aboutsummaryrefslogtreecommitdiff
path: root/src/backend/access/nbtree
diff options
context:
space:
mode:
Diffstat (limited to 'src/backend/access/nbtree')
-rw-r--r--src/backend/access/nbtree/nbtree.c32
-rw-r--r--src/backend/access/nbtree/nbtsearch.c70
-rw-r--r--src/backend/access/nbtree/nbtsort.c2
-rw-r--r--src/backend/access/nbtree/nbtutils.c97
4 files changed, 119 insertions, 82 deletions
diff --git a/src/backend/access/nbtree/nbtree.c b/src/backend/access/nbtree/nbtree.c
index 765659887af..fdff960c130 100644
--- a/src/backend/access/nbtree/nbtree.c
+++ b/src/backend/access/nbtree/nbtree.c
@@ -228,6 +228,8 @@ btgettuple(IndexScanDesc scan, ScanDirection dir)
BTScanOpaque so = (BTScanOpaque) scan->opaque;
bool res;
+ Assert(scan->heapRelation != NULL);
+
/* btree indexes are never lossy */
scan->xs_recheck = false;
@@ -289,6 +291,8 @@ btgetbitmap(IndexScanDesc scan, TIDBitmap *tbm)
int64 ntids = 0;
ItemPointer heapTid;
+ Assert(scan->heapRelation == NULL);
+
/* Each loop iteration performs another primitive index scan */
do
{
@@ -393,6 +397,34 @@ btrescan(IndexScanDesc scan, ScanKey scankey, int nscankeys,
BTScanPosInvalidate(so->currPos);
}
+ /*
+ * We prefer to eagerly drop leaf page pins before btgettuple returns.
+ * This avoids making VACUUM wait to acquire a cleanup lock on the page.
+ *
+ * We cannot safely drop leaf page pins during index-only scans due to a
+ * race condition involving VACUUM setting pages all-visible in the VM.
+ * It's also unsafe for plain index scans that use a non-MVCC snapshot.
+ *
+ * When we drop pins eagerly, the mechanism that marks so->killedItems[]
+ * index tuples LP_DEAD has to deal with concurrent TID recycling races.
+ * The scheme used to detect unsafe TID recycling won't work when scanning
+ * unlogged relations (since it involves saving an affected page's LSN).
+ * Opt out of eager pin dropping during unlogged relation scans for now
+ * (this is preferable to opting out of kill_prior_tuple LP_DEAD setting).
+ *
+ * Also opt out of dropping leaf page pins eagerly during bitmap scans.
+ * Pins cannot be held for more than an instant during bitmap scans either
+ * way, so we might as well avoid wasting cycles on acquiring page LSNs.
+ *
+ * See nbtree/README section on making concurrent TID recycling safe.
+ *
+ * Note: so->dropPin should never change across rescans.
+ */
+ so->dropPin = (!scan->xs_want_itup &&
+ IsMVCCSnapshot(scan->xs_snapshot) &&
+ RelationNeedsWAL(scan->indexRelation) &&
+ scan->heapRelation != NULL);
+
so->markItemIndex = -1;
so->needPrimScan = false;
so->scanBehind = false;
diff --git a/src/backend/access/nbtree/nbtsearch.c b/src/backend/access/nbtree/nbtsearch.c
index fe9a3886913..070f14c8b91 100644
--- a/src/backend/access/nbtree/nbtsearch.c
+++ b/src/backend/access/nbtree/nbtsearch.c
@@ -25,7 +25,7 @@
#include "utils/rel.h"
-static void _bt_drop_lock_and_maybe_pin(IndexScanDesc scan, BTScanPos sp);
+static inline void _bt_drop_lock_and_maybe_pin(Relation rel, BTScanOpaque so);
static Buffer _bt_moveright(Relation rel, Relation heaprel, BTScanInsert key,
Buffer buf, bool forupdate, BTStack stack,
int access);
@@ -57,24 +57,29 @@ static bool _bt_endpoint(IndexScanDesc scan, ScanDirection dir);
/*
* _bt_drop_lock_and_maybe_pin()
*
- * Unlock the buffer; and if it is safe to release the pin, do that, too.
- * This will prevent vacuum from stalling in a blocked state trying to read a
- * page when a cursor is sitting on it.
- *
- * See nbtree/README section on making concurrent TID recycling safe.
+ * Unlock so->currPos.buf. If scan is so->dropPin, drop the pin, too.
+ * Dropping the pin prevents VACUUM from blocking on acquiring a cleanup lock.
