/*
* lazy_vacuum_page() -- free dead tuples on a page
* and repair its fragmentation.
*
* Caller must hold pin and buffer cleanup lock on the buffer.
*
* tupindex is the index in vacrelstats->dead_tuples of the first dead
* tuple for this page. We assume the rest follow sequentially.
* The return value is the first tupindex after the tuples of this page.
*/
static int
lazy_vacuum_page(Relation onerel, BlockNumber blkno, Buffer buffer,
int tupindex, LVRelStats *vacrelstats)
{
Page page = BufferGetPage(buffer);
OffsetNumber unused[MaxOffsetNumber];
int uncnt = 0;
MIRROREDLOCK_BUFMGR_MUST_ALREADY_BE_HELD;
START_CRIT_SECTION();
for (; tupindex < vacrelstats->num_dead_tuples; tupindex++)
{
BlockNumber tblk;
OffsetNumber toff;
ItemId itemid;
tblk = ItemPointerGetBlockNumber(&vacrelstats->dead_tuples[tupindex]);
if (tblk != blkno)
break; /* past end of tuples for this block */
toff = ItemPointerGetOffsetNumber(&vacrelstats->dead_tuples[tupindex]);
itemid = PageGetItemId(page, toff);
ItemIdSetUnused(itemid);
unused[uncnt++] = toff;
}
PageRepairFragmentation(page);
MarkBufferDirty(buffer);
/* XLOG stuff */
if (!onerel->rd_istemp)
{
XLogRecPtr recptr;
recptr = log_heap_clean(onerel, buffer,
NULL, 0, NULL, 0,
unused, uncnt,
false);
PageSetLSN(page, recptr);
PageSetTLI(page, ThisTimeLineID);
}
END_CRIT_SECTION();
return tupindex;
}
/*
* Perform the actual page changes needed by heap_page_prune.
* It is expected that the caller has suitable pin and lock on the
* buffer, and is inside a critical section.
*
* This is split out because it is also used by heap_xlog_clean()
* to replay the WAL record when needed after a crash. Note that the
* arguments are identical to those of log_heap_clean().
*/
void
heap_page_prune_execute(Buffer buffer,
OffsetNumber *redirected, int nredirected,
OffsetNumber *nowdead, int ndead,
OffsetNumber *nowunused, int nunused)
{
Page page = (Page) BufferGetPage(buffer);
OffsetNumber *offnum;
int i;
/* Update all redirected line pointers */
offnum = redirected;
for (i = 0; i < nredirected; i++)
{
OffsetNumber fromoff = *offnum++;
OffsetNumber tooff = *offnum++;
ItemId fromlp = PageGetItemId(page, fromoff);
ItemIdSetRedirect(fromlp, tooff);
}
/* Update all now-dead line pointers */
offnum = nowdead;
for (i = 0; i < ndead; i++)
{
OffsetNumber off = *offnum++;
ItemId lp = PageGetItemId(page, off);
ItemIdSetDead(lp);
}
/* Update all now-unused line pointers */
offnum = nowunused;
for (i = 0; i < nunused; i++)
{
OffsetNumber off = *offnum++;
ItemId lp = PageGetItemId(page, off);
ItemIdSetUnused(lp);
}
/*
* Finally, repair any fragmentation, and update the page's hint bit about
* whether it has free pointers.
*/
PageRepairFragmentation(page);
}
/*
* PageIndexTupleDeleteNoCompact
*
* Remove the specified tuple from an index page, but set its line pointer
* to "unused" instead of compacting it out, except that it can be removed
* if it's the last line pointer on the page.
*
* This is used for index AMs that require that existing TIDs of live tuples
* remain unchanged, and are willing to allow unused line pointers instead.
*/
void
PageIndexTupleDeleteNoCompact(Page page, OffsetNumber offnum)
{
PageHeader phdr = (PageHeader) page;
char *addr;
ItemId tup;
Size size;
unsigned offset;
int nline;
/*
* As with PageRepairFragmentation, paranoia seems justified.
