/* * _bt_moveright() -- move right in the btree if necessary. * * When we follow a pointer to reach a page, it is possible that * the page has changed in the meanwhile. If this happens, we're * guaranteed that the page has "split right" -- that is, that any * data that appeared on the page originally is either on the page * or strictly to the right of it. * * This routine decides whether or not we need to move right in the * tree by examining the high key entry on the page. If that entry * is strictly less than the scankey, or <= the scankey in the nextkey=true * case, then we followed the wrong link and we need to move right. * * The passed scankey must be an insertion-type scankey (see nbtree/README), * but it can omit the rightmost column(s) of the index. * * When nextkey is false (the usual case), we are looking for the first * item >= scankey. When nextkey is true, we are looking for the first * item strictly greater than scankey. * * On entry, we have the buffer pinned and a lock of the type specified by * 'access'. If we move right, we release the buffer and lock and acquire * the same on the right sibling. Return value is the buffer we stop at. */ Buffer _bt_moveright(Relation rel, Buffer buf, int keysz, ScanKey scankey, bool nextkey, int access) { Page page; BTPageOpaque opaque; int32 cmpval; MIRROREDLOCK_BUFMGR_MUST_ALREADY_BE_HELD; page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); /* * When nextkey = false (normal case): if the scan key that brought us to * this page is > the high key stored on the page, then the page has split * and we need to move right. (If the scan key is equal to the high key, * we might or might not need to move right; have to scan the page first * anyway.) * * When nextkey = true: move right if the scan key is >= page's high key. * * The page could even have split more than once, so scan as far as * needed. * * We also have to move right if we followed a link that brought us to a * dead page. */ cmpval = nextkey ? 0 : 1; while (!P_RIGHTMOST(opaque) && (P_IGNORE(opaque) || _bt_compare(rel, keysz, scankey, page, P_HIKEY) >= cmpval)) { /* step right one page */ BlockNumber rblkno = opaque->btpo_next; buf = _bt_relandgetbuf(rel, buf, rblkno, access); page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); } if (P_IGNORE(opaque)) elog(ERROR, "fell off the end of index \"%s\"", RelationGetRelationName(rel)); return buf; }
/** * @brief Get the next smaller item from the old index * * Process flow * -# Examine the max offset position in the page * -# Search the next item * -# If the item has deleted flag, seearch the next one * -# If we can't find items any more, read the leaf page on the right side * and search the next again * * These members are updated: * - page : page which includes picked-up item * - offnum : item offset number of the picked-up item * * @param reader [in/out] BTReader structure * @return next index tuple, or null if no more tuples */ static IndexTuple BTReaderGetNextItem(BTReader *reader) { OffsetNumber maxoff; ItemId itemid; BTPageOpaque opaque; /* * If any leaf page isn't read, the state is treated like as EOF */ if (reader->blkno == InvalidBlockNumber) return NULL; maxoff = PageGetMaxOffsetNumber(reader->page); for (;;) { /* * If no one items are picked up, offnum is set to InvalidOffsetNumber. */ if (reader->offnum == InvalidOffsetNumber) { opaque = (BTPageOpaque) PageGetSpecialPointer(reader->page); reader->offnum = P_FIRSTDATAKEY(opaque); } else reader->offnum = OffsetNumberNext(reader->offnum); if (reader->offnum <= maxoff) { itemid = PageGetItemId(reader->page, reader->offnum); /* Ignore dead items */ if (ItemIdIsDead(itemid)) continue; return (IndexTuple) PageGetItem(reader->page, itemid); } else { /* The end of the leaf page. Go right. */ opaque = (BTPageOpaque) PageGetSpecialPointer(reader->page); if (P_RIGHTMOST(opaque)) return NULL; /* No more index tuples */ BTReaderReadPage(reader, opaque->btpo_next); maxoff = PageGetMaxOffsetNumber(reader->page); } } }
/* * _bt_moveright() -- move right in the btree if necessary. * * When we follow a pointer to reach a page, it is possible that * the page has changed in the meanwhile. If this happens, we're * guaranteed that the page has "split right" -- that is, that any * data that appeared on the page originally is either on the page * or strictly to the right of it. * * This routine decides whether or not we need to move right in the * tree by examining the high key entry on the page. If that entry * is strictly less than one we expect to be on the page, then our * picture of the page is incorrect and we need to move right. * * On entry, we have the buffer pinned and a lock of the proper type. * If we move right, we release the buffer and lock and acquire the * same on the right sibling. Return value is the buffer we stop at. */ Buffer _bt_moveright(Relation rel, Buffer buf, int keysz, ScanKey scankey, int access) { Page page; BTPageOpaque opaque; page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); /* * If the scan key that brought us to this page is > the high key * stored on the page, then the page has split and we need to move * right. (If the scan key is equal to the high key, we might or * might not need to move right; have to scan the page first anyway.) * It could even have split more than once, so scan as far as needed. * * We also have to move right if we followed a link that brought us to a * dead page. */ while (!P_RIGHTMOST(opaque) && (P_IGNORE(opaque) || _bt_compare(rel, keysz, scankey, page, P_HIKEY) > 0)) { /* step right one page */ BlockNumber rblkno = opaque->btpo_next; _bt_relbuf(rel, buf); buf = _bt_getbuf(rel, rblkno, access); page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); } if (P_IGNORE(opaque)) elog(ERROR, "fell off the end of \"%s\"", RelationGetRelationName(rel)); return buf; }
/* * Subroutine to pre-check whether a page deletion is safe, that is, its * parent page would be left in a valid or deletable state. * * "target" is the page we wish to delete, and "stack" is a search stack * leading to it (approximately). Note that we will update the stack * entry(s) to reflect current downlink positions --- this is harmless and * indeed saves later search effort in _bt_pagedel. * * Note: it's OK to release page locks after checking, because a safe * deletion can't become unsafe due to concurrent activity. A non-rightmost * page cannot become rightmost unless there's a concurrent page deletion, * but only VACUUM does page deletion and we only allow one VACUUM on an index * at a time. An only child could acquire a sibling (of the same parent) only * by being split ... but that would make it a non-rightmost child so the * deletion is still safe. */ static bool _bt_parent_deletion_safe(Relation rel, BlockNumber target, BTStack stack) { BlockNumber parent; OffsetNumber poffset, maxoff; Buffer pbuf; Page page; BTPageOpaque opaque; MIRROREDLOCK_BUFMGR_MUST_ALREADY_BE_HELD; /* * In recovery mode, assume the deletion being replayed is valid. We * can't always check it because we won't have a full search stack, and we * should complain if there's a problem, anyway. */ if (InRecovery) return true; /* Locate the parent's downlink (updating the stack entry if needed) */ ItemPointerSet(&(stack->bts_btentry.t_tid), target, P_HIKEY); pbuf = _bt_getstackbuf(rel, stack, BT_READ); if (pbuf == InvalidBuffer) elog(ERROR, "failed to re-find parent key in index \"%s\" for deletion target page %u", RelationGetRelationName(rel), target); parent = stack->bts_blkno; poffset = stack->bts_offset; page = BufferGetPage(pbuf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); maxoff = PageGetMaxOffsetNumber(page); /* * If the target is the rightmost child of its parent, then we can't * delete, unless it's also the only child. */ if (poffset >= maxoff) { /* It's rightmost child... */ if (poffset == P_FIRSTDATAKEY(opaque)) { /* * It's only child, so safe if parent would itself be removable. * We have to check the parent itself, and then recurse to test * the conditions at the parent's parent. */ if (P_RIGHTMOST(opaque) || P_ISROOT(opaque)) { _bt_relbuf(rel, pbuf); return false; } _bt_relbuf(rel, pbuf); return _bt_parent_deletion_safe(rel, parent, stack->bts_parent); } else { /* Unsafe to delete */ _bt_relbuf(rel, pbuf); return false; } } else { /* Not rightmost child, so safe to delete */ _bt_relbuf(rel, pbuf); return true; } }
/* * _bt_gettrueroot() -- Get the true root page of the btree. * * This is the same as the BT_READ case of _bt_getroot(), except * we follow the true-root link not the fast-root link. * * By the time we acquire lock on the root page, it might have been split and * not be the true root anymore. This is okay for the present uses of this * routine; we only really need to be able to move up at least one tree level * from whatever non-root page we were at. If we ever do need to lock the * one true root page, we could loop here, re-reading the metapage on each * failure. (Note that it wouldn't do to hold the lock on the metapage while * moving to the root --- that'd deadlock against any concurrent root split.) */ Buffer _bt_gettrueroot(Relation rel) { Buffer metabuf; Page metapg; BTPageOpaque metaopaque; Buffer rootbuf; Page rootpage; BTPageOpaque rootopaque; BlockNumber rootblkno; uint32 rootlevel; BTMetaPageData *metad; MIRROREDLOCK_BUFMGR_MUST_ALREADY_BE_HELD; /* * We don't try to use cached metapage data here, since (a) this path is * not performance-critical, and (b) if we are here it suggests our cache * is out-of-date anyway. In light of point (b), it's probably safest to * actively flush any cached metapage info. */ if (rel->rd_amcache) pfree(rel->rd_amcache); rel->rd_amcache = NULL; metabuf = _bt_getbuf(rel, BTREE_METAPAGE, BT_READ); metapg = BufferGetPage(metabuf); metaopaque = (BTPageOpaque) PageGetSpecialPointer(metapg); metad = BTPageGetMeta(metapg); if (!(metaopaque->btpo_flags & BTP_META) || metad->btm_magic != BTREE_MAGIC) ereport(ERROR, (errcode(ERRCODE_INDEX_CORRUPTED), errmsg("index \"%s\" is not a btree", RelationGetRelationName(rel)))); if (metad->btm_version != BTREE_VERSION) ereport(ERROR, (errcode(ERRCODE_INDEX_CORRUPTED), errmsg("version mismatch in index \"%s\": file version %d, code version %d", RelationGetRelationName(rel), metad->btm_version, BTREE_VERSION))); /* if no root page initialized yet, fail */ if (metad->btm_root == P_NONE) { _bt_relbuf(rel, metabuf); return InvalidBuffer; } rootblkno = metad->btm_root; rootlevel = metad->btm_level; /* * We are done with the metapage; arrange to release it via first * _bt_relandgetbuf call */ rootbuf = metabuf; for (;;) { rootbuf = _bt_relandgetbuf(rel, rootbuf, rootblkno, BT_READ); rootpage = BufferGetPage(rootbuf); rootopaque = (BTPageOpaque) PageGetSpecialPointer(rootpage); if (!P_IGNORE(rootopaque)) break; /* it's dead, Jim. step right one page */ if (P_RIGHTMOST(rootopaque)) elog(ERROR, "no live root page found in index \"%s\"", RelationGetRelationName(rel)); rootblkno = rootopaque->btpo_next; } /* Note: can't check btpo.level on deleted pages */ if (rootopaque->btpo.level != rootlevel) elog(ERROR, "root page %u of index \"%s\" has level %u, expected %u", rootblkno, RelationGetRelationName(rel), rootopaque->btpo.level, rootlevel); return rootbuf; }