*/
-static void
-_bt_drop_lock_and_maybe_pin(IndexScanDesc scan, BTScanPos sp)
+static inline void
+_bt_drop_lock_and_maybe_pin(Relation rel, BTScanOpaque so)
{
- _bt_unlockbuf(scan->indexRelation, sp->buf);
-
- if (IsMVCCSnapshot(scan->xs_snapshot) &&
- RelationNeedsWAL(scan->indexRelation) &&
- !scan->xs_want_itup)
+ if (!so->dropPin)
{
- ReleaseBuffer(sp->buf);
- sp->buf = InvalidBuffer;
+ /* Just drop the lock (not the pin) */
+ _bt_unlockbuf(rel, so->currPos.buf);
+ return;
}
+
+ /*
+ * Drop both the lock and the pin.
+ *
+ * Have to set so->currPos.lsn so that _bt_killitems has a way to detect
+ * when concurrent heap TID recycling by VACUUM might have taken place.
+ */
+ Assert(RelationNeedsWAL(rel));
+ so->currPos.lsn = BufferGetLSNAtomic(so->currPos.buf);
+ _bt_relbuf(rel, so->currPos.buf);
+ so->currPos.buf = InvalidBuffer;
}
/*
@@ -866,8 +871,8 @@ _bt_compare(Relation rel,
* if backwards scan, the last item) in the tree that satisfies the
* qualifications in the scan key. On success exit, data about the
* matching tuple(s) on the page has been loaded into so->currPos. We'll
- * drop all locks and hold onto a pin on page's buffer, except when
- * _bt_drop_lock_and_maybe_pin dropped the pin to avoid blocking VACUUM.
+ * drop all locks and hold onto a pin on page's buffer, except during
+ * so->dropPin scans, when we drop both the lock and the pin.
* _bt_returnitem sets the next item to return to scan on success exit.
*
* If there are no matching items in the index, we return false, with no
@@ -1610,7 +1615,13 @@ _bt_readpage(IndexScanDesc scan, ScanDirection dir, OffsetNumber offnum,
so->currPos.currPage = BufferGetBlockNumber(so->currPos.buf);
so->currPos.prevPage = opaque->btpo_prev;
so->currPos.nextPage = opaque->btpo_next;
+ /* delay setting so->currPos.lsn until _bt_drop_lock_and_maybe_pin */
+ so->currPos.dir = dir;
+ so->currPos.nextTupleOffset = 0;
+ /* either moreRight or moreLeft should be set now (may be unset later) */
+ Assert(ScanDirectionIsForward(dir) ? so->currPos.moreRight :
+ so->currPos.moreLeft);
Assert(!P_IGNORE(opaque));
Assert(BTScanPosIsPinned(so->currPos));
Assert(!so->needPrimScan);
@@ -1626,14 +1637,6 @@ _bt_readpage(IndexScanDesc scan, ScanDirection dir, OffsetNumber offnum,
so->currPos.currPage);
}
- /* initialize remaining currPos fields related to current page */
- so->currPos.lsn = BufferGetLSNAtomic(so->currPos.buf);
- so->currPos.dir = dir;
- so->currPos.nextTupleOffset = 0;
- /* either moreLeft or moreRight should be set now (may be unset later) */
- Assert(ScanDirectionIsForward(dir) ? so->currPos.moreRight :
- so->currPos.moreLeft);
-
PredicateLockPage(rel, so->currPos.currPage, scan->xs_snapshot);
/* initialize local variables */
@@ -2107,10 +2110,9 @@ _bt_returnitem(IndexScanDesc scan, BTScanOpaque so)
*
* Wrapper on _bt_readnextpage that performs final steps for the current page.
*
- * On entry, if so->currPos.buf is valid the buffer is pinned but not locked.
- * If there's no pin held, it's because _bt_drop_lock_and_maybe_pin dropped
- * the pin eagerly earlier on. The scan must have so->currPos.currPage set to
- * a valid block, in any case.
+ * On entry, so->currPos must be valid. Its buffer will be pinned, though
+ * never locked. (Actually, when so->dropPin there won't even be a pin held,
+ * though so->currPos.currPage must still be set to a valid block number.)
*/
static bool
_bt_steppage(IndexScanDesc scan, ScanDirection dir)
@@ -2251,12 +2253,14 @@ _bt_readfirstpage(IndexScanDesc scan, OffsetNumber offnum, ScanDirection dir)
*/
if (_bt_readpage(scan, dir, offnum, true))
{
+ Relation rel = scan->indexRelation;
+
/*
* _bt_readpage succeeded. Drop the lock (and maybe the pin) on
* so->currPos.buf in preparation for btgettuple returning tuples.
