYTE *pBuffer = GetBuffer(m_pRawBufferSize);
CMediaBuffer *pOutBuffer = new CMediaBuffer(pBuffer, m_pRawBufferSize, true);
pOutBuffer->SetLength(0);
DMO_OUTPUT_DATA_BUFFER OutputBufferStructs[1];
memset(&OutputBufferStructs[0], 0, sizeof(DMO_OUTPUT_DATA_BUFFER));
OutputBufferStructs[0].pBuffer = pOutBuffer;
hr = m_pDMO->ProcessOutput(0, 1, OutputBufferStructs, &dwStatus);
if (FAILED(hr)) {
ReleaseBuffer(pBuffer);
DbgLog((LOG_TRACE, 10, L"-> ProcessOutput failed with hr: %x", hr));
return S_FALSE;
}
if (hr == S_FALSE) {
ReleaseBuffer(pBuffer);
return S_FALSE;
}
LAVFrame *pFrame = nullptr;
AllocateFrame(&pFrame);
BITMAPINFOHEADER *pBMI = nullptr;
videoFormatTypeHandler(mtOut, &pBMI);
pFrame->width = pBMI->biWidth;
pFrame->height = pBMI->biHeight;
pFrame->format = m_OutPixFmt;
pFrame->key_frame = (OutputBufferStructs[0].dwStatus & DMO_OUTPUT_DATA_BUFFERF_SYNCPOINT);
AVRational display_aspect_ratio;
int64_t num = (int64_t)m_StreamAR.num * pBMI->biWidth;
int64_t den = (int64_t)m_StreamAR.den * pBMI->biHeight;
av_reduce(&display_aspect_ratio.num, &display_aspect_ratio.den, num, den, INT_MAX);
BYTE contentType = 0;
DWORD dwPropSize = 1;
pOutBuffer->GetProperty(WM_SampleExtensionGUID_ContentType, &contentType, &dwPropSize);
pFrame->interlaced = !!(contentType & WM_CT_INTERLACED);
pFrame->repeat = !!(contentType & WM_CT_REPEAT_FIRST_FIELD);
LAVDeintFieldOrder fo = m_pSettings->GetDeintFieldOrder();
pFrame->tff = (fo == DeintFieldOrder_Auto) ? !!(contentType & WM_CT_TOP_FIELD_FIRST) : (fo == DeintFieldOrder_TopFieldFirst);
if (pFrame->interlaced && !m_bInterlaced)
m_bInterlaced = TRUE;
pFrame->interlaced = (pFrame->interlaced || (m_bInterlaced && m_pSettings->GetDeinterlacingMode() == DeintMode_Aggressive) || m_pSettings->GetDeinterlacingMode() == DeintMode_Force) && !(m_pSettings->GetDeinterlacingMode() == DeintMode_Disable);
if (m_bManualReorder) {
if (!m_timestampQueue.empty()) {
pFrame->rtStart = m_timestampQueue.front();
m_timestampQueue.pop();
if (OutputBufferStructs[0].dwStatus & DMO_OUTPUT_DATA_BUFFERF_TIMELENGTH) {
pFrame->rtStop = pFrame->rtStart + OutputBufferStructs[0].rtTimelength;
}
}
} else {
if (OutputBufferStructs[0].dwStatus & DMO_OUTPUT_DATA_BUFFERF_TIME) {
pFrame->rtStart = OutputBufferStructs[0].rtTimestamp;
if (OutputBufferStructs[0].dwStatus & DMO_OUTPUT_DATA_BUFFERF_TIMELENGTH) {
pFrame->rtStop = pFrame->rtStart + OutputBufferStructs[0].rtTimelength;
}
}
}
// Check alignment
// If not properly aligned, we need to make the data aligned.
int alignment = (m_OutPixFmt == LAVPixFmt_NV12) ? 16 : 32;
if ((pFrame->width % alignment) != 0) {
AllocLAVFrameBuffers(pFrame);
size_t ySize = pFrame->width * pFrame->height;
memcpy_plane(pFrame->data[0], pBuffer, pFrame->width, pFrame->stride[0], pFrame->height);
if (m_OutPixFmt == LAVPixFmt_NV12) {
memcpy_plane(pFrame->data[1], pBuffer+ySize, pFrame->width, pFrame->stride[1], pFrame->height / 2);
} else if (m_OutPixFmt == LAVPixFmt_YUV420) {
size_t uvSize = ySize / 4;
memcpy_plane(pFrame->data[2], pBuffer+ySize, pFrame->width / 2, pFrame->stride[2], pFrame->height / 2);
memcpy_plane(pFrame->data[1], pBuffer+ySize+uvSize, pFrame->width / 2, pFrame->stride[1], pFrame->height / 2);
}
ReleaseBuffer(pBuffer);
} else {
if (m_OutPixFmt == LAVPixFmt_NV12) {
pFrame->data[0] = pBuffer;
pFrame->data[1] = pBuffer + pFrame->width * pFrame->height;
pFrame->stride[0] = pFrame->stride[1] = pFrame->width;
} else if (m_OutPixFmt == LAVPixFmt_YUV420) {
pFrame->data[0] = pBuffer;
pFrame->data[2] = pBuffer + pFrame->width * pFrame->height;
pFrame->data[1] = pFrame->data[2] + (pFrame->width / 2) * (pFrame->height / 2);
pFrame->stride[0] = pFrame->width;
pFrame->stride[1] = pFrame->stride[2] = pFrame->width / 2;
}
pFrame->destruct = wmv9_buffer_destruct;
pFrame->priv_data = this;
}
pFrame->flags |= LAV_FRAME_FLAG_BUFFER_MODIFY;
Deliver(pFrame);
SafeRelease(&pOutBuffer);
if (OutputBufferStructs[0].dwStatus & DMO_OUTPUT_DATA_BUFFERF_INCOMPLETE)
return ProcessOutput();
return hr;
}
/*
* Search the relation 'rel' for tuple using the index.
*
* If a matching tuple is found, lock it with lockmode, fill the slot with its
* contents, and return true. Return false otherwise.
*/
bool
RelationFindReplTupleByIndex(Relation rel, Oid idxoid,
LockTupleMode lockmode,
TupleTableSlot *searchslot,
TupleTableSlot *outslot)
{
HeapTuple scantuple;
ScanKeyData skey[INDEX_MAX_KEYS];
IndexScanDesc scan;
SnapshotData snap;
TransactionId xwait;
Relation idxrel;
bool found;
/* Open the index. */
idxrel = index_open(idxoid, RowExclusiveLock);
/* Start an index scan. */
InitDirtySnapshot(snap);
scan = index_beginscan(rel, idxrel, &snap,
IndexRelationGetNumberOfKeyAttributes(idxrel),
0);
/* Build scan key. */
build_replindex_scan_key(skey, rel, idxrel, searchslot);
retry:
found = false;
index_rescan(scan, skey, IndexRelationGetNumberOfKeyAttributes(idxrel), NULL, 0);
/* Try to find the tuple */
if ((scantuple = index_getnext(scan, ForwardScanDirection)) != NULL)
{
found = true;
ExecStoreTuple(scantuple, outslot, InvalidBuffer, false);
ExecMaterializeSlot(outslot);
xwait = TransactionIdIsValid(snap.xmin) ?
snap.xmin : snap.xmax;
/*
* If the tuple is locked, wait for locking transaction to finish and
* retry.