/* * _bt_moveright() -- move right in the btree if necessary. * * When we follow a pointer to reach a page, it is possible that * the page has changed in the meanwhile. If this happens, we're * guaranteed that the page has "split right" -- that is, that any * data that appeared on the page originally is either on the page * or strictly to the right of it. * * This routine decides whether or not we need to move right in the * tree by examining the high key entry on the page. If that entry * is strictly less than the scankey, or <= the scankey in the nextkey=true * case, then we followed the wrong link and we need to move right. * * The passed scankey must be an insertion-type scankey (see nbtree/README), * but it can omit the rightmost column(s) of the index. * * When nextkey is false (the usual case), we are looking for the first * item >= scankey. When nextkey is true, we are looking for the first * item strictly greater than scankey. * * If forupdate is true, we will attempt to finish any incomplete splits * that we encounter. This is required when locking a target page for an * insertion, because we don't allow inserting on a page before the split * is completed. 'stack' is only used if forupdate is true. * * On entry, we have the buffer pinned and a lock of the type specified by * 'access'. If we move right, we release the buffer and lock and acquire * the same on the right sibling. Return value is the buffer we stop at. */ Buffer _bt_moveright(Relation rel, Buffer buf, int keysz, ScanKey scankey, bool nextkey, bool forupdate, BTStack stack, int access) { Page page; BTPageOpaque opaque; int32 cmpval; /* * When nextkey = false (normal case): if the scan key that brought us to * this page is > the high key stored on the page, then the page has split * and we need to move right. (If the scan key is equal to the high key, * we might or might not need to move right; have to scan the page first * anyway.) * * When nextkey = true: move right if the scan key is >= page's high key. * * The page could even have split more than once, so scan as far as * needed. * * We also have to move right if we followed a link that brought us to a * dead page. */ cmpval = nextkey ? 0 : 1; for (;;) { page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); if (P_RIGHTMOST(opaque)) break; /* * Finish any incomplete splits we encounter along the way. */ if (forupdate && P_INCOMPLETE_SPLIT(opaque)) { BlockNumber blkno = BufferGetBlockNumber(buf); /* upgrade our lock if necessary */ if (access == BT_READ) { LockBuffer(buf, BUFFER_LOCK_UNLOCK); LockBuffer(buf, BT_WRITE); } if (P_INCOMPLETE_SPLIT(opaque)) _bt_finish_split(rel, buf, stack); else _bt_relbuf(rel, buf); /* re-acquire the lock in the right mode, and re-check */ buf = _bt_getbuf(rel, blkno, access); continue; } if (P_IGNORE(opaque) || _bt_compare(rel, keysz, scankey, page, P_HIKEY) >= cmpval) { /* step right one page */ buf = _bt_relandgetbuf(rel, buf, opaque->btpo_next, access); continue; } else break; } if (P_IGNORE(opaque)) elog(ERROR, "fell off the end of index \"%s\"", RelationGetRelationName(rel)); return buf; }
/* * _bt_endpoint() -- Find the first or last page in the index, and scan * from there to the first key satisfying all the quals. * * This is used by _bt_first() to set up a scan when we've determined * that the scan must start at the beginning or end of the index (for * a forward or backward scan respectively). Exit conditions are the * same as for _bt_first(). */ static bool _bt_endpoint(IndexScanDesc scan, ScanDirection dir) { Relation rel = scan->indexRelation; BTScanOpaque so = (BTScanOpaque) scan->opaque; Buffer buf; Page page; BTPageOpaque opaque; OffsetNumber start; BTScanPosItem *currItem; /* * Scan down to the leftmost or rightmost leaf page. This is a simplified * version of _bt_search(). We don't maintain a stack since we know we * won't need it. */ buf = _bt_get_endpoint(rel, 0, ScanDirectionIsBackward(dir)); if (!BufferIsValid(buf)) { /* * Empty index. Lock the whole relation, as nothing finer to lock * exists. */ PredicateLockRelation(rel, scan->xs_snapshot); so->currPos.buf = InvalidBuffer; return false; } PredicateLockPage(rel, BufferGetBlockNumber(buf), scan->xs_snapshot); page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); Assert(P_ISLEAF(opaque)); if (ScanDirectionIsForward(dir)) { /* There could be dead pages to the left, so not this: */ /* Assert(P_LEFTMOST(opaque)); */ start = P_FIRSTDATAKEY(opaque); } else if (ScanDirectionIsBackward(dir)) { Assert(P_RIGHTMOST(opaque)); start = PageGetMaxOffsetNumber(page); } else { elog(ERROR, "invalid scan direction: %d", (int) dir); start = 0; /* keep compiler quiet */ } /* remember which buffer we have pinned */ so->currPos.buf = buf; /* initialize moreLeft/moreRight appropriately for scan direction */ if (ScanDirectionIsForward(dir)) { so->currPos.moreLeft = false; so->currPos.moreRight = true; } else { so->currPos.moreLeft = true; so->currPos.moreRight = false; } so->numKilled = 0; /* just paranoia */ so->markItemIndex = -1; /* ditto */ /* * Now load data from the first page of the scan. */ if (!_bt_readpage(scan, dir, start)) { /* * There's no actually-matching data on this page. Try to advance to * the next page. Return false if there's no matching data at all. */ if (!_bt_steppage(scan, dir)) return false; } /* Drop the lock, but not pin, on the current page */ LockBuffer(so->currPos.buf, BUFFER_LOCK_UNLOCK); /* OK, itemIndex says what to return */ currItem = &so->currPos.items[so->currPos.itemIndex]; scan->xs_ctup.t_self = currItem->heapTid; if (scan->xs_want_itup) scan->xs_itup = (IndexTuple) (so->currTuples + currItem->tupleOffset); return true; }
/* * _bt_get_endpoint() -- Find the first or last page on a given tree level * * If the index is empty, we will return InvalidBuffer; any other failure * condition causes ereport(). We will not return a dead page. * * The returned buffer is pinned and read-locked. */ Buffer _bt_get_endpoint(Relation rel, uint32 level, bool rightmost) { Buffer buf; Page page; BTPageOpaque opaque; OffsetNumber offnum; BlockNumber blkno; IndexTuple itup; /* * If we are looking for a leaf page, okay to descend from fast root; * otherwise better descend from true root. (There is no point in being * smarter about intermediate levels.) */ if (level == 0) buf = _bt_getroot(rel, BT_READ); else buf = _bt_gettrueroot(rel); if (!BufferIsValid(buf)) return InvalidBuffer; page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); for (;;) { /* * If we landed on a deleted page, step right to find a live page * (there must be one). Also, if we want the rightmost page, step * right if needed to get to it (this could happen if the page split * since we obtained a pointer to it). */ while (P_IGNORE(opaque) || (rightmost && !P_RIGHTMOST(opaque))) { blkno = opaque->btpo_next; if (blkno == P_NONE) elog(ERROR, "fell off the end of index \"%s\"", RelationGetRelationName(rel)); buf = _bt_relandgetbuf(rel, buf, blkno, BT_READ); page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); } /* Done? */ if (opaque->btpo.level == level) break; if (opaque->btpo.level < level) elog(ERROR, "btree level %u not found in index \"%s\"", level, RelationGetRelationName(rel)); /* Descend to leftmost or rightmost child page */ if (rightmost) offnum = PageGetMaxOffsetNumber(page); else offnum = P_FIRSTDATAKEY(opaque); itup = (IndexTuple) PageGetItem(page, PageGetItemId(page, offnum)); blkno = ItemPointerGetBlockNumber(&(itup->t_tid)); buf = _bt_relandgetbuf(rel, buf, blkno, BT_READ); page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); } return buf; }
/* * _bt_getroot() -- Get the root page of the btree. * * Since the root page can move around the btree file, we have to read * its location from the metadata page, and then read the root page * itself. If no root page exists yet, we have to create one. The * standard class of race conditions exists here; I think I covered * them all in the Hopi Indian rain dance of lock requests below. * * The access type parameter (BT_READ or BT_WRITE) controls whether * a new root page will be created or not. If access = BT_READ, * and no root page exists, we just return InvalidBuffer. For * BT_WRITE, we try to create the root page if it doesn't exist. * NOTE that the returned root page will have only a read lock set * on it even if access = BT_WRITE! * * The returned page is not necessarily the true root --- it could be * a "fast root" (a page that is alone in its level due to deletions). * Also, if the root page is split while we are "in flight" to it, * what we will return is the old root, which is now just the leftmost * page on a probably-not-very-wide level. For most purposes this is * as good as or better than the true root, so we do not bother to * insist on finding the true root. We do, however, guarantee to * return a live (not deleted or half-dead) page. * * On successful return, the root page is pinned and read-locked. * The metadata page is not locked or pinned on exit. */ Buffer _bt_getroot(Relation rel, int access) { Buffer metabuf; Page metapg; BTPageOpaque metaopaque; Buffer rootbuf; Page rootpage; BTPageOpaque rootopaque; BlockNumber rootblkno; uint32 rootlevel; BTMetaPageData *metad; metabuf = _bt_getbuf(rel, BTREE_METAPAGE, BT_READ); metapg = BufferGetPage(metabuf); metaopaque = (BTPageOpaque) PageGetSpecialPointer(metapg); metad = BTPageGetMeta(metapg); /* sanity-check the metapage */ if (!