*/
Assert(BTScanPosIsPinned(so->currPos));
- _bt_drop_lock_and_maybe_pin(scan, &so->currPos);
+ _bt_drop_lock_and_maybe_pin(rel, so);
return true;
}
@@ -2294,8 +2298,8 @@ _bt_readfirstpage(IndexScanDesc scan, OffsetNumber offnum, ScanDirection dir)
*
* On success exit, so->currPos is updated to contain data from the next
* interesting page, and we return true. We hold a pin on the buffer on
- * success exit, except when _bt_drop_lock_and_maybe_pin decided it was safe
- * to eagerly drop the pin (to avoid blocking VACUUM).
+ * success exit (except during so->dropPin index scans, when we drop the pin
+ * eagerly to avoid blocking VACUUM).
*
* If there are no more matching records in the given direction, we drop all
* locks and pins, invalidate so->currPos, and return false.
@@ -2413,7 +2417,7 @@ _bt_readnextpage(IndexScanDesc scan, BlockNumber blkno,
*/
Assert(so->currPos.currPage == blkno);
Assert(BTScanPosIsPinned(so->currPos));
- _bt_drop_lock_and_maybe_pin(scan, &so->currPos);
+ _bt_drop_lock_and_maybe_pin(rel, so);
return true;
}
diff --git a/src/backend/access/nbtree/nbtsort.c b/src/backend/access/nbtree/nbtsort.c
index 3794cc924ad..9d70e89c1f3 100644
--- a/src/backend/access/nbtree/nbtsort.c
+++ b/src/backend/access/nbtree/nbtsort.c
@@ -105,7 +105,7 @@ typedef struct BTShared
int scantuplesortstates;
/* Query ID, for report in worker processes */
- uint64 queryid;
+ int64 queryid;
/*
* workersdonecv is used to monitor the progress of workers. All parallel
diff --git a/src/backend/access/nbtree/nbtutils.c b/src/backend/access/nbtree/nbtutils.c
index 1a15dfcb7d3..c71d1b6f2e1 100644
--- a/src/backend/access/nbtree/nbtutils.c
+++ b/src/backend/access/nbtree/nbtutils.c
@@ -3330,87 +3330,85 @@ _bt_checkkeys_look_ahead(IndexScanDesc scan, BTReadPageState *pstate,
* current page and killed tuples thereon (generally, this should only be
* called if so->numKilled > 0).
*
- * The caller does not have a lock on the page and may or may not have the
- * page pinned in a buffer. Note that read-lock is sufficient for setting
- * LP_DEAD status (which is only a hint).
+ * Caller should not have a lock on the so->currPos page, but must hold a
+ * buffer pin when !so->dropPin. When we return, it still won't be locked.
+ * It'll continue to hold whatever pins were held before calling here.
*
- * We match items by heap TID before assuming they are the right ones to
- * delete. We cope with cases where items have moved right due to insertions.
- * If an item has moved off the current page due to a split, we'll fail to
- * find it and do nothing (this is not an error case --- we assume the item
- * will eventually get marked in a future indexscan).
+ * We match items by heap TID before assuming they are the right ones to set
+ * LP_DEAD. If the scan is one that holds a buffer pin on the target page
+ * continuously from initially reading the items until applying this function
+ * (if it is a !so->dropPin scan), VACUUM cannot have deleted any items on the
+ * page, so the page's TIDs can't have been recycled by now. There's no risk
+ * that we'll confuse a new index tuple that happens to use a recycled TID
+ * with a now-removed tuple with the same TID (that used to be on this same
+ * page). We can't rely on that during scans that drop buffer pins eagerly
+ * (so->dropPin scans), though, so we must condition setting LP_DEAD bits on
+ * the page LSN having not changed since back when _bt_readpage saw the page.
+ * We totally give up on setting LP_DEAD bits when the page LSN changed.
*
- * Note that if we hold a pin on the target page continuously from initially
- * reading the items until applying this function, VACUUM cannot have deleted
- * any items from the page, and so there is no need to search left from the
- * recorded offset. (This observation also guarantees that the item is still
- * the right one to delete, which might otherwise be questionable since heap
- * TIDs can get recycled.) This holds true even if the page has been modified
- * by inserts and page splits, so there is no need to consult the LSN.
- *
- * If the pin was released after reading the page, then we re-read it. If it
- * has been modified since we read it (as determined by the LSN), we dare not
- * flag any entries because it is possible that the old entry was vacuumed
- * away and the TID was re-used by a completely different heap tuple.