*/
if (TransactionIdIsValid(xwait))
{
XactLockTableWait(xwait, NULL, NULL, XLTW_None);
goto retry;
}
}
/* Found tuple, try to lock it in the lockmode. */
if (found)
{
Buffer buf;
HeapUpdateFailureData hufd;
HTSU_Result res;
HeapTupleData locktup;
ItemPointerCopy(&outslot->tts_tuple->t_self, &locktup.t_self);
PushActiveSnapshot(GetLatestSnapshot());
res = heap_lock_tuple(rel, &locktup, GetCurrentCommandId(false),
lockmode,
LockWaitBlock,
false /* don't follow updates */ ,
&buf, &hufd);
/* the tuple slot already has the buffer pinned */
ReleaseBuffer(buf);
PopActiveSnapshot();
switch (res)
{
case HeapTupleMayBeUpdated:
break;
case HeapTupleUpdated:
/* XXX: Improve handling here */
if (ItemPointerIndicatesMovedPartitions(&hufd.ctid))
ereport(LOG,
(errcode(ERRCODE_T_R_SERIALIZATION_FAILURE),
errmsg("tuple to be locked was already moved to another partition due to concurrent update, retrying")));
else
ereport(LOG,
(errcode(ERRCODE_T_R_SERIALIZATION_FAILURE),
errmsg("concurrent update, retrying")));
goto retry;
case HeapTupleInvisible:
elog(ERROR, "attempted to lock invisible tuple");
default:
elog(ERROR, "unexpected heap_lock_tuple status: %u", res);
break;
}
}
index_endscan(scan);
/* Don't release lock until commit. */
index_close(idxrel, NoLock);
return found;
}
/*
* Search the relation 'rel' for tuple using the sequential scan.
*
* If a matching tuple is found, lock it with lockmode, fill the slot with its
* contents, and return true. Return false otherwise.
*
* Note that this stops on the first matching tuple.
*
* This can obviously be quite slow on tables that have more than few rows.
*/
bool
RelationFindReplTupleSeq(Relation rel, LockTupleMode lockmode,
TupleTableSlot *searchslot, TupleTableSlot *outslot)
{
HeapTuple scantuple;
HeapScanDesc scan;
SnapshotData snap;
TransactionId xwait;
bool found;
TupleDesc desc = RelationGetDescr(rel);
Assert(equalTupleDescs(desc, outslot->tts_tupleDescriptor));
/* Start a heap scan. */
InitDirtySnapshot(snap);
scan = heap_beginscan(rel, &snap, 0, NULL);
retry:
found = false;
heap_rescan(scan, NULL);
/* Try to find the tuple */
while ((scantuple = heap_getnext(scan, ForwardScanDirection)) != NULL)
{
if (!tuple_equals_slot(desc, scantuple, searchslot))
continue;
found = true;
ExecStoreTuple(scantuple, outslot, InvalidBuffer, false);
ExecMaterializeSlot(outslot);
xwait = TransactionIdIsValid(snap.xmin) ?
snap.xmin : snap.xmax;
/*
* If the tuple is locked, wait for locking transaction to finish and
* retry.
*/
if (TransactionIdIsValid(xwait))
{
XactLockTableWait(xwait, NULL, NULL, XLTW_None);
goto retry;
}
}
/* Found tuple, try to lock it in the lockmode. */
if (found)
{
Buffer buf;
HeapUpdateFailureData hufd;
HTSU_Result res;
HeapTupleData locktup;
ItemPointerCopy(&outslot->tts_tuple->t_self, &locktup.t_self);
PushActiveSnapshot(GetLatestSnapshot());
res = heap_lock_tuple(rel, &locktup, GetCurrentCommandId(false),
lockmode,
LockWaitBlock,
false /* don't follow updates */ ,
&buf, &hufd);
/* the tuple slot already has the buffer pinned */
ReleaseBuffer(buf);
PopActiveSnapshot();
switch (res)
{
case HeapTupleMayBeUpdated:
break;
case HeapTupleUpdated:
/* XXX: Improve handling here */
if (ItemPointerIndicatesMovedPartitions(&hufd.ctid))
ereport(LOG,
(errcode(ERRCODE_T_R_SERIALIZATION_FAILURE),
errmsg("tuple to be locked was already moved to another partition due to concurrent update, retrying")));
else
ereport(LOG,
(errcode(ERRCODE_T_R_SERIALIZATION_FAILURE),
errmsg("concurrent update, retrying")));
goto retry;
case HeapTupleInvisible:
elog(ERROR, "attempted to lock invisible tuple");
default:
elog(ERROR, "unexpected heap_lock_tuple status: %u", res);
break;
}
}
heap_endscan(scan);
return found;
}
void C_TCPServerConnection::run()
{
assert(m_pMsgConsolePrompter);
m_pMsgConsolePrompter->PromptMessageLineBy(_T("TCP_CONNECTION_0"));
socket().setReceiveTimeout(Poco::Timespan(90,0));
try
{
char* pszCmdBuffer = NULL;
while (true)
{
long nCmdLen = 0;
recvData(socket(), &pszCmdBuffer, nCmdLen);
if (nCmdLen == 0)
{
if (!m_szClientID.empty())
{
C_NetCmdLogOff logOff;
logOff.HandleRequest(socket(), NULL, m_DBOperate, m_szClientID.c_str());
}
m_pMsgConsolePrompter->PromptMessageLineBy(_T("TCP_CONNECTION_1"));
break;
}
if (nCmdLen > 0)
{
C_NetCommandDecomposer netCmdDecomposer;
if(!netCmdDecomposer.Decompose(pszCmdBuffer))
{
ReleaseBuffer(&pszCmdBuffer);
continue;
}
tstring szCmdName = netCmdDecomposer.GetCmdName().c_str();
C_NetCommand* pRequestCmd = m_NetCmder.CreateCommand(szCmdName.c_str(), netCmdDecomposer.GetCmdType().c_str());
if(pRequestCmd == NULL)
{
ReleaseBuffer(&pszCmdBuffer);
continue;
}
if(szCmdName ==_T("log on") || szCmdName ==_T("log off") || szCmdName == _T("fetch channels") || szCmdName == _T("fetch content"))
{
if(!m_bLogOn)
{
C_NetCmdLogOn *pLogOn = dynamic_cast<C_NetCmdLogOn*>(pRequestCmd);
if(pLogOn == NULL)
continue;
tstring szMac = netCmdDecomposer.GetCmdPara(_T("id"));
if( pLogOn->HandleRequest(socket(), pszCmdBuffer, m_DBOperate, szMac.c_str()) )
{
m_szClientID = szMac;
m_bLogOn = true;
ReleaseBuffer(&pszCmdBuffer);
pLogOn->Release();
C_AreaEvent::GetInstance()->PushEvent(socket(), szMac.c_str());
}
continue;
}
}
pRequestCmd->HandleRequest(socket(), pszCmdBuffer, m_DBOperate, m_szClientID.c_str());
pRequestCmd->Release();
ReleaseBuffer(&pszCmdBuffer);
}
}
}
catch (Poco::Exception& exc)
{
if (!m_szClientID.empty())
{
C_NetCmdLogOff logOff;
logOff.HandleRequest(socket(), NULL, m_DBOperate, m_szClientID.c_str());
}
m_pMsgConsolePrompter->PromptMessageLineBy(_T("TCP_CONNECTION_2"), exc.what());
return;
}
}
/*
* workhorse
*/
static bytea *
get_raw_page_internal(text *relname, ForkNumber forknum, BlockNumber blkno)
{
bytea *raw_page;
RangeVar *relrv;
Relation rel;
char *raw_page_data;
Buffer buf;
if (!superuser())
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
(errmsg("must be superuser to use raw functions"))));
relrv = makeRangeVarFromNameList(textToQualifiedNameList(relname));
rel = relation_openrv(relrv, AccessShareLock);
/* Check that this relation has storage */
if (rel->rd_rel->relkind == RELKIND_VIEW)
ereport(ERROR,
(errcode(ERRCODE_WRONG_OBJECT_TYPE),
errmsg("cannot get raw page from view \"%s\"",
RelationGetRelationName(rel))));
if (rel->rd_rel->relkind == RELKIND_COMPOSITE_TYPE)
ereport(ERROR,
(errcode(ERRCODE_WRONG_OBJECT_TYPE),
errmsg("cannot get raw page from composite type \"%s\"",
RelationGetRelationName(rel))));
/*
* Reject attempts to read non-local temporary relations; we would be
* likely to get wrong data since we have no visibility into the owning
* session's local buffers.