(metaopaque->btpo_flags & BTP_META) || metad->btm_magic != BTREE_MAGIC) ereport(ERROR, (errcode(ERRCODE_INDEX_CORRUPTED), errmsg("index \"%s\" is not a btree", RelationGetRelationName(rel)))); if (metad->btm_version != BTREE_VERSION) ereport(ERROR, (errcode(ERRCODE_INDEX_CORRUPTED), errmsg("version mismatch in index \"%s\": file version %d, code version %d", RelationGetRelationName(rel), metad->btm_version, BTREE_VERSION))); /* if no root page initialized yet, do it */ if (metad->btm_root == P_NONE) { /* If access = BT_READ, caller doesn't want us to create root yet */ if (access == BT_READ) { _bt_relbuf(rel, metabuf); return InvalidBuffer; } /* trade in our read lock for a write lock */ LockBuffer(metabuf, BUFFER_LOCK_UNLOCK); LockBuffer(metabuf, BT_WRITE); /* * Race condition: if someone else initialized the metadata between * the time we released the read lock and acquired the write lock, we * must avoid doing it again. */ if (metad->btm_root != P_NONE) { /* * Metadata initialized by someone else. In order to guarantee no * deadlocks, we have to release the metadata page and start all * over again. (Is that really true? But it's hardly worth trying * to optimize this case.) */ _bt_relbuf(rel, metabuf); return _bt_getroot(rel, access); } /* * Get, initialize, write, and leave a lock of the appropriate type on * the new root page. Since this is the first page in the tree, it's * a leaf as well as the root. */ rootbuf = _bt_getbuf(rel, P_NEW, BT_WRITE); rootblkno = BufferGetBlockNumber(rootbuf); rootpage = BufferGetPage(rootbuf); _bt_pageinit(rootpage, BufferGetPageSize(rootbuf)); rootopaque = (BTPageOpaque) PageGetSpecialPointer(rootpage); rootopaque->btpo_prev = rootopaque->btpo_next = P_NONE; rootopaque->btpo_flags = (BTP_LEAF | BTP_ROOT); rootopaque->btpo.level = 0; /* NO ELOG(ERROR) till meta is updated */ START_CRIT_SECTION(); metad->btm_root = rootblkno; metad->btm_level = 0; metad->btm_fastroot = rootblkno; metad->btm_fastlevel = 0; /* XLOG stuff */ if (!rel->rd_istemp) { xl_btree_newroot xlrec; XLogRecPtr recptr; XLogRecData rdata; xlrec.node = rel->rd_node; xlrec.rootblk = rootblkno; xlrec.level = 0; rdata.data = (char *) &xlrec; rdata.len = SizeOfBtreeNewroot; rdata.buffer = InvalidBuffer; rdata.next = NULL; recptr = XLogInsert(RM_BTREE_ID, XLOG_BTREE_NEWROOT, &rdata); PageSetLSN(rootpage, recptr); PageSetTLI(rootpage, ThisTimeLineID); PageSetLSN(metapg, recptr); PageSetTLI(metapg, ThisTimeLineID); } END_CRIT_SECTION(); _bt_wrtnorelbuf(rel, rootbuf); /* * swap root write lock for read lock. There is no danger of anyone * else accessing the new root page while it's unlocked, since no one * else knows where it is yet. */ LockBuffer(rootbuf, BUFFER_LOCK_UNLOCK); LockBuffer(rootbuf, BT_READ); /* okay, metadata is correct, write and release it */ _bt_wrtbuf(rel, metabuf); } else { rootblkno = metad->btm_fastroot; Assert(rootblkno != P_NONE); rootlevel = metad->btm_fastlevel; /* * We are done with the metapage; arrange to release it via first * _bt_relandgetbuf call */ rootbuf = metabuf; for (;;) { rootbuf = _bt_relandgetbuf(rel, rootbuf, rootblkno, BT_READ); rootpage = BufferGetPage(rootbuf); rootopaque = (BTPageOpaque) PageGetSpecialPointer(rootpage); if (!P_IGNORE(rootopaque)) break; /* it's dead, Jim. step right one page */ if (P_RIGHTMOST(rootopaque)) elog(ERROR, "no live root page found in \"%s\"", RelationGetRelationName(rel)); rootblkno = rootopaque->btpo_next; } /* Note: can't check btpo.level on deleted pages */ if (rootopaque->btpo.level != rootlevel) elog(ERROR, "root page %u of \"%s\" has level %u, expected %u", rootblkno, RelationGetRelationName(rel), rootopaque->btpo.level, rootlevel); } /* * By here, we have a pin and read lock on the root page, and no lock set * on the metadata page. Return the root page's buffer. */ return rootbuf; }
/* * btvacuumpage --- VACUUM one page * * This processes a single page for btvacuumscan(). In some cases we * must go back and re-examine previously-scanned pages; this routine * recurses when necessary to handle that case. * * blkno is the page to process. orig_blkno is the highest block number * reached by the outer btvacuumscan loop (the same as blkno, unless we * are recursing to re-examine a previous page). */ static void btvacuumpage(BTVacState *vstate, BlockNumber blkno, BlockNumber orig_blkno) { MIRROREDLOCK_BUFMGR_DECLARE; IndexVacuumInfo *info = vstate->info; IndexBulkDeleteResult *stats = vstate->stats; IndexBulkDeleteCallback callback = vstate->callback; void *callback_state = vstate->callback_state; Relation rel = info->index; bool delete_now; BlockNumber recurse_to; Buffer buf; Page page; BTPageOpaque opaque; restart: delete_now = false; recurse_to = P_NONE; /* call vacuum_delay_point while not holding any buffer lock */ vacuum_delay_point(); /* * We can't use _bt_getbuf() here because it always applies * _bt_checkpage(), which will barf on an all-zero page. We want to * recycle all-zero pages, not fail. Also, we want to use a nondefault * buffer access strategy. */ // -------- MirroredLock ---------- MIRROREDLOCK_BUFMGR_LOCK; buf = ReadBufferWithStrategy(rel, blkno, info->strategy); LockBuffer(buf, BT_READ); page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); if (!PageIsNew(page)) _bt_checkpage(rel, buf); /* * If we are recursing, the only case we want to do anything with is a * live leaf page having the current vacuum cycle ID. Any other state * implies we already saw the page (eg, deleted it as being empty). In * particular, we don't want to risk adding it to freePages twice. */ if (blkno != orig_blkno) { if (_bt_page_recyclable(page) || P_IGNORE(opaque) || !P_ISLEAF(opaque) || opaque->btpo_cycleid != vstate->cycleid) { _bt_relbuf(rel, buf); MIRROREDLOCK_BUFMGR_UNLOCK; // -------- MirroredLock ---------- return; } } /* Page is valid, see what to do with it */ if (_bt_page_recyclable(page)) { /* Okay to recycle this page */ if (vstate->nFreePages < vstate->maxFreePages) vstate->freePages[vstate->nFreePages++] = blkno; vstate->totFreePages++; stats->pages_deleted++; } else if (P_ISDELETED(opaque)) { /* Already deleted, but can't recycle yet */ stats->pages_deleted++; } else if (P_ISHALFDEAD(opaque)) { /* Half-dead, try to delete */ delete_now = true; } else if (P_ISLEAF(opaque)) { OffsetNumber deletable[MaxOffsetNumber]; int ndeletable; OffsetNumber offnum, minoff, maxoff; /* * Trade in the initial read lock for a super-exclusive write lock on * this page. We must get such a lock on every leaf page over the * course of the vacuum scan, whether or not it actually contains any * deletable tuples --- see nbtree/README. */ LockBuffer(buf, BUFFER_LOCK_UNLOCK); LockBufferForCleanup(buf); /* * Check whether we need to recurse back to earlier pages. What we * are concerned about is a page split that happened since we started * the vacuum scan. If the split moved some tuples to a lower page * then we might have missed 'em. If so, set up for tail recursion. * (Must do this before possibly clearing btpo_cycleid below!) */ if (vstate->cycleid != 0 && opaque->btpo_cycleid == vstate->cycleid && !(opaque->btpo_flags & BTP_SPLIT_END) && !P_RIGHTMOST(opaque) && opaque->btpo_next < orig_blkno) recurse_to = opaque->btpo_next; /* * Scan over all items to see which ones need deleted according to the * callback function. */ ndeletable = 0; minoff = P_FIRSTDATAKEY(opaque); maxoff = PageGetMaxOffsetNumber(page); if (callback) { for (offnum = minoff; offnum <= maxoff; offnum = OffsetNumberNext(offnum)) { IndexTuple itup; ItemPointer htup; itup = (IndexTuple) PageGetItem(page, PageGetItemId(page, offnum)); htup = &(itup->t_tid); if (callback(htup, callback_state)) deletable[ndeletable++] = offnum; } } /* * Apply any needed deletes. We issue just one _bt_delitems() call * per page, so as to minimize WAL traffic. */ if (ndeletable > 0) { _bt_delitems(rel, buf, deletable, ndeletable, true); stats->tuples_removed += ndeletable; /* must recompute maxoff */ maxoff = PageGetMaxOffsetNumber(page); } else { /* * If the page has been split during this vacuum cycle, it seems * worth expending a write to clear btpo_cycleid even if we don't * have any deletions to do. (If we do, _bt_delitems takes care * of this.) This ensures we won't process the page again. * * We treat this like a hint-bit update because there's no need to * WAL-log it. */ if (vstate->cycleid != 0 && opaque->btpo_cycleid == vstate->cycleid) { opaque->btpo_cycleid = 0; SetBufferCommitInfoNeedsSave(buf); } } /* * If it's now empty, try to delete; else count the live tuples. We * don't delete when recursing, though, to avoid putting entries into * freePages out-of-order (doesn't seem worth any extra code to handle * the case). */ if (minoff > maxoff) delete_now = (blkno == orig_blkno); else stats->num_index_tuples += maxoff - minoff + 1; } if (delete_now) { MemoryContext oldcontext; int ndel; /* Run pagedel in a temp context to avoid memory leakage */ MemoryContextReset(vstate->pagedelcontext); oldcontext = MemoryContextSwitchTo(vstate->pagedelcontext); ndel = _bt_pagedel(rel, buf, NULL, info->vacuum_full); /* count only this page, else may double-count parent */ if (ndel) stats->pages_deleted++; /* * During VACUUM FULL it's okay to recycle deleted pages immediately, * since there can be no other transactions scanning the index. Note * that we will only recycle the current page and not any parent pages * that _bt_pagedel might have recursed to; this seems reasonable in * the name of simplicity. (Trying to do otherwise would mean we'd * have to sort the list of recyclable pages we're building.) */ if (ndel && info->vacuum_full) { if (vstate->nFreePages < vstate->maxFreePages) vstate->freePages[vstate->nFreePages++] = blkno; vstate->totFreePages++; } MemoryContextSwitchTo(oldcontext); /* pagedel released buffer, so we shouldn't */ } else _bt_relbuf(rel, buf); MIRROREDLOCK_BUFMGR_UNLOCK; // -------- MirroredLock ---------- /* * This is really tail recursion, but if the compiler is too stupid to * optimize it as such, we'd eat an uncomfortably large amount of stack * space per recursion level (due to the deletable[] array). A failure is * improbable since the number of levels isn't likely to be large ... but * just in case, let's hand-optimize into a loop. */ if (recurse_to != P_NONE) { blkno = recurse_to; goto restart; } }
/* * _bt_endpoint() -- Find the first or last key in the index. * * This is used by _bt_first() to set up a scan when we've determined * that the scan must start at the beginning or end of the index (for * a forward or backward scan respectively). */ static bool _bt_endpoint(IndexScanDesc scan, ScanDirection dir) { Relation rel; Buffer buf; Page page; BTPageOpaque opaque; ItemPointer current; OffsetNumber maxoff; OffsetNumber start; BlockNumber blkno; BTItem btitem; IndexTuple itup; BTScanOpaque so; bool res; bool continuescan; rel = scan->indexRelation; current = &(scan->currentItemData); so = (BTScanOpaque) scan->opaque; /* * Scan down to the leftmost or rightmost leaf page. This is a * simplified version of _bt_search(). We don't maintain a stack * since we know we won't need it. */ buf = _bt_get_endpoint(rel, 0, ScanDirectionIsBackward(dir)); if (!BufferIsValid(buf)) { /* empty index... */ ItemPointerSetInvalid(current); so->btso_curbuf = InvalidBuffer; return false; } blkno = BufferGetBlockNumber(buf); page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); Assert(P_ISLEAF(opaque)); maxoff = PageGetMaxOffsetNumber(page); if (ScanDirectionIsForward(dir)) { /* There could be dead pages to the left, so not this: */ /* Assert(P_LEFTMOST(opaque)); */ start = P_FIRSTDATAKEY(opaque); } else if (ScanDirectionIsBackward(dir)) { Assert(P_RIGHTMOST(opaque)); start = PageGetMaxOffsetNumber(page); if (start < P_FIRSTDATAKEY(opaque)) /* watch out for empty * page */ start = P_FIRSTDATAKEY(opaque); } else { elog(ERROR, "invalid scan direction: %d", (int) dir); start = 0; /* keep compiler quiet */ } ItemPointerSet(current, blkno, start); /* remember which buffer we have pinned */ so->btso_curbuf = buf; /* * Left/rightmost page could be empty due to deletions, if so step * till we find a nonempty page. */ if (start > maxoff) { if (!_bt_step(scan, &buf, dir)) return false; start = ItemPointerGetOffsetNumber(current); page = BufferGetPage(buf); } btitem = (BTItem) PageGetItem(page, PageGetItemId(page, start)); itup = &(btitem->bti_itup); /* see if we picked a winner */ if (_bt_checkkeys(scan, itup, dir, &continuescan)) { /* yes, return it */ scan->xs_ctup.t_self = itup->t_tid; res = true; } else if (continuescan) { /* no, but there might be another one that is */ res = _bt_next(scan, dir); } else { /* no tuples in the index match this scan key */ ItemPointerSetInvalid(current); so->btso_curbuf = InvalidBuffer; _bt_relbuf(rel, buf); res = false; } return res; }