+ * We give up much less often during !so->dropPin scans, but it still happens.
+ * We cope with cases where items have moved right due to insertions. If an
+ * item has moved off the current page due to a split, we'll fail to find it
+ * and just give up on it.
*/
void
_bt_killitems(IndexScanDesc scan)
{
+ Relation rel = scan->indexRelation;
BTScanOpaque so = (BTScanOpaque) scan->opaque;
Page page;
BTPageOpaque opaque;
OffsetNumber minoff;
OffsetNumber maxoff;
- int i;
int numKilled = so->numKilled;
bool killedsomething = false;
- bool droppedpin PG_USED_FOR_ASSERTS_ONLY;
+ Buffer buf;
+ Assert(numKilled > 0);
Assert(BTScanPosIsValid(so->currPos));
+ Assert(scan->heapRelation != NULL); /* can't be a bitmap index scan */
- /*
- * Always reset the scan state, so we don't look for same items on other
- * pages.
- */
+ /* Always invalidate so->killedItems[] before leaving so->currPos */
so->numKilled = 0;
- if (BTScanPosIsPinned(so->currPos))
+ if (!so->dropPin)
{
/*
* We have held the pin on this page since we read the index tuples,
* so all we need to do is lock it. The pin will have prevented
- * re-use of any TID on the page, so there is no need to check the
- * LSN.
+ * concurrent VACUUMs from recycling any of the TIDs on the page.
*/
- droppedpin = false;
- _bt_lockbuf(scan->indexRelation, so->currPos.buf, BT_READ);
-
- page = BufferGetPage(so->currPos.buf);
+ Assert(BTScanPosIsPinned(so->currPos));
+ buf = so->currPos.buf;
+ _bt_lockbuf(rel, buf, BT_READ);
}
else
{
- Buffer buf;
+ XLogRecPtr latestlsn;
- droppedpin = true;
- /* Attempt to re-read the buffer, getting pin and lock. */
- buf = _bt_getbuf(scan->indexRelation, so->currPos.currPage, BT_READ);
+ Assert(!BTScanPosIsPinned(so->currPos));
+ Assert(RelationNeedsWAL(rel));
+ buf = _bt_getbuf(rel, so->currPos.currPage, BT_READ);
- page = BufferGetPage(buf);
- if (BufferGetLSNAtomic(buf) == so->currPos.lsn)
- so->currPos.buf = buf;
- else
+ latestlsn = BufferGetLSNAtomic(buf);
+ Assert(!XLogRecPtrIsInvalid(so->currPos.lsn));
+ Assert(so->currPos.lsn <= latestlsn);
+ if (so->currPos.lsn != latestlsn)
{
- /* Modified while not pinned means hinting is not safe. */
- _bt_relbuf(scan->indexRelation, buf);
+ /* Modified, give up on hinting */
+ _bt_relbuf(rel, buf);
return;
}
+
+ /* Unmodified, hinting is safe */
}
+ page = BufferGetPage(buf);
opaque = BTPageGetOpaque(page);
minoff = P_FIRSTDATAKEY(opaque);
maxoff = PageGetMaxOffsetNumber(page);
- for (i = 0; i < numKilled; i++)
+ for (int i = 0; i < numKilled; i++)
{
int itemIndex = so->killedItems[i];
BTScanPosItem *kitem = &so->currPos.items[itemIndex];
@@ -3442,7 +3440,7 @@ _bt_killitems(IndexScanDesc scan)
* correctness.
*
* Note that the page may have been modified in almost any way
- * since we first read it (in the !droppedpin case), so it's
+ * since we first read it (in the !so->dropPin case), so it's
* possible that this posting list tuple wasn't a posting list
* tuple when we first encountered its heap TIDs.
*/
@@ -3458,7 +3456,7 @@ _bt_killitems(IndexScanDesc scan)
* though only in the common case where the page can't
* have been concurrently modified
*/
- Assert(kitem->indexOffset == offnum || !droppedpin);
+ Assert(kitem->indexOffset == offnum || !so->dropPin);
/*
* Read-ahead to later kitems here.
@@ -3522,10 +3520,13 @@ _bt_killitems(IndexScanDesc scan)
if (killedsomething)
{
opaque->btpo_flags |= BTP_HAS_GARBAGE;
- MarkBufferDirtyHint(so->currPos.buf, true);
+ MarkBufferDirtyHint(buf, true);
}
- _bt_unlockbuf(scan->indexRelation, so->currPos.buf);
+ if (!so->dropPin)
+ _bt_unlockbuf(rel, buf);
+ else
+ _bt_relbuf(rel, buf);
}