rewriteheap.c

postgresql8.3.4源码,开源数据库

	new_tuple->t_data->t_infomask |=
		old_tuple->t_data->t_infomask & HEAP_XACT_MASK;

	/*
	 * While we have our hands on the tuple, we may as well freeze any
	 * very-old xmin or xmax, so that future VACUUM effort can be saved.
	 *
	 * Note we abuse heap_freeze_tuple() a bit here, since it's expecting to
	 * be given a pointer to a tuple in a disk buffer.	It happens though that
	 * we can get the right things to happen by passing InvalidBuffer for the
	 * buffer.
	 */
	heap_freeze_tuple(new_tuple->t_data, state->rs_freeze_xid, InvalidBuffer);

	/*
	 * Invalid ctid means that ctid should point to the tuple itself. We'll
	 * override it later if the tuple is part of an update chain.
	 */
	ItemPointerSetInvalid(&new_tuple->t_data->t_ctid);

	/*
	 * If the tuple has been updated, check the old-to-new mapping hash table.
	 */
	if (!(old_tuple->t_data->t_infomask & (HEAP_XMAX_INVALID |
										   HEAP_IS_LOCKED)) &&
		!(ItemPointerEquals(&(old_tuple->t_self),
							&(old_tuple->t_data->t_ctid))))
	{
		OldToNewMapping mapping;

		memset(&hashkey, 0, sizeof(hashkey));
		hashkey.xmin = HeapTupleHeaderGetXmax(old_tuple->t_data);
		hashkey.tid = old_tuple->t_data->t_ctid;

		mapping = (OldToNewMapping)
			hash_search(state->rs_old_new_tid_map, &hashkey,
						HASH_FIND, NULL);

		if (mapping != NULL)
		{
			/*
			 * We've already copied the tuple that t_ctid points to, so we can
			 * set the ctid of this tuple to point to the new location, and
			 * insert it right away.
			 */
			new_tuple->t_data->t_ctid = mapping->new_tid;

			/* We don't need the mapping entry anymore */
			hash_search(state->rs_old_new_tid_map, &hashkey,
						HASH_REMOVE, &found);
			Assert(found);
		}
		else
		{
			/*
			 * We haven't seen the tuple t_ctid points to yet. Stash this
			 * tuple into unresolved_tups to be written later.
			 */
			UnresolvedTup unresolved;

			unresolved = hash_search(state->rs_unresolved_tups, &hashkey,
									 HASH_ENTER, &found);
			Assert(!found);

			unresolved->old_tid = old_tuple->t_self;
			unresolved->tuple = heap_copytuple(new_tuple);

			/*
			 * We can't do anything more now, since we don't know where the
			 * tuple will be written.
			 */
			MemoryContextSwitchTo(old_cxt);
			return;
		}
	}

	/*
	 * Now we will write the tuple, and then check to see if it is the B tuple
	 * in any new or known pair.  When we resolve a known pair, we will be
	 * able to write that pair's A tuple, and then we have to check if it
	 * resolves some other pair.  Hence, we need a loop here.
	 */
	old_tid = old_tuple->t_self;
	free_new = false;

	for (;;)
	{
		ItemPointerData new_tid;

		/* Insert the tuple and find out where it's put in new_heap */
		raw_heap_insert(state, new_tuple);
		new_tid = new_tuple->t_self;

		/*
		 * If the tuple is the updated version of a row, and the prior version
		 * wouldn't be DEAD yet, then we need to either resolve the prior
		 * version (if it's waiting in rs_unresolved_tups), or make an entry
		 * in rs_old_new_tid_map (so we can resolve it when we do see it). The
		 * previous tuple's xmax would equal this one's xmin, so it's
		 * RECENTLY_DEAD if and only if the xmin is not before OldestXmin.
		 */
		if ((new_tuple->t_data->t_infomask & HEAP_UPDATED) &&
			!TransactionIdPrecedes(HeapTupleHeaderGetXmin(new_tuple->t_data),
								   state->rs_oldest_xmin))
		{
			/*
			 * Okay, this is B in an update pair.  See if we've seen A.
			 */
			UnresolvedTup unresolved;

			memset(&hashkey, 0, sizeof(hashkey));
			hashkey.xmin = HeapTupleHeaderGetXmin(new_tuple->t_data);
			hashkey.tid = old_tid;

			unresolved = hash_search(state->rs_unresolved_tups, &hashkey,
									 HASH_FIND, NULL);

			if (unresolved != NULL)
			{
				/*
				 * We have seen and memorized the previous tuple already. Now
				 * that we know where we inserted the tuple its t_ctid points
				 * to, fix its t_ctid and insert it to the new heap.
				 */
				if (free_new)
					heap_freetuple(new_tuple);
				new_tuple = unresolved->tuple;
				free_new = true;
				old_tid = unresolved->old_tid;
				new_tuple->t_data->t_ctid = new_tid;

				/*
				 * We don't need the hash entry anymore, but don't free its
				 * tuple just yet.
				 */
				hash_search(state->rs_unresolved_tups, &hashkey,
							HASH_REMOVE, &found);
				Assert(found);

				/* loop back to insert the previous tuple in the chain */
				continue;
			}
			else
			{
				/*
				 * Remember the new tid of this tuple. We'll use it to set the
				 * ctid when we find the previous tuple in the chain.
				 */
				OldToNewMapping mapping;

				mapping = hash_search(state->rs_old_new_tid_map, &hashkey,
									  HASH_ENTER, &found);
				Assert(!found);

				mapping->new_tid = new_tid;
			}
		}

		/* Done with this (chain of) tuples, for now */
		if (free_new)
			heap_freetuple(new_tuple);
		break;
	}