/** * @brief Read the left-most leaf page by walking down on index tree structure * from root node. * * Process flow * -# Open index file and read meta page * -# Get block number of root page * -# Read "fast root" page * -# Read left child page until reaching left-most leaf page * * After calling this function, the members of BTReader are the following: * - smgr : Smgr relation of the existing index file. * - blkno : block number of left-most leaf page. If there is no leaf page, * InvalidBlockNumber is set. * - offnum : InvalidOffsetNumber is set. * - page : Left-most leaf page, or undefined if no leaf page. * * @param reader [in/out] B-Tree index reader * @return true iff there are some tuples */ static bool BTReaderInit(BTReader *reader, Relation rel) { BTPageOpaque metaopaque; BTMetaPageData *metad; BTPageOpaque opaque; BlockNumber blkno; /* * HACK: We cannot use smgropen because smgrs returned from it * will be closed automatically when we assign a new file node. * * XXX: It might be better to open the previous relfilenode with * smgropen *after* RelationSetNewRelfilenode. */ memset(&reader->smgr, 0, sizeof(reader->smgr)); #if PG_VERSION_NUM >= 90100 reader->smgr.smgr_rnode.node = rel->rd_node; reader->smgr.smgr_rnode.backend = rel->rd_backend == MyBackendId ? MyBackendId : InvalidBackendId; #else reader->smgr.smgr_rnode = rel->rd_node; #endif reader->smgr.smgr_which = 0; /* md.c */ reader->blkno = InvalidBlockNumber; reader->offnum = InvalidOffsetNumber; reader->page = palloc(BLCKSZ); /* * Read meta page and check sanity of it. * * XXX: It might be better to do REINDEX against corrupted indexes * instead of raising errors because we've spent long time for data * loading... */ BTReaderReadPage(reader, BTREE_METAPAGE); metaopaque = (BTPageOpaque) PageGetSpecialPointer(reader->page); metad = BTPageGetMeta(reader->page); if (!(metaopaque->btpo_flags & BTP_META) || metad->btm_magic != BTREE_MAGIC) ereport(ERROR, (errcode(ERRCODE_INDEX_CORRUPTED), errmsg("index \"%s\" is not a reader", RelationGetRelationName(rel)))); if (metad->btm_version != BTREE_VERSION) ereport(ERROR, (errcode(ERRCODE_INDEX_CORRUPTED), errmsg("version mismatch in index \"%s\": file version %d," " code version %d", RelationGetRelationName(rel), metad->btm_version, BTREE_VERSION))); if (metad->btm_root == P_NONE) { /* No root page; We ignore the index in the subsequent build. */ reader->blkno = InvalidBlockNumber; return false; } /* Go to the fast root page. */ blkno = metad->btm_fastroot; BTReaderReadPage(reader, blkno); opaque = (BTPageOpaque) PageGetSpecialPointer(reader->page); /* Walk down to the left-most leaf page */ while (!P_ISLEAF(opaque)) { ItemId firstid; IndexTuple itup; /* Get the block number of the left child */ firstid = PageGetItemId(reader->page, P_FIRSTDATAKEY(opaque)); itup = (IndexTuple) PageGetItem(reader->page, firstid); blkno = ItemPointerGetBlockNumber(&(itup->t_tid)); /* Go down to children */ for (;;) { BTReaderReadPage(reader, blkno); opaque = (BTPageOpaque) PageGetSpecialPointer(reader->page); if (!P_IGNORE(opaque)) break; if (P_RIGHTMOST(opaque)) { /* We reach end of the index without any valid leaves. */ reader->blkno = InvalidBlockNumber; return false; } blkno = opaque->btpo_next; } } return true; }
/* * _bt_pagedel() -- Delete a page from the b-tree. * * This action unlinks the page from the b-tree structure, removing all * pointers leading to it --- but not touching its own left and right links. * The page cannot be physically reclaimed right away, since other processes * may currently be trying to follow links leading to the page; they have to * be allowed to use its right-link to recover. See nbtree/README. * * On entry, the target buffer must be pinned and read-locked. This lock and * pin will be dropped before exiting. * * Returns the number of pages successfully deleted (zero on failure; could * be more than one if parent blocks were deleted). * * NOTE: this leaks memory. Rather than trying to clean up everything * carefully, it's better to run it in a temp context that can be reset * frequently. */ int _bt_pagedel(Relation rel, Buffer buf, bool vacuum_full) { BlockNumber target, leftsib, rightsib, parent; OffsetNumber poffset, maxoff; uint32 targetlevel, ilevel; ItemId itemid; BTItem targetkey, btitem; ScanKey itup_scankey; BTStack stack; Buffer lbuf, rbuf, pbuf; bool parent_half_dead; bool parent_one_child; bool rightsib_empty; Buffer metabuf = InvalidBuffer; Page metapg = NULL; BTMetaPageData *metad = NULL; Page page; BTPageOpaque opaque; /* * We can never delete rightmost pages nor root pages. While at it, check * that page is not already deleted and is empty. */ page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); if (P_RIGHTMOST(opaque) || P_ISROOT(opaque) || P_ISDELETED(opaque) || P_FIRSTDATAKEY(opaque) <= PageGetMaxOffsetNumber(page)) { _bt_relbuf(rel, buf); return 0; } /* * Save info about page, including a copy of its high key (it must have * one, being non-rightmost). */ target = BufferGetBlockNumber(buf); targetlevel = opaque->btpo.level; leftsib = opaque->btpo_prev; itemid = PageGetItemId(page, P_HIKEY); targetkey = CopyBTItem((BTItem) PageGetItem(page, itemid)); /* * We need to get an approximate pointer to the page's parent page. Use * the standard search mechanism to search for the page's high key; this * will give us a link to either the current parent or someplace to its * left (if there are multiple equal high keys). To avoid deadlocks, we'd * better drop the target page lock first. */ _bt_relbuf(rel, buf); /* we need a scan key to do our search, so build one */ itup_scankey = _bt_mkscankey(rel, &(targetkey->bti_itup)); /* find the leftmost leaf page containing this key */ stack = _bt_search(rel, rel->rd_rel->relnatts, itup_scankey, false, &lbuf, BT_READ); /* don't need a pin on that either */ _bt_relbuf(rel, lbuf); /* * If we are trying to delete an interior page, _bt_search did more than * we needed. Locate the stack item pointing to our parent level. */ ilevel = 0; for (;;) { if (stack == NULL) elog(ERROR, "not enough stack items"); if (ilevel == targetlevel) break; stack = stack->bts_parent; ilevel++; } /* * We have to lock the pages we need to modify in the standard order: * moving right, then up. Else we will deadlock against other writers. * * So, we need to find and write-lock the current left sibling of the * target page. The sibling that was current a moment ago could have * split, so we may have to move right. This search could fail if either * the sibling or the target page was deleted by someone else meanwhile; * if so, give up. (Right now, that should never happen, since page * deletion is only done in VACUUM and there shouldn't be multiple VACUUMs * concurrently on the same table.) */ if (leftsib != P_NONE) { lbuf = _bt_getbuf(rel, leftsib, BT_WRITE); page = BufferGetPage(lbuf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); while (P_ISDELETED(opaque) || opaque->btpo_next != target) { /* step right one page */ leftsib = opaque->btpo_next; _bt_relbuf(rel, lbuf); if (leftsib == P_NONE) { elog(LOG, "no left sibling (concurrent deletion?) in \"%s\"", RelationGetRelationName(rel)); return 0; } lbuf = _bt_getbuf(rel, leftsib, BT_WRITE); page = BufferGetPage(lbuf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); } } else lbuf = InvalidBuffer; /* * Next write-lock the target page itself. It should be okay to take just * a write lock not a superexclusive lock, since no scans would stop on an * empty page. */ buf = _bt_getbuf(rel, target, BT_WRITE); page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); /* * Check page is still empty etc, else abandon deletion. The empty check * is necessary since someone else might have inserted into it while we * didn't have it locked; the others are just for paranoia's sake. */ if (P_RIGHTMOST(opaque) || P_ISROOT(opaque) || P_ISDELETED(opaque) || P_FIRSTDATAKEY(opaque) <= PageGetMaxOffsetNumber(page)) { _bt_relbuf(rel, buf); if (BufferIsValid(lbuf)) _bt_relbuf(rel, lbuf); return 0; } if (opaque->btpo_prev != leftsib) elog(ERROR, "left link changed unexpectedly in block %u of \"%s\"", target, RelationGetRelationName(rel)); /* * And next write-lock the (current) right sibling. */ rightsib = opaque->btpo_next; rbuf = _bt_getbuf(rel, rightsib, BT_WRITE); /* * Next find and write-lock the current parent of the target page. This is * essentially the same as the corresponding step of splitting. However, * it's possible for the search to fail (for reasons explained in README). * If that happens, we recover by searching the whole parent level, which * is a tad inefficient but doesn't happen often enough to be a problem. */ ItemPointerSet(&(stack->bts_btitem.bti_itup.t_tid), target, P_HIKEY); pbuf = _bt_getstackbuf(rel, stack, BT_WRITE); if (pbuf == InvalidBuffer) { /* Find the leftmost page in the parent level */ pbuf = _bt_get_endpoint(rel, opaque->btpo.level + 1, false); stack->bts_blkno = BufferGetBlockNumber(pbuf); stack->bts_offset = InvalidOffsetNumber; _bt_relbuf(rel, pbuf); /* and repeat search from there */ pbuf = _bt_getstackbuf(rel, stack, BT_WRITE); if (pbuf == InvalidBuffer) elog(ERROR, "failed to re-find parent key in \"%s\" for deletion target page %u", RelationGetRelationName(rel), target); } parent = stack->bts_blkno; poffset = stack->bts_offset; /* * If the target is the rightmost child of its parent, then we can't * delete, unless it's also the only child --- in which case the parent * changes to half-dead status. */ page = BufferGetPage(pbuf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); maxoff = PageGetMaxOffsetNumber(page); parent_half_dead = false; parent_one_child = false; if (poffset >= maxoff) { if (poffset == P_FIRSTDATAKEY(opaque)) parent_half_dead = true; else { _bt_relbuf(rel, pbuf); _bt_relbuf(rel, rbuf); _bt_relbuf(rel, buf); if (BufferIsValid(lbuf)) _bt_relbuf(rel, lbuf); return 0; } } else { /* Will there be exactly one child left in this parent? */ if (OffsetNumberNext(P_FIRSTDATAKEY(opaque)) == maxoff) parent_one_child = true; } /* * If we are deleting the next-to-last page on the target's level, then * the rightsib is a candidate to become the new fast root. (In theory, it * might be possible to push the fast root even further down, but the odds * of doing so are slim, and the locking considerations daunting.) * * We can safely acquire a lock on the metapage here --- see comments for * _bt_newroot(). */ if (leftsib == P_NONE) { page = BufferGetPage(rbuf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); Assert(opaque->btpo.level == targetlevel); if (P_RIGHTMOST(opaque)) { /* rightsib will be the only one left on the level */ metabuf = _bt_getbuf(rel, BTREE_METAPAGE, BT_WRITE); metapg = BufferGetPage(metabuf); metad = BTPageGetMeta(metapg); /* * The expected case here is btm_fastlevel == targetlevel+1; if * the fastlevel is <= targetlevel, something is wrong, and we * choose to overwrite it to fix it. */ if (metad->btm_fastlevel > targetlevel + 1) { /* no update wanted */ _bt_relbuf(rel, metabuf); metabuf = InvalidBuffer; } } } /* * Here we begin doing the deletion. */ /* No ereport(ERROR) until changes are logged */ START_CRIT_SECTION(); /* * Update