*/
if (RELATION_IS_OTHER_TEMP(rel))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("cannot access temporary tables of other sessions")));
if (blkno >= RelationGetNumberOfBlocks(rel))
elog(ERROR, "block number %u is out of range for relation \"%s\"",
blkno, RelationGetRelationName(rel));
/* Initialize buffer to copy to */
raw_page = (bytea *) palloc(BLCKSZ + VARHDRSZ);
SET_VARSIZE(raw_page, BLCKSZ + VARHDRSZ);
raw_page_data = VARDATA(raw_page);
/* Take a verbatim copy of the page */
buf = ReadBufferExtended(rel, forknum, blkno, RBM_NORMAL, NULL);
LockBuffer(buf, BUFFER_LOCK_SHARE);
memcpy(raw_page_data, BufferGetPage(buf), BLCKSZ);
LockBuffer(buf, BUFFER_LOCK_UNLOCK);
ReleaseBuffer(buf);
relation_close(rel, AccessShareLock);
return raw_page;
}
/*
* pg_prewarm(regclass, mode text, fork text,
* first_block int8, last_block int8)
*
* The first argument is the relation to be prewarmed; the second controls
* how prewarming is done; legal options are 'prefetch', 'read', and 'buffer'.
* The third is the name of the relation fork to be prewarmed. The fourth
* and fifth arguments specify the first and last block to be prewarmed.
* If the fourth argument is NULL, it will be taken as 0; if the fifth argument
* is NULL, it will be taken as the number of blocks in the relation. The
* return value is the number of blocks successfully prewarmed.
*/
Datum
pg_prewarm(PG_FUNCTION_ARGS)
{
Oid relOid;
text *forkName;
text *type;
int64 first_block;
int64 last_block;
int64 nblocks;
int64 blocks_done = 0;
int64 block;
Relation rel;
ForkNumber forkNumber;
char *forkString;
char *ttype;
PrewarmType ptype;
AclResult aclresult;
/* Basic sanity checking. */
if (PG_ARGISNULL(0))
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("relation cannot be null")));
relOid = PG_GETARG_OID(0);
if (PG_ARGISNULL(1))
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
(errmsg("prewarm type cannot be null"))));
type = PG_GETARG_TEXT_P(1);
ttype = text_to_cstring(type);
if (strcmp(ttype, "prefetch") == 0)
ptype = PREWARM_PREFETCH;
else if (strcmp(ttype, "read") == 0)
ptype = PREWARM_READ;
else if (strcmp(ttype, "buffer") == 0)
ptype = PREWARM_BUFFER;
else
{
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("invalid prewarm type"),
errhint("Valid prewarm types are \"prefetch\", \"read\", and \"buffer\".")));
PG_RETURN_INT64(0); /* Placate compiler. */
}
if (PG_ARGISNULL(2))
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
(errmsg("relation fork cannot be null"))));
forkName = PG_GETARG_TEXT_P(2);
forkString = text_to_cstring(forkName);
forkNumber = forkname_to_number(forkString);
/* Open relation and check privileges. */
rel = relation_open(relOid, AccessShareLock);
aclresult = pg_class_aclcheck(relOid, GetUserId(), ACL_SELECT);
if (aclresult != ACLCHECK_OK)
aclcheck_error(aclresult, ACL_KIND_CLASS, get_rel_name(relOid));
/* Check that the fork exists. */
RelationOpenSmgr(rel);
if (!smgrexists(rel->rd_smgr, forkNumber))
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("fork \"%s\" does not exist for this relation",
forkString)));
/* Validate block numbers, or handle nulls. */
nblocks = RelationGetNumberOfBlocksInFork(rel, forkNumber);
if (PG_ARGISNULL(3))
first_block = 0;
else
{
first_block = PG_GETARG_INT64(3);
if (first_block < 0 || first_block >= nblocks)
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("starting block number must be between 0 and " INT64_FORMAT,
nblocks - 1)));
}
if (PG_ARGISNULL(4))
last_block = nblocks - 1;
else
{
last_block = PG_GETARG_INT64(4);
if (last_block < 0 || last_block >= nblocks)
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("ending block number must be between 0 and " INT64_FORMAT,
nblocks - 1)));
}
/* Now we're ready to do the real work. */
if (ptype == PREWARM_PREFETCH)
{
#ifdef USE_PREFETCH
/*
* In prefetch mode, we just hint the OS to read the blocks, but we
* don't know whether it really does it, and we don't wait for it to
* finish.
*
* It would probably be better to pass our prefetch requests in chunks
* of a megabyte or maybe even a whole segment at a time, but there's
* no practical way to do that at present without a gross modularity
* violation, so we just do this.
*/
for (block = first_block; block <= last_block; ++block)
{
CHECK_FOR_INTERRUPTS();
PrefetchBuffer(rel, forkNumber, block);
++blocks_done;
}
#else
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("prefetch is not supported by this build")));
#endif
}
else if (ptype == PREWARM_READ)
{
/*
* In read mode, we actually read the blocks, but not into shared
* buffers. This is more portable than prefetch mode (it works
* everywhere) and is synchronous.
*/
for (block = first_block; block <= last_block; ++block)
{
CHECK_FOR_INTERRUPTS();
smgrread(rel->rd_smgr, forkNumber, block, blockbuffer.data);
++blocks_done;
}
}
else if (ptype == PREWARM_BUFFER)
{
/*
* In buffer mode, we actually pull the data into shared_buffers.
*/
for (block = first_block; block <= last_block; ++block)
{
Buffer buf;
CHECK_FOR_INTERRUPTS();
buf = ReadBufferExtended(rel, forkNumber, block, RBM_NORMAL, NULL);
ReleaseBuffer(buf);
++blocks_done;
}
}
/* Close relation, release lock. */
relation_close(rel, AccessShareLock);
PG_RETURN_INT64(blocks_done);
}
FarString& FarString::TrimRight()
{
FarSF::RTrim (GetBuffer());
ReleaseBuffer();
return * this;
}
/*
* 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++;
}
FarString& FarString::Unquote()
{
FarSF::Unquote( GetBuffer() );
ReleaseBuffer();
return * this;
}
/*
* scans posting tree and deletes empty pages
*/
static bool
ginScanToDelete(GinVacuumState *gvs, BlockNumber blkno, bool isRoot,
DataPageDeleteStack *parent, OffsetNumber myoff)
{
DataPageDeleteStack *me;
Buffer buffer;
Page page;
bool meDelete = FALSE;
bool isempty;
if (isRoot)
{
me = parent;
}
else
{
if (!parent->child)
{
me = (DataPageDeleteStack *) palloc0(sizeof(DataPageDeleteStack));
me->parent = parent;
parent->child = me;
me->leftBlkno = InvalidBlockNumber;
}
else
me = parent->child;
}
buffer = ReadBufferExtended(gvs->index, MAIN_FORKNUM, blkno,
RBM_NORMAL, gvs->strategy);
page = BufferGetPage(buffer);
Assert(GinPageIsData(page));
if (!GinPageIsLeaf(page))
{
OffsetNumber i;
me->blkno = blkno;
for (i = FirstOffsetNumber; i <= GinPageGetOpaque(page)->maxoff; i++)
{
PostingItem *pitem = GinDataPageGetPostingItem(page, i);
if (ginScanToDelete(gvs, PostingItemGetBlockNumber(pitem), FALSE, me, i))
i--;
}
}
if (GinPageIsLeaf(page))
isempty = GinDataLeafPageIsEmpty(page);
else
isempty = GinPageGetOpaque(page)->maxoff < FirstOffsetNumber;
if (isempty)
{
/* we never delete the left- or rightmost branch */
if (me->leftBlkno != InvalidBlockNumber && !GinPageRightMost(page))
{
Assert(!isRoot);
ginDeletePage(gvs, blkno, me->leftBlkno, me->parent->blkno, myoff, me->parent->isRoot);
meDelete = TRUE;
}
}
ReleaseBuffer(buffer);
if (!meDelete)
me->leftBlkno = blkno;
return meDelete;
}
/*
* Try to find parent for current stack position, returns correct
* parent and child's offset in stack->parent.