	MemoryContextSwitchTo(old_cxt);
}

/*
 * Register a dead tuple with an ongoing rewrite. Dead tuples are not
 * copied to the new table, but we still make note of them so that we
 * can release some resources earlier.
 */
void
rewrite_heap_dead_tuple(RewriteState state, HeapTuple old_tuple)
{
	/*
	 * If we have already seen an earlier tuple in the update chain that
	 * points to this tuple, let's forget about that earlier tuple. It's in
	 * fact dead as well, our simple xmax < OldestXmin test in
	 * HeapTupleSatisfiesVacuum just wasn't enough to detect it. It happens
	 * when xmin of a tuple is greater than xmax, which sounds
	 * counter-intuitive but is perfectly valid.
	 *
	 * We don't bother to try to detect the situation the other way round,
	 * when we encounter the dead tuple first and then the recently dead one
	 * that points to it. If that happens, we'll have some unmatched entries
	 * in the UnresolvedTups hash table at the end. That can happen anyway,
	 * because a vacuum might have removed the dead tuple in the chain before
	 * us.
	 */
	UnresolvedTup unresolved;
	TidHashKey	hashkey;
	bool		found;

	memset(&hashkey, 0, sizeof(hashkey));
	hashkey.xmin = HeapTupleHeaderGetXmin(old_tuple->t_data);
	hashkey.tid = old_tuple->t_self;

	unresolved = hash_search(state->rs_unresolved_tups, &hashkey,
							 HASH_FIND, NULL);

	if (unresolved != NULL)
	{
		/* Need to free the contained tuple as well as the hashtable entry */
		heap_freetuple(unresolved->tuple);
		hash_search(state->rs_unresolved_tups, &hashkey,
					HASH_REMOVE, &found);
		Assert(found);
	}
}

/*
 * Insert a tuple to the new relation.	This has to track heap_insert
 * and its subsidiary functions!
 *
 * t_self of the tuple is set to the new TID of the tuple. If t_ctid of the
 * tuple is invalid on entry, it's replaced with the new TID as well (in
 * the inserted data only, not in the caller's copy).
 */
static void
raw_heap_insert(RewriteState state, HeapTuple tup)
{
	Page		page = state->rs_buffer;
	Size		pageFreeSpace,
				saveFreeSpace;
	Size		len;
	OffsetNumber newoff;
	HeapTuple	heaptup;

	/*
	 * If the new tuple is too big for storage or contains already toasted
	 * out-of-line attributes from some other relation, invoke the toaster.
	 *
	 * Note: below this point, heaptup is the data we actually intend to store
	 * into the relation; tup is the caller's original untoasted data.
	 */
	if (state->rs_new_rel->rd_rel->relkind == RELKIND_TOASTVALUE)
	{
		/* toast table entries should never be recursively toasted */
		Assert(!HeapTupleHasExternal(tup));
		heaptup = tup;
	}
	else if (HeapTupleHasExternal(tup) || tup->t_len > TOAST_TUPLE_THRESHOLD)
		heaptup = toast_insert_or_update(state->rs_new_rel, tup, NULL,
										 state->rs_use_wal, false);
	else
		heaptup = tup;

	len = MAXALIGN(heaptup->t_len);		/* be conservative */

	/*
	 * 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(state->rs_new_rel,
												   HEAP_DEFAULT_FILLFACTOR);

	/* Now we can check to see if there's enough free space already. */
	if (state->rs_buffer_valid)
	{
		pageFreeSpace = PageGetHeapFreeSpace(page);

		if (len + saveFreeSpace > pageFreeSpace)
		{
			/* Doesn't fit, so write out the existing page */

			/* XLOG stuff */
			if (state->rs_use_wal)
				log_newpage(&state->rs_new_rel->rd_node,
							state->rs_blockno,
							page);

			/*
			 * Now write the page. We say isTemp = true even if it's not a
			 * temp table, because there's no need for smgr to schedule an
			 * fsync for this write; we'll do it ourselves in
			 * end_heap_rewrite.
			 */
			RelationOpenSmgr(state->rs_new_rel);
			smgrextend(state->rs_new_rel->rd_smgr, state->rs_blockno,
					   (char *) page, true);

			state->rs_blockno++;
			state->rs_buffer_valid = false;
		}
	}

	if (!state->rs_buffer_valid)
	{
		/* Initialize a new empty page */
		PageInit(page, BLCKSZ, 0);
		state->rs_buffer_valid = true;
	}

	/* And now we can insert the tuple into the page */
	newoff = PageAddItem(page, (Item) heaptup->t_data, len,
						 InvalidOffsetNumber, false, true);
	if (newoff == InvalidOffsetNumber)
		elog(ERROR, "failed to add tuple");

	/* Update caller's t_self to the actual position where it was stored */
	ItemPointerSet(&(tup->t_self), state->rs_blockno, newoff);

	/*
	 * Insert the correct position into CTID of the stored tuple, too, if the
	 * caller didn't supply a valid CTID.
	 */
	if (!ItemPointerIsValid(&tup->t_data->t_ctid))
	{
		ItemId		newitemid;
		HeapTupleHeader onpage_tup;

		newitemid = PageGetItemId(page, newoff);
		onpage_tup = (HeapTupleHeader) PageGetItem(page, newitemid);

		onpage_tup->t_ctid = tup->t_self;
	}

	/* If heaptup is a private copy, release it. */
	if (heaptup != tup)
		heap_freetuple(heaptup);
}