parent. The normal case is a tad tricky because we want to * delete the target's downlink and the *following* key. Easiest way is * to copy the right sibling's downlink over the target downlink, and then * delete the following item. */ page = BufferGetPage(pbuf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); if (parent_half_dead) { PageIndexTupleDelete(page, poffset); opaque->btpo_flags |= BTP_HALF_DEAD; } else { OffsetNumber nextoffset; itemid = PageGetItemId(page, poffset); btitem = (BTItem) PageGetItem(page, itemid); Assert(ItemPointerGetBlockNumber(&(btitem->bti_itup.t_tid)) == target); ItemPointerSet(&(btitem->bti_itup.t_tid), rightsib, P_HIKEY); nextoffset = OffsetNumberNext(poffset); /* This part is just for double-checking */ itemid = PageGetItemId(page, nextoffset); btitem = (BTItem) PageGetItem(page, itemid); if (ItemPointerGetBlockNumber(&(btitem->bti_itup.t_tid)) != rightsib) elog(PANIC, "right sibling is not next child in \"%s\"", RelationGetRelationName(rel)); PageIndexTupleDelete(page, nextoffset); } /* * Update siblings' side-links. Note the target page's side-links will * continue to point to the siblings. */ if (BufferIsValid(lbuf)) { page = BufferGetPage(lbuf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); Assert(opaque->btpo_next == target); opaque->btpo_next = rightsib; } page = BufferGetPage(rbuf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); Assert(opaque->btpo_prev == target); opaque->btpo_prev = leftsib; rightsib_empty = (P_FIRSTDATAKEY(opaque) > PageGetMaxOffsetNumber(page)); /* * Mark the page itself deleted. It can be recycled when all current * transactions are gone; or immediately if we're doing VACUUM FULL. */ page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); opaque->btpo_flags |= BTP_DELETED; opaque->btpo.xact = vacuum_full ? FrozenTransactionId : ReadNewTransactionId(); /* And update the metapage, if needed */ if (BufferIsValid(metabuf)) { metad->btm_fastroot = rightsib; metad->btm_fastlevel = targetlevel; } /* XLOG stuff */ if (!rel->rd_istemp) { xl_btree_delete_page xlrec; xl_btree_metadata xlmeta; uint8 xlinfo; XLogRecPtr recptr; XLogRecData rdata[5]; XLogRecData *nextrdata; xlrec.target.node = rel->rd_node; ItemPointerSet(&(xlrec.target.tid), parent, poffset); xlrec.deadblk = target; xlrec.leftblk = leftsib; xlrec.rightblk = rightsib; rdata[0].data = (char *) &xlrec; rdata[0].len = SizeOfBtreeDeletePage; rdata[0].buffer = InvalidBuffer; rdata[0].next = nextrdata = &(rdata[1]); if (BufferIsValid(metabuf)) { xlmeta.root = metad->btm_root; xlmeta.level = metad->btm_level; xlmeta.fastroot = metad->btm_fastroot; xlmeta.fastlevel = metad->btm_fastlevel; nextrdata->data = (char *) &xlmeta; nextrdata->len = sizeof(xl_btree_metadata); nextrdata->buffer = InvalidBuffer; nextrdata->next = nextrdata + 1; nextrdata++; xlinfo = XLOG_BTREE_DELETE_PAGE_META; } else xlinfo = XLOG_BTREE_DELETE_PAGE; nextrdata->data = NULL; nextrdata->len = 0; nextrdata->next = nextrdata + 1; nextrdata->buffer = pbuf; nextrdata->buffer_std = true; nextrdata++; nextrdata->data = NULL; nextrdata->len = 0; nextrdata->buffer = rbuf; nextrdata->buffer_std = true; nextrdata->next = NULL; if (BufferIsValid(lbuf)) { nextrdata->next = nextrdata + 1; nextrdata++; nextrdata->data = NULL; nextrdata->len = 0; nextrdata->buffer = lbuf; nextrdata->buffer_std = true; nextrdata->next = NULL; } recptr = XLogInsert(RM_BTREE_ID, xlinfo, rdata); if (BufferIsValid(metabuf)) { PageSetLSN(metapg, recptr); PageSetTLI(metapg, ThisTimeLineID); } page = BufferGetPage(pbuf); PageSetLSN(page, recptr); PageSetTLI(page, ThisTimeLineID); page = BufferGetPage(rbuf); PageSetLSN(page, recptr); PageSetTLI(page, ThisTimeLineID); page = BufferGetPage(buf); PageSetLSN(page, recptr); PageSetTLI(page, ThisTimeLineID); if (BufferIsValid(lbuf)) { page = BufferGetPage(lbuf); PageSetLSN(page, recptr); PageSetTLI(page, ThisTimeLineID); } } END_CRIT_SECTION(); /* Write and release buffers */ if (BufferIsValid(metabuf)) _bt_wrtbuf(rel, metabuf); _bt_wrtbuf(rel, pbuf); _bt_wrtbuf(rel, rbuf); _bt_wrtbuf(rel, buf); if (BufferIsValid(lbuf)) _bt_wrtbuf(rel, lbuf); /* * If parent became half dead, recurse to try to delete it. Otherwise, if * right sibling is empty and is now the last child of the parent, recurse * to try to delete it. (These cases cannot apply at the same time, * though the second case might itself recurse to the first.) */ if (parent_half_dead) { buf = _bt_getbuf(rel, parent, BT_READ); return _bt_pagedel(rel, buf, vacuum_full) + 1; } if (parent_one_child && rightsib_empty) { buf = _bt_getbuf(rel, rightsib, BT_READ); return _bt_pagedel(rel, buf, vacuum_full) + 1; } return 1; }
/* * _bt_gettrueroot() -- Get the true root page of the btree. * * This is the same as the BT_READ case of _bt_getroot(), except * we follow the true-root link not the fast-root link. * * By the time we acquire lock on the root page, it might have been split and * not be the true root anymore. This is okay for the present uses of this * routine; we only really need to be able to move up at least one tree level * from whatever non-root page we were at. If we ever do need to lock the * one true root page, we could loop here, re-reading the metapage on each * failure. (Note that it wouldn't do to hold the lock on the metapage while * moving to the root --- that'd deadlock against any concurrent root split.) */ Buffer _bt_gettrueroot(Relation rel) { Buffer metabuf; Page metapg; BTPageOpaque metaopaque; Buffer rootbuf; Page rootpage; BTPageOpaque rootopaque; BlockNumber rootblkno; uint32 rootlevel; BTMetaPageData *metad; metabuf = _bt_getbuf(rel, BTREE_METAPAGE, BT_READ); metapg = BufferGetPage(metabuf); metaopaque = (BTPageOpaque) PageGetSpecialPointer(metapg); metad = BTPageGetMeta(metapg); if (!(metaopaque->btpo_flags & BTP_META) || metad->btm_magic != BTREE_MAGIC) ereport(ERROR, (errcode(ERRCODE_INDEX_CORRUPTED), errmsg("index \"%s\" is not a btree", RelationGetRelationName(rel)))); if (metad->btm_version != BTREE_VERSION) ereport(ERROR, (errcode(ERRCODE_INDEX_CORRUPTED), errmsg("version mismatch in index \"%s\": file version %d, code version %d", RelationGetRelationName(rel), metad->btm_version, BTREE_VERSION))); /* if no root page initialized yet, fail */ if (metad->btm_root == P_NONE) { _bt_relbuf(rel, metabuf); return InvalidBuffer; } rootblkno = metad->btm_root; rootlevel = metad->btm_level; /* * We are done with the metapage; arrange to release it via first * _bt_relandgetbuf call */ rootbuf = metabuf; for (;;) { rootbuf = _bt_relandgetbuf(rel, rootbuf, rootblkno, BT_READ); rootpage = BufferGetPage(rootbuf); rootopaque = (BTPageOpaque) PageGetSpecialPointer(rootpage); if (!P_IGNORE(rootopaque)) break; /* it's dead, Jim. step right one page */ if (P_RIGHTMOST(rootopaque)) elog(ERROR, "no live root page found in \"%s\"", RelationGetRelationName(rel)); rootblkno = rootopaque->btpo_next; } /* Note: can't check btpo.level on deleted pages */ if (rootopaque->btpo.level != rootlevel) elog(ERROR, "root page %u of \"%s\" has level %u, expected %u", rootblkno, RelationGetRelationName(rel), rootopaque->btpo.level, rootlevel); return rootbuf; }
/* * _bt_pagedel() -- Delete a page from the b-tree, if legal to do so. * * This action unlinks the page from the b-tree structure, removing all * pointers leading to it --- but not touching its own left and right links. * The page cannot be physically reclaimed right away, since other processes * may currently be trying to follow links leading to the page; they have to * be allowed to use its right-link to recover. See nbtree/README. * * On entry, the target buffer must be pinned and locked (either read or write * lock is OK). This lock and pin will be dropped before exiting. * * The "stack" argument can be a search stack leading (approximately) to the * target page, or NULL --- outside callers typically pass NULL since they * have not done such a search, but internal recursion cases pass the stack * to avoid duplicated search effort. * * Returns the number of pages successfully deleted (zero if page cannot * be deleted now; could be more than one if parent pages were deleted too). * * NOTE: this leaks memory. Rather than trying to clean up everything * carefully, it's better to run it in a temp context that can be reset * frequently. */ int _bt_pagedel(Relation rel, Buffer buf, BTStack stack, bool vacuum_full) { int result; BlockNumber target, leftsib, rightsib, parent; OffsetNumber poffset, maxoff; uint32 targetlevel, ilevel; ItemId itemid; IndexTuple targetkey, itup; ScanKey itup_scankey; Buffer lbuf, rbuf, pbuf; bool parent_half_dead; bool parent_one_child; bool rightsib_empty; Buffer metabuf = InvalidBuffer; Page metapg = NULL; BTMetaPageData *metad = NULL; Page page; BTPageOpaque opaque; MIRROREDLOCK_BUFMGR_MUST_ALREADY_BE_HELD; // Fetch gp_persistent_relation_node information that will be added to XLOG record. RelationFetchGpRelationNodeForXLog(rel); /* * We can never delete rightmost pages nor root pages. While at it, check * that page is not already deleted and is empty. */ page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); if (P_RIGHTMOST(opaque) || P_ISROOT(opaque) || P_ISDELETED(opaque) || P_FIRSTDATAKEY(opaque) <= PageGetMaxOffsetNumber(page)) { /* Should never fail to delete a half-dead page */ Assert(!P_ISHALFDEAD(opaque)); _bt_relbuf(rel, buf); return 0; } /* * Save info about page, including a copy of its high key (it must have * one, being non-rightmost). */ target = BufferGetBlockNumber(buf); targetlevel = opaque->btpo.level; leftsib = opaque->btpo_prev; itemid = PageGetItemId(page, P_HIKEY); targetkey = CopyIndexTuple((IndexTuple) PageGetItem(page, itemid)); /* * To avoid deadlocks, we'd better drop the target page lock before going * further. */ _bt_relbuf(rel, buf); /* * We need an approximate pointer to the page's parent page. We use the * standard search mechanism to search for the page's high key; this will * give us a link to either the current parent or someplace to its left * (if there are multiple equal high keys). In recursion cases, the * caller already generated a search stack and we can just re-use that * work. */ if (stack == NULL) { if (!InRecovery) { /* we need an insertion scan key to do our search, so build one */ itup_scankey = _bt_mkscankey(rel, targetkey); /* find the leftmost leaf page containing this key */ stack = _bt_search(rel, rel->rd_rel->relnatts, itup_scankey, false, &lbuf, BT_READ); /* don't need a pin on that either */ _bt_relbuf(rel, lbuf); /* * If we are trying to delete an interior page, _bt_search did * more than we needed. Locate the stack item pointing to our * parent level. */ ilevel = 0; for (;;) { if (stack == NULL) elog(ERROR, "not enough stack items"); if (ilevel == targetlevel) break; stack = stack->bts_parent; ilevel++; } } else { /* * During WAL recovery, we can't use _bt_search (for one reason, * it might invoke user-defined comparison functions that expect * facilities not available in recovery mode). Instead, just set * up a dummy stack pointing to the left end of the parent tree * level, from which _bt_getstackbuf will walk right to the parent * page. Painful, but we don't care too much about performance in * this scenario. */ pbuf = _bt_get_endpoint(rel, targetlevel + 1, false); stack = (BTStack) palloc(sizeof(BTStackData)); stack->bts_blkno = BufferGetBlockNumber(pbuf); stack->bts_offset = InvalidOffsetNumber; /* bts_btentry will be initialized below */ stack->bts_parent = NULL; _bt_relbuf(rel, pbuf); } } /* * We cannot delete a page that is the rightmost child of its immediate * parent, unless it is the only child --- in which case the parent has to * be deleted too, and the same condition applies recursively to it. We * have to check this condition all the way up before trying to delete. We * don't need to re-test when deleting a non-leaf page, though. */ if (targetlevel == 0 && !