* Function should never release root page to prevent conflicts
* with vacuum process
*/
void
ginFindParents(GinBtree btree, GinBtreeStack *stack,
BlockNumber rootBlkno)
{
Page page;
Buffer buffer;
BlockNumber blkno,
leftmostBlkno;
OffsetNumber offset;
GinBtreeStack *root = stack->parent;
GinBtreeStack *ptr;
if (!root)
{
/* XLog mode... */
root = (GinBtreeStack *) palloc(sizeof(GinBtreeStack));
root->blkno = rootBlkno;
root->buffer = ReadBuffer(btree->index, rootBlkno);
LockBuffer(root->buffer, GIN_EXCLUSIVE);
root->parent = NULL;
}
else
{
/*
* find root, we should not release root page until update is
* finished!!
*/
while (root->parent)
{
ReleaseBuffer(root->buffer);
root = root->parent;
}
Assert(root->blkno == rootBlkno);
Assert(BufferGetBlockNumber(root->buffer) == rootBlkno);
LockBuffer(root->buffer, GIN_EXCLUSIVE);
}
root->off = InvalidOffsetNumber;
page = BufferGetPage(root->buffer);
Assert(!GinPageIsLeaf(page));
/* check trivial case */
if ((root->off = btree->findChildPtr(btree, page, stack->blkno, InvalidOffsetNumber)) != InvalidOffsetNumber)
{
stack->parent = root;
return;
}
leftmostBlkno = blkno = btree->getLeftMostPage(btree, page);
LockBuffer(root->buffer, GIN_UNLOCK);
Assert(blkno != InvalidBlockNumber);
for (;;)
{
buffer = ReadBuffer(btree->index, blkno);
LockBuffer(buffer, GIN_EXCLUSIVE);
page = BufferGetPage(buffer);
if (GinPageIsLeaf(page))
elog(ERROR, "Lost path");
leftmostBlkno = btree->getLeftMostPage(btree, page);
while ((offset = btree->findChildPtr(btree, page, stack->blkno, InvalidOffsetNumber)) == InvalidOffsetNumber)
{
blkno = GinPageGetOpaque(page)->rightlink;
LockBuffer(buffer, GIN_UNLOCK);
ReleaseBuffer(buffer);
if (blkno == InvalidBlockNumber)
break;
buffer = ReadBuffer(btree->index, blkno);
LockBuffer(buffer, GIN_EXCLUSIVE);
page = BufferGetPage(buffer);
}
if (blkno != InvalidBlockNumber)
{
ptr = (GinBtreeStack *) palloc(sizeof(GinBtreeStack));
ptr->blkno = blkno;
ptr->buffer = buffer;
ptr->parent = root; /* it's may be wrong, but in next call we will
* correct */
ptr->off = offset;
stack->parent = ptr;
return;
}
blkno = leftmostBlkno;
}
}
STDMETHODIMP CDecWMV9MFT::ProcessOutput()
{
HRESULT hr = S_OK;
DWORD dwStatus = 0;
MFT_OUTPUT_STREAM_INFO outputInfo = {0};
m_pMFT->GetOutputStreamInfo(0, &outputInfo);
IMFMediaBuffer *pMFBuffer = nullptr;
ASSERT(!(outputInfo.dwFlags & MFT_OUTPUT_STREAM_PROVIDES_SAMPLES));
MFT_OUTPUT_DATA_BUFFER OutputBuffer = {0};
if (!(outputInfo.dwFlags & MFT_OUTPUT_STREAM_PROVIDES_SAMPLES)) {
pMFBuffer = GetBuffer(outputInfo.cbSize);
if (!pMFBuffer) { DbgLog((LOG_TRACE, 10, L"Unable to allocate media buffere")); return E_FAIL; }
IMFSample *pSampleOut = nullptr;
hr = MF.CreateSample(&pSampleOut);
if (FAILED(hr)) { DbgLog((LOG_TRACE, 10, L"Unable to allocate MF sample, hr: 0x%x", hr)); ReleaseBuffer(pMFBuffer); return E_FAIL; }
pSampleOut->AddBuffer(pMFBuffer);
OutputBuffer.pSample = pSampleOut;
}
hr = m_pMFT->ProcessOutput(0, 1, &OutputBuffer, &dwStatus);
// We don't process events, just release them
SafeRelease(&OutputBuffer.pEvents);
// handle stream format changes
if (hr == MF_E_TRANSFORM_STREAM_CHANGE || OutputBuffer.dwStatus == MFT_OUTPUT_DATA_BUFFER_FORMAT_CHANGE ) {
SafeRelease(&OutputBuffer.pSample);
ReleaseBuffer(pMFBuffer);
hr = SelectOutputType();
if (FAILED(hr)) {
DbgLog((LOG_TRACE, 10, L"-> Failed to handle stream change, hr: %x", hr));
return E_FAIL;
}
// try again with the new type, it should work now!
return ProcessOutput();
}
// the MFT generated no output, discard the sample and return
if (hr == MF_E_TRANSFORM_NEED_MORE_INPUT || OutputBuffer.dwStatus == MFT_OUTPUT_DATA_BUFFER_NO_SAMPLE) {
SafeRelease(&OutputBuffer.pSample);
ReleaseBuffer(pMFBuffer);
return S_FALSE;
}
// unknown error condition
if (FAILED(hr)) {
DbgLog((LOG_TRACE, 10, L"-> ProcessOutput failed with hr: %x", hr));
SafeRelease(&OutputBuffer.pSample);
ReleaseBuffer(pMFBuffer);
return E_FAIL;
}
LAVFrame *pFrame = nullptr;
AllocateFrame(&pFrame);
IMFMediaType *pMTOut = nullptr;
m_pMFT->GetOutputCurrentType(0, &pMTOut);
MFGetAttributeSize(pMTOut, MF_MT_FRAME_SIZE, (UINT32 *)&pFrame->width, (UINT32 *)&pFrame->height);
pFrame->format = m_OutPixFmt;
AVRational pixel_aspect_ratio = {1, 1};
MFGetAttributeRatio(pMTOut, MF_MT_PIXEL_ASPECT_RATIO, (UINT32*)&pixel_aspect_ratio.num, (UINT32*)&pixel_aspect_ratio.den);
AVRational display_aspect_ratio = {0, 0};
av_reduce(&display_aspect_ratio.num, &display_aspect_ratio.den, (int64_t)pixel_aspect_ratio.num * pFrame->width, (int64_t)pixel_aspect_ratio.den * pFrame->height, INT_MAX);
pFrame->interlaced = MFGetAttributeUINT32(OutputBuffer.pSample, MFSampleExtension_Interlaced, FALSE);
pFrame->repeat = MFGetAttributeUINT32(OutputBuffer.pSample, MFSampleExtension_RepeatFirstField, FALSE);
LAVDeintFieldOrder fo = m_pSettings->GetDeintFieldOrder();
pFrame->tff = (fo == DeintFieldOrder_Auto) ? !MFGetAttributeUINT32(OutputBuffer.pSample, MFSampleExtension_BottomFieldFirst, FALSE) : (fo == DeintFieldOrder_TopFieldFirst);
if (pFrame->interlaced && !m_bInterlaced)
m_bInterlaced = TRUE;
pFrame->interlaced = (pFrame->interlaced || (m_bInterlaced && m_pSettings->GetDeinterlacingMode() == DeintMode_Aggressive) || m_pSettings->GetDeinterlacingMode() == DeintMode_Force) && !(m_pSettings->GetDeinterlacingMode() == DeintMode_Disable);