_bt_parent_deletion_safe(rel, target, stack)) return 0; /* * We have to lock the pages we need to modify in the standard order: * moving right, then up. Else we will deadlock against other writers. * * So, we need to find and write-lock the current left sibling of the * target page. The sibling that was current a moment ago could have * split, so we may have to move right. This search could fail if either * the sibling or the target page was deleted by someone else meanwhile; * if so, give up. (Right now, that should never happen, since page * deletion is only done in VACUUM and there shouldn't be multiple VACUUMs * concurrently on the same table.) */ if (leftsib != P_NONE) { lbuf = _bt_getbuf(rel, leftsib, BT_WRITE); page = BufferGetPage(lbuf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); while (P_ISDELETED(opaque) || opaque->btpo_next != target) { /* step right one page */ leftsib = opaque->btpo_next; _bt_relbuf(rel, lbuf); if (leftsib == P_NONE) { elog(LOG, "no left sibling (concurrent deletion?) in \"%s\"", RelationGetRelationName(rel)); return 0; } lbuf = _bt_getbuf(rel, leftsib, BT_WRITE); page = BufferGetPage(lbuf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); } } else lbuf = InvalidBuffer; /* * Next write-lock the target page itself. It should be okay to take just * a write lock not a superexclusive lock, since no scans would stop on an * empty page. */ buf = _bt_getbuf(rel, target, BT_WRITE); page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); /* * Check page is still empty etc, else abandon deletion. The empty check * is necessary since someone else might have inserted into it while we * didn't have it locked; the others are just for paranoia's sake. */ if (P_RIGHTMOST(opaque) || P_ISROOT(opaque) || P_ISDELETED(opaque) || P_FIRSTDATAKEY(opaque) <= PageGetMaxOffsetNumber(page)) { _bt_relbuf(rel, buf); if (BufferIsValid(lbuf)) _bt_relbuf(rel, lbuf); return 0; } if (opaque->btpo_prev != leftsib) elog(ERROR, "left link changed unexpectedly in block %u of index \"%s\"", target, RelationGetRelationName(rel)); /* * And next write-lock the (current) right sibling. */ rightsib = opaque->btpo_next; rbuf = _bt_getbuf(rel, rightsib, BT_WRITE); page = BufferGetPage(rbuf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); if (opaque->btpo_prev != target) elog(ERROR, "right sibling's left-link doesn't match: " "block %u links to %u instead of expected %u in index \"%s\"", rightsib, opaque->btpo_prev, target, RelationGetRelationName(rel)); /* * Next find and write-lock the current parent of the target page. This is * essentially the same as the corresponding step of splitting. */ ItemPointerSet(&(stack->bts_btentry.t_tid), target, P_HIKEY); pbuf = _bt_getstackbuf(rel, stack, BT_WRITE); if (pbuf == InvalidBuffer) elog(ERROR, "failed to re-find parent key in index \"%s\" for deletion target page %u", RelationGetRelationName(rel), target); parent = stack->bts_blkno; poffset = stack->bts_offset; /* * If the target is the rightmost child of its parent, then we can't * delete, unless it's also the only child --- in which case the parent * changes to half-dead status. The "can't delete" case should have been * detected by _bt_parent_deletion_safe, so complain if we see it now. */ page = BufferGetPage(pbuf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); maxoff = PageGetMaxOffsetNumber(page); parent_half_dead = false; parent_one_child = false; if (poffset >= maxoff) { if (poffset == P_FIRSTDATAKEY(opaque)) parent_half_dead = true; else elog(ERROR, "failed to delete rightmost child %u of block %u in index \"%s\"", target, parent, RelationGetRelationName(rel)); } else { /* Will there be exactly one child left in this parent? */ if (OffsetNumberNext(P_FIRSTDATAKEY(opaque)) == maxoff) parent_one_child = true; } /* * If we are deleting the next-to-last page on the target's level, then * the rightsib is a candidate to become the new fast root. (In theory, it * might be possible to push the fast root even further down, but the odds * of doing so are slim, and the locking considerations daunting.) * * We don't support handling this in the case where the parent is becoming * half-dead, even though it theoretically could occur. * * We can safely acquire a lock on the metapage here --- see comments for * _bt_newroot(). */ if (leftsib == P_NONE && !parent_half_dead) { page = BufferGetPage(rbuf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); Assert(opaque->btpo.level == targetlevel); if (P_RIGHTMOST(opaque)) { /* rightsib will be the only one left on the level */ metabuf = _bt_getbuf(rel, BTREE_METAPAGE, BT_WRITE); metapg = BufferGetPage(metabuf); metad = BTPageGetMeta(metapg); /* * The expected case here is btm_fastlevel == targetlevel+1; if * the fastlevel is <= targetlevel, something is wrong, and we * choose to overwrite it to fix it. */ if (metad->btm_fastlevel > targetlevel + 1) { /* no update wanted */ _bt_relbuf(rel, metabuf); metabuf = InvalidBuffer; } } } /* * Check that the parent-page index items we're about to delete/overwrite * contain what we expect. This can fail if the index has become * corrupt for some reason. We want to throw any error before entering * the critical section --- otherwise it'd be a PANIC. * * The test on the target item is just an Assert because _bt_getstackbuf * should have guaranteed it has the expected contents. The test on the * next-child downlink is known to sometimes fail in the field, though. */ page = BufferGetPage(pbuf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); #ifdef USE_ASSERT_CHECKING itemid = PageGetItemId(page, poffset); itup = (IndexTuple) PageGetItem(page, itemid); Assert(ItemPointerGetBlockNumber(&(itup->t_tid)) == target); #endif if (!parent_half_dead) { OffsetNumber nextoffset; nextoffset = OffsetNumberNext(poffset); itemid = PageGetItemId(page, nextoffset); itup = (IndexTuple) PageGetItem(page, itemid); if (ItemPointerGetBlockNumber(&(itup->t_tid)) != rightsib) elog(ERROR, "right sibling %u of block %u is not next child %u of block %u in index \"%s\"", rightsib, target, ItemPointerGetBlockNumber(&(itup->t_tid)), parent, RelationGetRelationName(rel)); } /* * Here we begin doing the deletion. */ /* No ereport(ERROR) until changes are logged */ START_CRIT_SECTION(); /* * Update parent. The normal case is a tad tricky because we want to * delete the target's downlink and the *following* key. Easiest way is * to copy the right sibling's downlink over the target downlink, and then * delete the following item. */ if (parent_half_dead) { PageIndexTupleDelete(page, poffset); opaque->btpo_flags |= BTP_HALF_DEAD; } else { OffsetNumber nextoffset; itemid = PageGetItemId(page, poffset); itup = (IndexTuple) PageGetItem(page, itemid); ItemPointerSet(&(itup->t_tid), rightsib, P_HIKEY); nextoffset = OffsetNumberNext(poffset); PageIndexTupleDelete(page, nextoffset); } /* * Update siblings' side-links. Note the target page's side-links will * continue to point to the siblings. Asserts here are just rechecking * things we already verified above. */ if (BufferIsValid(lbuf)) { page = BufferGetPage(lbuf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); Assert(opaque->btpo_next == target); opaque->btpo_next = rightsib; } page = BufferGetPage(rbuf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); Assert(opaque->btpo_prev == target); opaque->btpo_prev = leftsib; rightsib_empty = (P_FIRSTDATAKEY(opaque) > PageGetMaxOffsetNumber(page)); /* * Mark the page itself deleted. It can be recycled when all current * transactions are gone; or immediately if we're doing VACUUM FULL. */ page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); opaque->btpo_flags &= ~BTP_HALF_DEAD; opaque->btpo_flags |= BTP_DELETED; opaque->btpo.xact = vacuum_full ? FrozenTransactionId : ReadNewTransactionId(); /* And update the metapage, if needed */ if (BufferIsValid(metabuf)) { metad->btm_fastroot = rightsib; metad->btm_fastlevel = targetlevel; MarkBufferDirty(metabuf); } /* Must mark buffers dirty before XLogInsert */ MarkBufferDirty(pbuf); MarkBufferDirty(rbuf); MarkBufferDirty(buf); if (BufferIsValid(lbuf)) MarkBufferDirty(lbuf); /* XLOG stuff */ if (!rel->rd_istemp) { xl_btree_delete_page xlrec; xl_btree_metadata xlmeta; uint8 xlinfo; XLogRecPtr recptr; XLogRecData rdata[5]; XLogRecData *nextrdata; xl_btreetid_set(&(xlrec.target), rel, parent, poffset); xlrec.deadblk = target; xlrec.leftblk = leftsib; xlrec.rightblk = rightsib; rdata[0].data = (char *) &xlrec; rdata[0].len = SizeOfBtreeDeletePage; rdata[0].buffer = InvalidBuffer; rdata[0].next = nextrdata = &(rdata[1]); if (BufferIsValid(metabuf)) { xlmeta.root = metad->btm_root; xlmeta.level = metad->btm_level; xlmeta.fastroot = metad->btm_fastroot; xlmeta.fastlevel = metad->btm_fastlevel; nextrdata->data = (char *) &xlmeta; nextrdata->len = sizeof(xl_btree_metadata); nextrdata->buffer = InvalidBuffer; nextrdata->next = nextrdata + 1; nextrdata++; xlinfo = XLOG_BTREE_DELETE_PAGE_META; } else if (parent_half_dead) xlinfo = XLOG_BTREE_DELETE_PAGE_HALF; else xlinfo = XLOG_BTREE_DELETE_PAGE; nextrdata->data = NULL; nextrdata->len = 0; nextrdata->next = nextrdata + 1; nextrdata->buffer = pbuf; nextrdata->buffer_std = true; nextrdata++; nextrdata->data = NULL; nextrdata->len = 0; nextrdata->buffer = rbuf; nextrdata->buffer_std = true; nextrdata->next = NULL; if (BufferIsValid(lbuf)) { nextrdata->next = nextrdata + 1; nextrdata++; nextrdata->data = NULL; nextrdata->len = 0; nextrdata->buffer = lbuf; nextrdata->buffer_std = true; nextrdata->next = NULL; } recptr = XLogInsert(RM_BTREE_ID, xlinfo, rdata); if (BufferIsValid(metabuf)) { PageSetLSN(metapg, recptr); PageSetTLI(metapg, ThisTimeLineID); } page = BufferGetPage(pbuf); PageSetLSN(page, recptr); PageSetTLI(page, ThisTimeLineID); page = BufferGetPage(rbuf); PageSetLSN(page, recptr); PageSetTLI(page, ThisTimeLineID); page = BufferGetPage(buf); PageSetLSN(page, recptr); PageSetTLI(page, ThisTimeLineID); if (BufferIsValid(lbuf)) { page = BufferGetPage(lbuf); PageSetLSN(page, recptr); PageSetTLI(page, ThisTimeLineID); } } END_CRIT_SECTION(); /* release metapage; send out relcache inval if metapage changed */ if (BufferIsValid(metabuf)) { CacheInvalidateRelcache(rel); _bt_relbuf(rel, metabuf); } /* can always release leftsib immediately */ if (BufferIsValid(lbuf)) _bt_relbuf(rel, lbuf); /* * If parent became half dead, recurse to delete it. Otherwise, if right * sibling is empty and is now the last child of the parent, recurse to * try to delete it. (These cases cannot apply at the same time, though * the second case might itself recurse to the first.) * * When recursing to parent, we hold the lock on the target page until * done. This delays any insertions into the keyspace that was just * effectively reassigned to the parent's right sibling. If we allowed * that, and there were enough such insertions before we finish deleting * the parent, page splits within that keyspace could lead to inserting * out-of-order keys into the grandparent level. It is thought that that * wouldn't have any serious consequences, but it still seems like a * pretty bad idea. */ if (parent_half_dead) { /* recursive call will release pbuf */ _bt_relbuf(rel, rbuf); result = _bt_pagedel(rel, pbuf, stack->bts_parent, vacuum_full) + 1; _bt_relbuf(rel, buf); } else if (parent_one_child && rightsib_empty) { _bt_relbuf(rel, pbuf); _bt_relbuf(rel, buf); /* recursive call will release rbuf */ result = _bt_pagedel(rel, rbuf, stack, vacuum_full) + 1; } else { _bt_relbuf(rel, pbuf); _bt_relbuf(rel, buf); _bt_relbuf(rel, rbuf); result = 1; } return result; }