pFrame->ext_format.VideoPrimaries = MFGetAttributeUINT32(pMTOut, MF_MT_VIDEO_PRIMARIES, MFVideoPrimaries_Unknown);
pFrame->ext_format.VideoTransferFunction = MFGetAttributeUINT32(pMTOut, MF_MT_TRANSFER_FUNCTION, MFVideoTransFunc_Unknown);
pFrame->ext_format.VideoTransferMatrix = MFGetAttributeUINT32(pMTOut, MF_MT_YUV_MATRIX, MFVideoTransferMatrix_Unknown);
pFrame->ext_format.VideoChromaSubsampling = MFGetAttributeUINT32(pMTOut, MF_MT_VIDEO_CHROMA_SITING, MFVideoChromaSubsampling_Unknown);
pFrame->ext_format.NominalRange = MFGetAttributeUINT32(pMTOut, MF_MT_VIDEO_NOMINAL_RANGE, MFNominalRange_Unknown);
// HACK: don't flag range=limited if its the only value set, since its also the implied default, this helps to avoid a reconnect
// The MFT always sets this value, even if the bitstream says nothing about it, causing a reconnect on every vc1/wmv3 file
if (pFrame->ext_format.value == 0x2000)
pFrame->ext_format.value = 0;
// Timestamps
if (m_bManualReorder) {
if (!m_timestampQueue.empty()) {
pFrame->rtStart = m_timestampQueue.front();
m_timestampQueue.pop();
LONGLONG llDuration = 0;
hr = OutputBuffer.pSample->GetSampleDuration(&llDuration);
if (SUCCEEDED(hr) && llDuration > 0) {
pFrame->rtStop = pFrame->rtStart + llDuration;
}
}
} else {
LONGLONG llTimestamp = 0;
hr = OutputBuffer.pSample->GetSampleTime(&llTimestamp);
if (SUCCEEDED(hr)) {
pFrame->rtStart = llTimestamp;
LONGLONG llDuration = 0;
hr = OutputBuffer.pSample->GetSampleDuration(&llDuration);
if (SUCCEEDED(hr) && llDuration > 0) {
pFrame->rtStop = pFrame->rtStart + llDuration;
}
}
}
SafeRelease(&pMTOut);
// Lock memory in the buffer
BYTE *pBuffer = nullptr;
pMFBuffer->Lock(&pBuffer, NULL, NULL);
// Check alignment
// If not properly aligned, we need to make the data aligned.
int alignment = (m_OutPixFmt == LAVPixFmt_NV12) ? 16 : 32;
if ((pFrame->width % alignment) != 0) {
hr = AllocLAVFrameBuffers(pFrame);
if (FAILED(hr)) {
pMFBuffer->Unlock();
ReleaseBuffer(pMFBuffer);
SafeRelease(&OutputBuffer.pSample);
return hr;
}
size_t ySize = pFrame->width * pFrame->height;
memcpy_plane(pFrame->data[0], pBuffer, pFrame->width, pFrame->stride[0], pFrame->height);
if (m_OutPixFmt == LAVPixFmt_NV12) {
memcpy_plane(pFrame->data[1], pBuffer + ySize, pFrame->width, pFrame->stride[1], pFrame->height / 2);
} else if (m_OutPixFmt == LAVPixFmt_YUV420) {
size_t uvSize = ySize / 4;
memcpy_plane(pFrame->data[2], pBuffer + ySize, pFrame->width / 2, pFrame->stride[2], pFrame->height / 2);
memcpy_plane(pFrame->data[1], pBuffer + ySize + uvSize, pFrame->width / 2, pFrame->stride[1], pFrame->height / 2);
}
pMFBuffer->Unlock();
ReleaseBuffer(pMFBuffer);
} else {
if (m_OutPixFmt == LAVPixFmt_NV12) {
pFrame->data[0] = pBuffer;
pFrame->data[1] = pBuffer + pFrame->width * pFrame->height;
pFrame->stride[0] = pFrame->stride[1] = pFrame->width;
} else if (m_OutPixFmt == LAVPixFmt_YUV420) {
pFrame->data[0] = pBuffer;
pFrame->data[2] = pBuffer + pFrame->width * pFrame->height;
pFrame->data[1] = pFrame->data[2] + (pFrame->width / 2) * (pFrame->height / 2);
pFrame->stride[0] = pFrame->width;
pFrame->stride[1] = pFrame->stride[2] = pFrame->width / 2;
}
pFrame->data[3] = (BYTE *)pMFBuffer;
pFrame->destruct = wmv9_buffer_destruct;
pFrame->priv_data = this;
}
pFrame->flags |= LAV_FRAME_FLAG_BUFFER_MODIFY;
Deliver(pFrame);
SafeRelease(&OutputBuffer.pSample);
if (OutputBuffer.dwStatus == MFT_OUTPUT_DATA_BUFFER_INCOMPLETE)
return ProcessOutput();
return hr;
}
void PersistentStore_ReadTuple(
PersistentStoreData *storeData,
PersistentStoreSharedData *storeSharedData,
ItemPointer readTid,
Datum *values,
HeapTuple *tupleCopy)
{
Relation persistentRel;
HeapTupleData tuple;
Buffer buffer;
bool *nulls;
#ifdef USE_ASSERT_CHECKING
if (storeSharedData == NULL ||
!PersistentStoreSharedData_EyecatcherIsValid(storeSharedData))
elog(ERROR, "Persistent store shared-memory not valid");
#endif
if (Debug_persistent_store_print)
elog(PersistentStore_DebugPrintLevel(),
"PersistentStore_ReadTuple: Going to read tuple at TID %s ('%s', shared data %p)",
ItemPointerToString(readTid),
storeData->tableName,
storeSharedData);
if (PersistentStore_IsZeroTid(readTid))
elog(ERROR, "TID for fetch persistent tuple is invalid (0,0) ('%s')",
storeData->tableName);
// UNDONE: I think the InRecovery test only applies to physical Master Mirroring on Standby.
/* Only test this outside of recovery scenarios */
if (!InRecovery
&&
(PersistentStore_IsZeroTid(&storeSharedData->maxTid)
||
ItemPointerCompare(
readTid,
&storeSharedData->maxTid) == 1 // Greater-than.
))
{
elog(ERROR, "TID %s for fetch persistent tuple is greater than the last known TID %s ('%s')",
ItemPointerToString(readTid),
ItemPointerToString2(&storeSharedData->maxTid),
storeData->tableName);
}
persistentRel = (*storeData->openRel)();
tuple.t_self = *readTid;
if (!heap_fetch(persistentRel, SnapshotAny,
&tuple, &buffer, false, NULL))
{
elog(ERROR, "Failed to fetch persistent tuple at %s (maximum known TID %s, '%s')",
ItemPointerToString(&tuple.t_self),
ItemPointerToString2(&storeSharedData->maxTid),
storeData->tableName);
}
*tupleCopy = heaptuple_copy_to(&tuple, NULL, NULL);
ReleaseBuffer(buffer);
/*
* In order to keep the tuples the exact same size to enable direct reuse of
* free tuples, we do not use NULLs.
*/
nulls = (bool*)palloc(storeData->numAttributes * sizeof(bool));
heap_deform_tuple(*tupleCopy, persistentRel->rd_att, values, nulls);
(*storeData->closeRel)(persistentRel);
if (Debug_persistent_store_print)
{
elog(PersistentStore_DebugPrintLevel(),
"PersistentStore_ReadTuple: Successfully read tuple at TID %s ('%s')",
ItemPointerToString(readTid),
storeData->tableName);
(*storeData->printTupleCallback)(
PersistentStore_DebugPrintLevel(),
"STORE READ TUPLE",
readTid,
values);
}
pfree(nulls);
}
LocalString::~LocalString() {
IOTJS_ASSERT(_strp != NULL);
ReleaseBuffer(const_cast<char*>(_strp));
}
/* ------------------------------------------------------
* pgstatindex()
*
* Usage: SELECT * FROM pgstatindex('t1_pkey');
* ------------------------------------------------------
*/
Datum
pgstatindex(PG_FUNCTION_ARGS)
{
text *relname = PG_GETARG_TEXT_P(0);
Relation rel;
RangeVar *relrv;
Datum result;
uint32 nblocks;
uint32 blkno;
BTIndexStat indexStat;
if (!superuser())
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
(errmsg("must be superuser to use pgstattuple functions"))));
relrv = makeRangeVarFromNameList(textToQualifiedNameList(relname));
rel = relation_openrv(relrv, AccessShareLock);
if (!IS_INDEX(rel) || !IS_BTREE(rel))
elog(ERROR, "pgstatindex() can be used only on b-tree index.");
/*
* Reject attempts to read non-local temporary relations; we would
* be likely to get wrong data since we have no visibility into the
* owning session's local buffers.
*/
if (isOtherTempNamespace(RelationGetNamespace(rel)))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("cannot access temporary tables of other sessions")));
/*-------------------
* Read a metapage
*-------------------
*/
{
Buffer buffer = ReadBuffer(rel, 0);
Page page = BufferGetPage(buffer);
BTMetaPageData *metad = BTPageGetMeta(page);
indexStat.magic = metad->btm_magic;
indexStat.version = metad->btm_version;
indexStat.root_blkno = metad->btm_root;
indexStat.level = metad->btm_level;
indexStat.fastroot = metad->btm_fastroot;
indexStat.fastlevel = metad->btm_fastlevel;
ReleaseBuffer(buffer);
}
nblocks = RelationGetNumberOfBlocks(rel);
/* -- init stat -- */
indexStat.fragments = 0;
indexStat.root_pages = 0;
indexStat.leaf_pages = 0;
indexStat.internal_pages = 0;
indexStat.empty_pages = 0;
indexStat.deleted_pages = 0;
indexStat.max_avail = 0;
indexStat.free_space = 0;
/*-----------------------
* Scan all blocks
*-----------------------
*/
for (blkno = 1; blkno < nblocks; blkno++)
{
Buffer buffer = ReadBuffer(rel, blkno);
BTPageStat stat;
CHECK_FOR_INTERRUPTS();
/* scan one page */
stat.blkno = blkno;
GetBTPageStatistics(blkno, buffer, &stat);
/*---------------------
* page status (type)
*---------------------
*/
switch (stat.type)
{
case 'd':
indexStat.deleted_pages++;
break;
case 'l':
indexStat.leaf_pages++;
break;
case 'i':
indexStat.internal_pages++;
break;
case 'e':
indexStat.empty_pages++;
break;
case 'r':
indexStat.root_pages++;
break;
default:
elog(ERROR, "unknown page status.");
}
/* -- leaf fragmentation -- */
indexStat.fragments += stat.fragments;
if (stat.type == 'l')
{
indexStat.max_avail += stat.max_avail;
indexStat.free_space += stat.free_size;
}
ReleaseBuffer(buffer);
}
relation_close(rel, AccessShareLock);
/*----------------------------
* Build a result tuple
*----------------------------
*/
{
TupleDesc tupleDesc;
int j;
char *values[PGSTATINDEX_NCOLUMNS];
HeapTuple tuple;
tupleDesc = RelationNameGetTupleDesc(PGSTATINDEX_TYPE);
j = 0;
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", indexStat.version);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", indexStat.level);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", (indexStat.root_pages +
indexStat.leaf_pages +
indexStat.internal_pages +
indexStat.deleted_pages +
indexStat.empty_pages) * BLCKSZ);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", indexStat.root_blkno);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", indexStat.internal_pages);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", indexStat.leaf_pages);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", indexStat.empty_pages);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", indexStat.deleted_pages);
values[j] = palloc(32);
if (indexStat.max_avail > 0)
snprintf(values[j++], 32, "%.2f",
100.0 - (double) indexStat.free_space / (double) indexStat.max_avail * 100.0);
else
snprintf(values[j++], 32, "NaN");
values[j] = palloc(32);
if (indexStat.leaf_pages > 0)
snprintf(values[j++], 32, "%.2f",
(double) indexStat.fragments / (double) indexStat.leaf_pages * 100.0);
else
snprintf(values[j++], 32, "NaN");
tuple = BuildTupleFromCStrings(TupleDescGetAttInMetadata(tupleDesc),
values);
result = TupleGetDatum(TupleDescGetSlot(tupleDesc), tuple);
}
PG_RETURN_DATUM(result);
}
Datum
readindex(PG_FUNCTION_ARGS)
{
FuncCallContext *funcctx;
readindexinfo *info;
MIRROREDLOCK_BUFMGR_DECLARE;
if (SRF_IS_FIRSTCALL())
{
Oid irelid = PG_GETARG_OID(0);
TupleDesc tupdesc;
MemoryContext oldcontext;
AttrNumber outattnum;
Relation irel;
TupleDesc itupdesc;
int i;
AttrNumber attno;
irel = index_open(irelid, AccessShareLock);
itupdesc = RelationGetDescr(irel);
outattnum = FIXED_COLUMN + itupdesc->natts;
funcctx = SRF_FIRSTCALL_INIT();
oldcontext = MemoryContextSwitchTo(funcctx->multi_call_memory_ctx);
tupdesc = CreateTemplateTupleDesc(outattnum, false);
attno = 1;
TupleDescInitEntry(tupdesc, attno++, "ictid", TIDOID, -1, 0);
TupleDescInitEntry(tupdesc, attno++, "hctid", TIDOID, -1, 0);
TupleDescInitEntry(tupdesc, attno++, "aotid", TEXTOID, -1, 0);
TupleDescInitEntry(tupdesc, attno++, "istatus", TEXTOID, -1, 0);
TupleDescInitEntry(tupdesc, attno++, "hstatus", TEXTOID, -1, 0);
for (i = 0; i < itupdesc->natts; i++)
{
Form_pg_attribute attr = itupdesc->attrs[i];
TupleDescInitEntry(tupdesc, attno++, NameStr(attr->attname), attr->atttypid, attr->atttypmod, 0);
}
funcctx->tuple_desc = BlessTupleDesc(tupdesc);
info = (readindexinfo *) palloc(sizeof(readindexinfo));
funcctx->user_fctx = (void *) info;
info->outattnum = outattnum;
info->irel = irel;
info->hrel = relation_open(irel->rd_index->indrelid, AccessShareLock);
if (info->hrel->rd_rel != NULL &&
(info->hrel->rd_rel->relstorage == 'a' ||
info->hrel->rd_rel->relstorage == 'c'))
{
relation_close(info->hrel, AccessShareLock);
info->hrel = NULL;
}
info->num_pages = RelationGetNumberOfBlocks(irel);
info->blkno = BTREE_METAPAGE + 1;
info->page = NULL;
MemoryContextSwitchTo(oldcontext);
}
funcctx = SRF_PERCALL_SETUP();
info = (readindexinfo *) funcctx->user_fctx;
while (info->blkno < info->num_pages)
{
Datum values[255];
bool nulls[255];
ItemPointerData itid;
HeapTuple tuple;
Datum result;
if (info->page == NULL)
{
MIRROREDLOCK_BUFMGR_LOCK;
info->buf = ReadBuffer(info->irel, info->blkno);
info->page = BufferGetPage(info->buf);
info->opaque = (BTPageOpaque) PageGetSpecialPointer(info->page);
info->minoff = P_FIRSTDATAKEY(info->opaque);
info->maxoff = PageGetMaxOffsetNumber(info->page);
info->offnum = info->minoff;
MIRROREDLOCK_BUFMGR_UNLOCK;
}
if (!P_ISLEAF(info->opaque) || info->offnum > info->maxoff)
{
ReleaseBuffer(info->buf);
info->page = NULL;
info->blkno++;
continue;
}
MemSet(nulls, false, info->outattnum * sizeof(bool));
ItemPointerSet(&itid, info->blkno, info->offnum);
values[0] = ItemPointerGetDatum(&itid);
readindextuple(info, values, nulls);
info->offnum = OffsetNumberNext(info->offnum);
tuple = heap_form_tuple(funcctx->tuple_desc, values, nulls);
result = HeapTupleGetDatum(tuple);
SRF_RETURN_NEXT(funcctx, result);
}
if (info->hrel != NULL)
relation_close(info->hrel, AccessShareLock);
index_close(info->irel, AccessShareLock);
SRF_RETURN_DONE(funcctx);
}
Datum
bt_page_items(PG_FUNCTION_ARGS)
{
text *relname = PG_GETARG_TEXT_P(0);
uint32 blkno = PG_GETARG_UINT32(1);
RangeVar *relrv;
Datum result;
char *values[BTPAGEITEMS_NCOLUMNS];
BTPageOpaque opaque;
HeapTuple tuple;
ItemId id;
FuncCallContext *fctx;
MemoryContext mctx;
struct user_args *uargs = NULL;
if (!superuser())
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
(errmsg("must be superuser to use pgstattuple functions"))));
if (SRF_IS_FIRSTCALL())
{
fctx = SRF_FIRSTCALL_INIT();
mctx = MemoryContextSwitchTo(fctx->multi_call_memory_ctx);
uargs = palloc(sizeof(struct user_args));
uargs->tupd = RelationNameGetTupleDesc(BTPAGEITEMS_TYPE);
uargs->offset = FirstOffsetNumber;
relrv = makeRangeVarFromNameList(textToQualifiedNameList(relname));
uargs->rel = relation_openrv(relrv, AccessShareLock);
if (!IS_INDEX(uargs->rel) || !IS_BTREE(uargs->rel))
elog(ERROR, "bt_page_items() can be used only on b-tree index.");
/*
* Reject attempts to read non-local temporary relations; we would
* be likely to get wrong data since we have no visibility into the
* owning session's local buffers.