/* * _bt_getroot() -- Get the root page of the btree. * * Since the root page can move around the btree file, we have to read * its location from the metadata page, and then read the root page * itself. If no root page exists yet, we have to create one. The * standard class of race conditions exists here; I think I covered * them all in the Hopi Indian rain dance of lock requests below. * * The access type parameter (BT_READ or BT_WRITE) controls whether * a new root page will be created or not. If access = BT_READ, * and no root page exists, we just return InvalidBuffer. For * BT_WRITE, we try to create the root page if it doesn't exist. * NOTE that the returned root page will have only a read lock set * on it even if access = BT_WRITE! * * The returned page is not necessarily the true root --- it could be * a "fast root" (a page that is alone in its level due to deletions). * Also, if the root page is split while we are "in flight" to it, * what we will return is the old root, which is now just the leftmost * page on a probably-not-very-wide level. For most purposes this is * as good as or better than the true root, so we do not bother to * insist on finding the true root. We do, however, guarantee to * return a live (not deleted or half-dead) page. * * On successful return, the root page is pinned and read-locked. * The metadata page is not locked or pinned on exit. */ Buffer _bt_getroot(Relation rel, int access) { Buffer metabuf; Page metapg; BTPageOpaque metaopaque; Buffer rootbuf; Page rootpage; BTPageOpaque rootopaque; BlockNumber rootblkno; uint32 rootlevel; BTMetaPageData *metad; MIRROREDLOCK_BUFMGR_MUST_ALREADY_BE_HELD; /* * Try to use previously-cached metapage data to find the root. This * normally saves one buffer access per index search, which is a very * helpful savings in bufmgr traffic and hence contention. */ if (rel->rd_amcache != NULL) { metad = (BTMetaPageData *) rel->rd_amcache; /* We shouldn't have cached it if any of these fail */ Assert(metad->btm_magic == BTREE_MAGIC); Assert(metad->btm_version == BTREE_VERSION); Assert(metad->btm_root != P_NONE); rootblkno = metad->btm_fastroot; Assert(rootblkno != P_NONE); rootlevel = metad->btm_fastlevel; rootbuf = _bt_getbuf(rel, rootblkno, BT_READ); rootpage = BufferGetPage(rootbuf); rootopaque = (BTPageOpaque) PageGetSpecialPointer(rootpage); /* * Since the cache might be stale, we check the page more carefully * here than normal. We *must* check that it's not deleted. If it's * not alone on its level, then we reject too --- this may be overly * paranoid but better safe than sorry. Note we don't check P_ISROOT, * because that's not set in a "fast root". */ if (!P_IGNORE(rootopaque) && rootopaque->btpo.level == rootlevel && P_LEFTMOST(rootopaque) && P_RIGHTMOST(rootopaque)) { /* OK, accept cached page as the root */ return rootbuf; } _bt_relbuf(rel, rootbuf); /* Cache is stale, throw it away */ if (rel->rd_amcache) pfree(rel->rd_amcache); rel->rd_amcache = NULL; } metabuf = _bt_getbuf(rel, BTREE_METAPAGE, BT_READ); metapg = BufferGetPage(metabuf); metaopaque = (BTPageOpaque) PageGetSpecialPointer(metapg); metad = BTPageGetMeta(metapg); /* sanity-check the metapage */ if (!(metaopaque->btpo_flags & BTP_META) || metad->btm_magic != BTREE_MAGIC) ereport(ERROR, (errcode(ERRCODE_INDEX_CORRUPTED), errmsg("index \"%s\" is not a btree", RelationGetRelationName(rel)))); if (metad->btm_version != BTREE_VERSION) ereport(ERROR, (errcode(ERRCODE_INDEX_CORRUPTED), errmsg("version mismatch in index \"%s\": file version %d, code version %d", RelationGetRelationName(rel), metad->btm_version, BTREE_VERSION))); /* if no root page initialized yet, do it */ if (metad->btm_root == P_NONE) { /* If access = BT_READ, caller doesn't want us to create root yet */ if (access == BT_READ) { _bt_relbuf(rel, metabuf); return InvalidBuffer; } // Fetch gp_persistent_relation_node information that will be added to XLOG record. RelationFetchGpRelationNodeForXLog(rel); /* trade in our read lock for a write lock */ LockBuffer(metabuf, BUFFER_LOCK_UNLOCK); LockBuffer(metabuf, BT_WRITE); /* * Race condition: if someone else initialized the metadata between * the time we released the read lock and acquired the write lock, we * must avoid doing it again. */ if (metad->btm_root != P_NONE) { /* * Metadata initialized by someone else. In order to guarantee no * deadlocks, we have to release the metadata page and start all * over again. (Is that really true? But it's hardly worth trying * to optimize this case.) */ _bt_relbuf(rel, metabuf); return _bt_getroot(rel, access); } /* * Get, initialize, write, and leave a lock of the appropriate type on * the new root page. Since this is the first page in the tree, it's * a leaf as well as the root. */ rootbuf = _bt_getbuf(rel, P_NEW, BT_WRITE); rootblkno = BufferGetBlockNumber(rootbuf); rootpage = BufferGetPage(rootbuf); rootopaque = (BTPageOpaque) PageGetSpecialPointer(rootpage); rootopaque->btpo_prev = rootopaque->btpo_next = P_NONE; rootopaque->btpo_flags = (BTP_LEAF | BTP_ROOT); rootopaque->btpo.level = 0; rootopaque->btpo_cycleid = 0; /* NO ELOG(ERROR) till meta is updated */ START_CRIT_SECTION(); metad->btm_root = rootblkno; metad->btm_level = 0; metad->btm_fastroot = rootblkno; metad->btm_fastlevel = 0; MarkBufferDirty(rootbuf); MarkBufferDirty(metabuf); /* XLOG stuff */ if (!rel->rd_istemp) { xl_btree_newroot xlrec; XLogRecPtr recptr; XLogRecData rdata; xl_btreenode_set(&(xlrec.btreenode), rel); xlrec.rootblk = rootblkno; xlrec.level = 0; rdata.data = (char *) &xlrec; rdata.len = SizeOfBtreeNewroot; rdata.buffer = InvalidBuffer; rdata.next = NULL; recptr = XLogInsert(RM_BTREE_ID, XLOG_BTREE_NEWROOT, &rdata); PageSetLSN(rootpage, recptr); PageSetTLI(rootpage, ThisTimeLineID); PageSetLSN(metapg, recptr); PageSetTLI(metapg, ThisTimeLineID); } END_CRIT_SECTION(); /* * Send out relcache inval for metapage change (probably unnecessary * here, but let's be safe). */ CacheInvalidateRelcache(rel); /* * swap root write lock for read lock. There is no danger of anyone * else accessing the new root page while it's unlocked, since no one * else knows where it is yet. */ LockBuffer(rootbuf, BUFFER_LOCK_UNLOCK); LockBuffer(rootbuf, BT_READ); /* okay, metadata is correct, release lock on it */ _bt_relbuf(rel, metabuf); } else { rootblkno = metad->btm_fastroot; Assert(rootblkno != P_NONE); rootlevel = metad->btm_fastlevel; /* * Cache the metapage data for next time */ rel->rd_amcache = MemoryContextAlloc(rel->rd_indexcxt, sizeof(BTMetaPageData)); memcpy(rel->rd_amcache, metad, sizeof(BTMetaPageData)); /* * We are done with the metapage; arrange to release it via first * _bt_relandgetbuf call */ rootbuf = metabuf; for (;;) { rootbuf = _bt_relandgetbuf(rel, rootbuf, rootblkno, BT_READ); rootpage = BufferGetPage(rootbuf); rootopaque = (BTPageOpaque) PageGetSpecialPointer(rootpage); if (!P_IGNORE(rootopaque)) break; /* it's dead, Jim. step right one page */ if (P_RIGHTMOST(rootopaque)) elog(ERROR, "no live root page found in index \"%s\"", RelationGetRelationName(rel)); rootblkno = rootopaque->btpo_next; } /* Note: can't check btpo.level on deleted pages */ if (rootopaque->btpo.level != rootlevel) elog(ERROR, "root page %u of index \"%s\" has level %u, expected %u", rootblkno, RelationGetRelationName(rel), rootopaque->btpo.level, rootlevel); } /* * By here, we have a pin and read lock on the root page, and no lock set * on the metadata page. Return the root page's buffer. */ return rootbuf; }
/* * _bt_step() -- Step one item in the requested direction in a scan on * the tree. * * *bufP is the current buffer (read-locked and pinned). If we change * pages, it's updated appropriately. * * If successful, update scan's currentItemData and return true. * If no adjacent record exists in the requested direction, * release buffer pin/locks and return false. */ bool _bt_step(IndexScanDesc scan, Buffer *bufP, ScanDirection dir) { Relation rel = scan->indexRelation; ItemPointer current = &(scan->currentItemData); BTScanOpaque so = (BTScanOpaque) scan->opaque; Page page; BTPageOpaque opaque; OffsetNumber offnum, maxoff; BlockNumber blkno; /* * Don't use ItemPointerGetOffsetNumber or you risk to get assertion * due to ability of ip_posid to be equal 0. */ offnum = current->ip_posid; page = BufferGetPage(*bufP); opaque = (BTPageOpaque) PageGetSpecialPointer(page); maxoff = PageGetMaxOffsetNumber(page); if (ScanDirectionIsForward(dir)) { if (!PageIsEmpty(page) && offnum < maxoff) offnum = OffsetNumberNext(offnum); else { /* Walk right to the next page with data */ for (;;) { /* if we're at end of scan, release the buffer and return */ if (P_RIGHTMOST(opaque)) { _bt_relbuf(rel, *bufP); ItemPointerSetInvalid(current); *bufP = so->btso_curbuf = InvalidBuffer; return false; } /* step right one page */ blkno = opaque->btpo_next; _bt_relbuf(rel, *bufP); *bufP = _bt_getbuf(rel, blkno, BT_READ); page = BufferGetPage(*bufP); opaque = (BTPageOpaque) PageGetSpecialPointer(page); if (!P_IGNORE(opaque)) { maxoff = PageGetMaxOffsetNumber(page); /* done if it's not empty */ offnum = P_FIRSTDATAKEY(opaque); if (!PageIsEmpty(page) && offnum <= maxoff) break; } } } } else /* backwards scan */ { if (offnum > P_FIRSTDATAKEY(opaque)) offnum = OffsetNumberPrev(offnum); else { /* * Walk left to the next page with data. This is much more * complex than the walk-right case because of the possibility * that the page to our left splits while we are in flight to * it, plus the possibility that the page we were on gets * deleted after we leave it. See nbtree/README for details. */ for (;;) { *bufP = _bt_walk_left(rel, *bufP); /* if we're at end of scan, return failure */ if (*bufP == InvalidBuffer) { ItemPointerSetInvalid(current); so->btso_curbuf = InvalidBuffer; return false; } page = BufferGetPage(*bufP); opaque = (BTPageOpaque) PageGetSpecialPointer(page); /* * Okay, we managed to move left to a non-deleted page. * Done if it's not half-dead and not empty. Else loop * back and do it all again. */ if (!P_IGNORE(opaque)) { maxoff = PageGetMaxOffsetNumber(page); offnum = maxoff; if (!PageIsEmpty(page) && maxoff >= P_FIRSTDATAKEY(opaque)) break; } } } } /* Update scan state */ so->btso_curbuf = *bufP; blkno = BufferGetBlockNumber(*bufP); ItemPointerSet(current, blkno, offnum); return true; }
/* * btvacuumpage --- VACUUM one page * * This processes a single page for btvacuumscan(). In some cases we * must go back and re-examine previously-scanned pages; this routine * recurses when necessary to handle that case. * * blkno is the page to process. orig_blkno is the highest block number * reached by the outer btvacuumscan loop (the same as blkno, unless we * are recursing to re-examine a previous page). */ static void btvacuumpage(BTVacState *vstate, BlockNumber blkno, BlockNumber orig_blkno) { IndexVacuumInfo *info = vstate->info; IndexBulkDeleteResult *stats = vstate->stats; IndexBulkDeleteCallback callback = vstate->callback; void *callback_state = vstate->callback_state; Relation rel = info->index; bool delete_now; BlockNumber recurse_to; Buffer buf; Page page; BTPageOpaque opaque = NULL; restart: delete_now = false; recurse_to = P_NONE; /* call vacuum_delay_point while not holding any buffer lock */ vacuum_delay_point(); /* * We can't use _bt_getbuf() here because it always applies * _bt_checkpage(), which will barf on an all-zero page. We want to * recycle all-zero pages, not fail. Also, we want to use a nondefault * buffer access strategy. */ buf = ReadBufferExtended(rel, MAIN_FORKNUM, blkno, RBM_NORMAL, info->strategy); LockBuffer(buf, BT_READ); page = BufferGetPage(buf); if (!PageIsNew(page)) { _bt_checkpage(rel, buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); } /* * If we are recursing, the only case we want to do anything with is a * live leaf page having the current vacuum cycle ID. Any other state * implies we already saw the page (eg, deleted it as being empty). */ if (blkno != orig_blkno) { if (_bt_page_recyclable(page) || P_IGNORE(opaque) || !P_ISLEAF(opaque) || opaque->btpo_cycleid != vstate->cycleid) { _bt_relbuf(rel, buf); return; } } /* Page is valid, see what to do with it */ if (_bt_page_recyclable(page)) { /* Okay to recycle this page */ RecordFreeIndexPage(rel, blkno); vstate->totFreePages++; stats->pages_deleted++; } else if (P_ISDELETED(opaque)) { /* Already deleted, but can't recycle yet */ stats->pages_deleted++; } else if (P_ISHALFDEAD(opaque)) { /* Half-dead, try to delete */ delete_now = true; } else if (P_ISLEAF(opaque)) { OffsetNumber deletable[MaxOffsetNumber]; int ndeletable; OffsetNumber offnum, minoff, maxoff; /* * Trade in the initial read lock for a super-exclusive write lock on * this page. We must get such a lock on every leaf page over the * course of the vacuum scan, whether or not it actually contains any * deletable tuples --- see nbtree/README. */ LockBuffer(buf, BUFFER_LOCK_UNLOCK); LockBufferForCleanup(buf); /* * Remember highest leaf page number we've taken cleanup lock on; see * notes in btvacuumscan */ if (blkno > vstate->lastBlockLocked) vstate->lastBlockLocked = blkno; /* * Check whether we need to recurse back to earlier pages. What we * are concerned about is a page split that happened since we started * the vacuum scan. If the split moved some tuples to a lower page * then we might have missed 'em. If so, set up for tail recursion. * (Must do this before possibly clearing btpo_cycleid below!) */ if (vstate->cycleid != 0 && opaque->btpo_cycleid == vstate->cycleid && !(opaque->btpo_flags & BTP_SPLIT_END) && !P_RIGHTMOST(opaque) && opaque->btpo_next < orig_blkno) recurse_to = opaque->btpo_next; /* * Scan over all items to see which ones need deleted according to the * callback function. */ ndeletable = 0; minoff = P_FIRSTDATAKEY(opaque); maxoff = PageGetMaxOffsetNumber(page); if (callback) { for (offnum = minoff; offnum <= maxoff; offnum = OffsetNumberNext(offnum)) { IndexTuple itup; ItemPointer htup; itup = (IndexTuple) PageGetItem(page, PageGetItemId(page, offnum)); htup = &(itup->t_tid); /* * During Hot Standby we currently assume that * XLOG_BTREE_VACUUM records do not produce conflicts. That is * only true as long as the callback function depends only * upon whether the index tuple refers to heap tuples removed * in the initial heap scan. When vacuum starts it derives a * value of OldestXmin. Backends taking later snapshots could * have a RecentGlobalXmin with a later xid than the vacuum's * OldestXmin, so it is possible that row versions deleted * after OldestXmin could be marked as killed by other * backends. The callback function *could* look at the index * tuple state in isolation and decide to delete the index * tuple, though currently it does not. If it ever did, we * would need to reconsider whether XLOG_BTREE_VACUUM records * should cause conflicts. If they did cause conflicts they * would be fairly harsh conflicts, since we haven't yet * worked out a way to pass a useful value for * latestRemovedXid on the XLOG_BTREE_VACUUM records. This * applies to *any* type of index that marks index tuples as * killed. */ if (callback(htup, callback_state)) deletable[ndeletable++] = offnum; } } /* * Apply any needed deletes. We issue just one _bt_delitems_vacuum() * call per page, so as to minimize WAL traffic. */ if (ndeletable > 0) { /* * Notice that the issued XLOG_BTREE_VACUUM WAL record includes * all information to the replay code to allow it to get a cleanup * lock on all pages between the previous lastBlockVacuumed and * this page. This ensures that WAL replay locks all leaf pages at * some point, which is important should non-MVCC scans be * requested. This is currently unused on standby, but we record * it anyway, so that the WAL contains the required information. * * Since we can visit leaf pages out-of-order when recursing, * replay might end up locking such pages an extra time, but it * doesn't seem worth the amount of bookkeeping it'd take to avoid * that. */ _bt_delitems_vacuum(rel, buf, deletable, ndeletable, vstate->lastBlockVacuumed); /* * Remember highest leaf page number we've issued a * XLOG_BTREE_VACUUM WAL record for. */ if (blkno > vstate->lastBlockVacuumed) vstate->lastBlockVacuumed = blkno; stats->tuples_removed += ndeletable; /* must recompute maxoff */ maxoff = PageGetMaxOffsetNumber(page); } else { /* * If the page has been split during this vacuum cycle, it seems * worth expending a write to clear btpo_cycleid even if we don't * have any deletions to do. (If we do, _bt_delitems_vacuum takes * care of this.) This ensures we won't process the page again. * * We treat this like a hint-bit update because there's no need to * WAL-log it. */ if (vstate->cycleid != 0 && opaque->btpo_cycleid == vstate->cycleid) { opaque->btpo_cycleid = 0; MarkBufferDirtyHint(buf, true); } } /* * If it's now empty, try to delete; else count the live tuples. We * don't delete when recursing, though, to avoid putting entries into * freePages out-of-order (doesn't seem worth any extra code to handle * the case). */ if (minoff > maxoff) delete_now = (blkno == orig_blkno); else stats->num_index_tuples += maxoff - minoff + 1; } if (delete_now) { MemoryContext oldcontext; int ndel; /* Run pagedel in a temp context to avoid memory leakage */ MemoryContextReset(vstate->pagedelcontext); oldcontext = MemoryContextSwitchTo(vstate->pagedelcontext); ndel = _bt_pagedel(rel, buf); /* count only this page, else may double-count parent */ if (ndel) stats->pages_deleted++; MemoryContextSwitchTo(oldcontext); /* pagedel released buffer, so we shouldn't */ } else _bt_relbuf(rel, buf); /* * This is really tail recursion, but if the compiler is too stupid to * optimize it as such, we'd eat an uncomfortably large amount of stack * space per recursion level (due to the deletable[] array). A failure is * improbable since the number of levels isn't likely to be large ... but * just in case, let's hand-optimize into a loop. */ if (recurse_to != P_NONE) { blkno = recurse_to; goto restart; } }
/* * _bt_walk_left() -- step left one page, if possible * * The given buffer must be pinned and read-locked. This will be dropped * before stepping left. On return, we have pin and read lock on the * returned page, instead. * * Returns InvalidBuffer if there is no page to the left (no lock is held * in that case). * * When working on a non-leaf level, it is possible for the returned page * to be half-dead; the caller should check that condition and step left * again if it's important. */ static Buffer _bt_walk_left(Relation rel, Buffer buf) { Page page; BTPageOpaque opaque; page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); for (;;) { BlockNumber obknum; BlockNumber lblkno; BlockNumber blkno; int tries; /* if we're at end of tree, release buf and return failure */ if (P_LEFTMOST(opaque)) { _bt_relbuf(rel, buf); break; } /* remember original page we are stepping left from */ obknum = BufferGetBlockNumber(buf); /* step left */ blkno = lblkno = opaque->btpo_prev; _bt_relbuf(rel, buf); /* check for interrupts while we're not holding any buffer lock */ CHECK_FOR_INTERRUPTS(); buf = _bt_getbuf(rel, blkno, BT_READ); page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); /* * If this isn't the page we want, walk right till we find what we * want --- but go no more than four hops (an arbitrary limit). If we * don't find the correct page by then, the most likely bet is that * the original page got deleted and isn't in the sibling chain at all * anymore, not that its left sibling got split more than four times. * * Note that it is correct to test P_ISDELETED not P_IGNORE here, * because half-dead pages are still in the sibling chain. Caller * must reject half-dead pages if wanted. */ tries = 0; for (;;) { if (!P_ISDELETED(opaque) && opaque->btpo_next == obknum) { /* Found desired page, return it */ return buf; } if (P_RIGHTMOST(opaque) || ++tries > 4) break; blkno = opaque->btpo_next; buf = _bt_relandgetbuf(rel, buf, blkno, BT_READ); page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); } /* Return to the original page to see what's up */ buf = _bt_relandgetbuf(rel, buf, obknum, BT_READ); page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); if (P_ISDELETED(opaque)) { /* * It was deleted. Move right to first nondeleted page (there * must be one); that is the page that has acquired the deleted * one's keyspace, so stepping left from it will take us where we * want to be. */ for (;;) { if (P_RIGHTMOST(opaque)) elog(ERROR, "fell off the end of index \"%s\"", RelationGetRelationName(rel)); blkno = opaque->btpo_next; buf = _bt_relandgetbuf(rel, buf, blkno, BT_READ); page = BufferGetPage(buf); opaque = (BTPageOpaque) PageGetSpecialPointer(page); if (!P_ISDELETED(opaque)) break; } /* * Now return to top of loop, resetting obknum to point to this * nondeleted page, and try again. */ } else { /* * It wasn't deleted; the explanation had better be that the page * to the left got split or deleted. Without this check, we'd go * into an infinite loop if there's anything wrong. */ if (opaque->btpo_prev == lblkno) elog(ERROR, "could not find left sibling of block %u in index \"%s\"", obknum, RelationGetRelationName(rel)); /* Okay to try again with new lblkno value */ } } return InvalidBuffer; }