*/
if (isOtherTempNamespace(RelationGetNamespace(uargs->rel)))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("cannot access temporary tables of other sessions")));
if (blkno == 0)
elog(ERROR, "Block 0 is a meta page.");
CHECK_RELATION_BLOCK_RANGE(uargs->rel, blkno);
uargs->buffer = ReadBuffer(uargs->rel, blkno);
uargs->page = BufferGetPage(uargs->buffer);
opaque = (BTPageOpaque) PageGetSpecialPointer(uargs->page);
if (P_ISDELETED(opaque))
elog(NOTICE, "bt_page_items(): this page is deleted.");
fctx->max_calls = PageGetMaxOffsetNumber(uargs->page);
fctx->user_fctx = uargs;
MemoryContextSwitchTo(mctx);
}
fctx = SRF_PERCALL_SETUP();
uargs = fctx->user_fctx;
if (fctx->call_cntr < fctx->max_calls)
{
IndexTuple itup;
id = PageGetItemId(uargs->page, uargs->offset);
if (!ItemIdIsValid(id))
elog(ERROR, "Invalid ItemId.");
itup = (IndexTuple) PageGetItem(uargs->page, id);
{
int j = 0;
BlockNumber blkno = BlockIdGetBlockNumber(&(itup->t_tid.ip_blkid));
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", uargs->offset);
values[j] = palloc(32);
snprintf(values[j++], 32, "(%u,%u)", blkno, itup->t_tid.ip_posid);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", (int) IndexTupleSize(itup));
values[j] = palloc(32);
snprintf(values[j++], 32, "%c", IndexTupleHasNulls(itup) ? 't' : 'f');
values[j] = palloc(32);
snprintf(values[j++], 32, "%c", IndexTupleHasVarwidths(itup) ? 't' : 'f');
{
int off;
char *dump;
char *ptr = (char *) itup + IndexInfoFindDataOffset(itup->t_info);
dump = palloc(IndexTupleSize(itup) * 3);
memset(dump, 0, IndexTupleSize(itup) * 3);
for (off = 0;
off < IndexTupleSize(itup) - IndexInfoFindDataOffset(itup->t_info);
off++)
{
if (dump[0] == '\0')
sprintf(dump, "%02x", *(ptr + off) & 0xff);
else
{
char buf[4];
sprintf(buf, " %02x", *(ptr + off) & 0xff);
strcat(dump, buf);
}
}
values[j] = dump;
}
tuple = BuildTupleFromCStrings(TupleDescGetAttInMetadata(uargs->tupd), values);
result = TupleGetDatum(TupleDescGetSlot(uargs->tupd), tuple);
}
uargs->offset = uargs->offset + 1;
SRF_RETURN_NEXT(fctx, result);
}
else
{
ReleaseBuffer(uargs->buffer);
relation_close(uargs->rel, AccessShareLock);
SRF_RETURN_DONE(fctx);
}
}
/*
* 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.
*
* 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 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;
bool fsm_vac = false;
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);
fsm_vac = false;
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 >= (Size)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);
fsm_vac = false;
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, fill_fsm, stats))
{
/* another cleanup process is running concurrently */
LockBuffer(metabuffer, GIN_UNLOCK);
fsm_vac = false;
break;
}
/* 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 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);
/*
* 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);
}
/*
* Recursive guts of FreeSpaceMapVacuum
*/
static uint8
fsm_vacuum_page(Relation rel, FSMAddress addr, bool *eof_p)
{
Buffer buf;
Page page;
uint8 max_avail;
/* Read the page if it exists, or return EOF */
buf = fsm_readbuf(rel, addr, false);
if (!BufferIsValid(buf))
{
*eof_p = true;
return 0;
}
else
*eof_p = false;
page = BufferGetPage(buf);
/*
* Recurse into children, and fix the information stored about them at
* this level.
*/
if (addr.level > FSM_BOTTOM_LEVEL)
{
int slot;
bool eof = false;
for (slot = 0; slot < SlotsPerFSMPage; slot++)
{
int child_avail;
CHECK_FOR_INTERRUPTS();
/* After we hit end-of-file, just clear the rest of the slots */
if (!eof)
child_avail = fsm_vacuum_page(rel, fsm_get_child(addr, slot), &eof);
else
child_avail = 0;
/* Update information about the child */
if (fsm_get_avail(page, slot) != child_avail)
{
LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
fsm_set_avail(BufferGetPage(buf), slot, child_avail);
MarkBufferDirtyHint(buf, false);
LockBuffer(buf, BUFFER_LOCK_UNLOCK);
}
}
}
max_avail = fsm_get_max_avail(BufferGetPage(buf));
/*
* Reset the next slot pointer. This encourages the use of low-numbered
* pages, increasing the chances that a later vacuum can truncate the
* relation.
*/
((FSMPage) PageGetContents(page))->fp_next_slot = 0;
ReleaseBuffer(buf);
return max_avail;
}
FarString& FarString::TrimLeft()
{
FarSF::LTrim (GetBuffer());
ReleaseBuffer();
return * this;
}
/* --------------------------------
* ExecStoreTuple
*
* This function is used to store a physical tuple into a specified
* slot in the tuple table.
*
* tuple: tuple to store
* slot: slot to store it in
* buffer: disk buffer if tuple is in a disk page, else InvalidBuffer
* shouldFree: true if ExecClearTuple should pfree() the tuple
* when done with it
*
* If 'buffer' is not InvalidBuffer, the tuple table code acquires a pin
* on the buffer which is held until the slot is cleared, so that the tuple
* won't go away on us.
*
* shouldFree is normally set 'true' for tuples constructed on-the-fly.
* It must always be 'false' for tuples that are stored in disk pages,
* since we don't want to try to pfree those.
*
* Another case where it is 'false' is when the referenced tuple is held
* in a tuple table slot belonging to a lower-level executor Proc node.
* In this case the lower-level slot retains ownership and responsibility
* for eventually releasing the tuple. When this method is used, we must
* be certain that the upper-level Proc node will lose interest in the tuple
* sooner than the lower-level one does! If you're not certain, copy the
* lower-level tuple with heap_copytuple and let the upper-level table
* slot assume ownership of the copy!
*
* Return value is just the passed-in slot pointer.
*
* NOTE: before PostgreSQL 8.1, this function would accept a NULL tuple
* pointer and effectively behave like ExecClearTuple (though you could
* still specify a buffer to pin, which would be an odd combination).
* This saved a couple lines of code in a few places, but seemed more likely
* to mask logic errors than to be really useful, so it's now disallowed.
* --------------------------------
*/
TupleTableSlot *
ExecStoreHeapTuple(HeapTuple tuple,
TupleTableSlot *slot,
Buffer buffer,
bool shouldFree)
{
/*
* sanity checks
*/
Assert(tuple != NULL);
Assert(slot != NULL);
Assert(slot->tts_tupleDescriptor != NULL);
/* passing shouldFree=true for a tuple on a disk page is not sane */
Assert(BufferIsValid(buffer) ? (!shouldFree) : true);
/*
* Actually we are storing a memtuple!
*/
if(is_heaptuple_memtuple(tuple))
{
Assert(buffer == InvalidBuffer);
return ExecStoreMemTuple((MemTuple) tuple, slot, shouldFree);
}
/*
* Free any old physical tuple belonging to the slot.
*/
free_heaptuple_memtuple(slot);
/*
* Store the new tuple into the specified slot.
*/
/* Clear tts_flags, here isempty set to false */
slot->PRIVATE_tts_flags = shouldFree ? TTS_SHOULDFREE : 0;
/* store the tuple */
slot->PRIVATE_tts_heaptuple = (void *) tuple;
/* Mark extracted state invalid */
slot->PRIVATE_tts_nvalid = 0;
/*
* If tuple is on a disk page, keep the page pinned as long as we hold a
* pointer into it. We assume the caller already has such a pin.
*
* This is coded to optimize the case where the slot previously held a
* tuple on the same disk page: in that case releasing and re-acquiring
* the pin is a waste of cycles. This is a common situation during
* seqscans, so it's worth troubling over.
*/
if (slot->tts_buffer != buffer)
{
if (BufferIsValid(slot->tts_buffer))
ReleaseBuffer(slot->tts_buffer);
slot->tts_buffer = buffer;
if (BufferIsValid(buffer))
IncrBufferRefCount(buffer);
}
return slot;
}
/* ------------------------------------------------------
* pgstatindex()
*
* Usage: SELECT * FROM pgstatindex('t1_pkey');
* ------------------------------------------------------
*/
Datum
pgstatindex(PG_FUNCTION_ARGS)
{
text *relname = PG_GETARG_TEXT_P(0);
Relation rel;
RangeVar *relrv;
Datum result;
BlockNumber nblocks;
BlockNumber blkno;
BTIndexStat indexStat;
if (!superuser())
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
(errmsg("must be superuser to use pgstattuple functions"))));
relrv = makeRangeVarFromNameList(textToQualifiedNameList(relname));
rel = relation_openrv(relrv, AccessShareLock);
if (!IS_INDEX(rel) || !IS_BTREE(rel))
elog(ERROR, "relation \"%s\" is not a btree index",
RelationGetRelationName(rel));
/*
* Reject attempts to read non-local temporary relations; we would be
* likely to get wrong data since we have no visibility into the owning
* session's local buffers.
*/
if (RELATION_IS_OTHER_TEMP(rel))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("cannot access temporary tables of other sessions")));
/*
* Read metapage
*/
{
Buffer buffer = ReadBuffer(rel, 0);
Page page = BufferGetPage(buffer);
BTMetaPageData *metad = BTPageGetMeta(page);
indexStat.version = metad->btm_version;
indexStat.level = metad->btm_level;
indexStat.root_blkno = metad->btm_root;
ReleaseBuffer(buffer);
}
/* -- init counters -- */
indexStat.root_pages = 0;
indexStat.internal_pages = 0;
indexStat.leaf_pages = 0;
indexStat.empty_pages = 0;
indexStat.deleted_pages = 0;
indexStat.max_avail = 0;
indexStat.free_space = 0;
indexStat.fragments = 0;
/*
* Scan all blocks except the metapage
*/
nblocks = RelationGetNumberOfBlocks(rel);
for (blkno = 1; blkno < nblocks; blkno++)
{
Buffer buffer;
Page page;
BTPageOpaque opaque;
/* Read and lock buffer */
buffer = ReadBuffer(rel, blkno);
LockBuffer(buffer, BUFFER_LOCK_SHARE);
page = BufferGetPage(buffer);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
/* Determine page type, and update totals */
if (P_ISLEAF(opaque))
{
int max_avail;
max_avail = BLCKSZ - (BLCKSZ - ((PageHeader) page)->pd_special + SizeOfPageHeaderData);
indexStat.max_avail += max_avail;
indexStat.free_space += PageGetFreeSpace(page);
indexStat.leaf_pages++;
/*
* If the next leaf is on an earlier block, it means a
* fragmentation.
*/
if (opaque->btpo_next != P_NONE && opaque->btpo_next < blkno)
indexStat.fragments++;
}
else if (P_ISDELETED(opaque))
indexStat.deleted_pages++;
else if (P_IGNORE(opaque))
indexStat.empty_pages++;
else if (P_ISROOT(opaque))
indexStat.root_pages++;
else
indexStat.internal_pages++;
/* Unlock and release buffer */
LockBuffer(buffer, BUFFER_LOCK_UNLOCK);
ReleaseBuffer(buffer);
}
relation_close(rel, AccessShareLock);
/*----------------------------
* Build a result tuple
*----------------------------
*/
{
TupleDesc tupleDesc;
int j;
char *values[10];
HeapTuple tuple;
/* Build a tuple descriptor for our result type */
if (get_call_result_type(fcinfo, NULL, &tupleDesc) != TYPEFUNC_COMPOSITE)
elog(ERROR, "return type must be a row type");
j = 0;
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", indexStat.version);
values[j] = palloc(32);
snprintf(values[j++], 32, "%d", indexStat.level);
values[j] = palloc(32);
snprintf(values[j++], 32, INT64_FORMAT,
(indexStat.root_pages +
indexStat.leaf_pages +
indexStat.internal_pages +
indexStat.deleted_pages +
indexStat.empty_pages) * BLCKSZ);
values[j] = palloc(32);
snprintf(values[j++], 32, "%u", indexStat.root_blkno);
values[j] = palloc(32);
snprintf(values[j++], 32, INT64_FORMAT, indexStat.internal_pages);
values[j] = palloc(32);
snprintf(values[j++], 32, INT64_FORMAT, indexStat.leaf_pages);
values[j] = palloc(32);
snprintf(values[j++], 32, INT64_FORMAT, indexStat.empty_pages);
values[j] = palloc(32);
snprintf(values[j++], 32, INT64_FORMAT, indexStat.deleted_pages);
values[j] = palloc(32);
snprintf(values[j++], 32, "%.2f", 100.0 - (double) indexStat.free_space / (double) indexStat.max_avail * 100.0);
values[j] = palloc(32);
snprintf(values[j++], 32, "%.2f", (double) indexStat.fragments / (double) indexStat.leaf_pages * 100.0);
tuple = BuildTupleFromCStrings(TupleDescGetAttInMetadata(tupleDesc),
values);
result = HeapTupleGetDatum(tuple);
}
PG_RETURN_DATUM(result);
}