*/
if (phdr->pd_lower < SizeOfPageHeaderData ||
phdr->pd_lower > phdr->pd_upper ||
phdr->pd_upper > phdr->pd_special ||
phdr->pd_special > BLCKSZ ||
phdr->pd_special != MAXALIGN(phdr->pd_special))
ereport(ERROR,
(errcode(ERRCODE_DATA_CORRUPTED),
errmsg("corrupted page pointers: lower = %u, upper = %u, special = %u",
phdr->pd_lower, phdr->pd_upper, phdr->pd_special)));
nline = PageGetMaxOffsetNumber(page);
if ((int) offnum <= 0 || (int) offnum > nline)
elog(ERROR, "invalid index offnum: %u", offnum);
tup = PageGetItemId(page, offnum);
Assert(ItemIdHasStorage(tup));
size = ItemIdGetLength(tup);
offset = ItemIdGetOffset(tup);
if (offset < phdr->pd_upper || (offset + size) > phdr->pd_special ||
offset != MAXALIGN(offset))
ereport(ERROR,
(errcode(ERRCODE_DATA_CORRUPTED),
errmsg("corrupted item pointer: offset = %u, size = %u",
offset, (unsigned int) size)));
/* Amount of space to actually be deleted */
size = MAXALIGN(size);
/*
* Either set the item pointer to "unused", or zap it if it's the last
* one. (Note: it's possible that the next-to-last one(s) are already
* unused, but we do not trouble to try to compact them out if so.)
*/
if ((int) offnum < nline)
ItemIdSetUnused(tup);
else
{
phdr->pd_lower -= sizeof(ItemIdData);
nline--; /* there's one less than when we started */
}
/*
* Now move everything between the old upper bound (beginning of tuple
* space) and the beginning of the deleted tuple forward, so that space in
* the middle of the page is left free. If we've just deleted the tuple
* at the beginning of tuple space, then there's no need to do the copy.
*/
/* beginning of tuple space */
addr = (char *) page + phdr->pd_upper;
if (offset > phdr->pd_upper)
memmove(addr + size, addr, offset - phdr->pd_upper);
/* adjust free space boundary pointer */
phdr->pd_upper += size;
/*
* Finally, we need to adjust the linp entries that remain.
*
* Anything that used to be before the deleted tuple's data was moved
* forward by the size of the deleted tuple.
*/
if (!PageIsEmpty(page))
{
int i;
for (i = 1; i <= nline; i++)
{
ItemId ii = PageGetItemId(phdr, i);
if (ItemIdHasStorage(ii) && ItemIdGetOffset(ii) <= offset)
ii->lp_off += size;
}
}
}
/*
* PageRepairFragmentation
*
* Frees fragmented space on a page.
* It doesn't remove unused line pointers! Please don't change this.
*
* This routine is usable for heap pages only, but see PageIndexMultiDelete.
*
* As a side effect, the page's PD_HAS_FREE_LINES hint bit is updated.
*/
void
PageRepairFragmentation(Page page)
{
Offset pd_lower = ((PageHeader) page)->pd_lower;
Offset pd_upper = ((PageHeader) page)->pd_upper;
Offset pd_special = ((PageHeader) page)->pd_special;
ItemId lp;
int nline,
nstorage,
nunused;
int i;
Size totallen;
/*
* It's worth the trouble to be more paranoid here than in most places,
* because we are about to reshuffle data in (what is usually) a shared
* disk buffer. If we aren't careful then corrupted pointers, lengths,
* etc could cause us to clobber adjacent disk buffers, spreading the data
* loss further. So, check everything.
*/
if (pd_lower < SizeOfPageHeaderData ||
pd_lower > pd_upper ||
pd_upper > pd_special ||
pd_special > BLCKSZ ||
pd_special != MAXALIGN(pd_special))
ereport(ERROR,
(errcode(ERRCODE_DATA_CORRUPTED),
errmsg("corrupted page pointers: lower = %u, upper = %u, special = %u",
pd_lower, pd_upper, pd_special)));
nline = PageGetMaxOffsetNumber(page);
nunused = nstorage = 0;
for (i = FirstOffsetNumber; i <= nline; i++)
{
lp = PageGetItemId(page, i);
if (ItemIdIsUsed(lp))
{
if (ItemIdHasStorage(lp))
nstorage++;
}
else
{
/* Unused entries should have lp_len = 0, but make sure */
ItemIdSetUnused(lp);
nunused++;
}
}
if (nstorage == 0)
{
/* Page is completely empty, so just reset it quickly */
((PageHeader) page)->pd_upper = pd_special;
}
else
{
/* Need to compact the page the hard way */
itemIdSortData itemidbase[MaxHeapTuplesPerPage];
itemIdSort itemidptr = itemidbase;
totallen = 0;
for (i = 0; i < nline; i++)
{
lp = PageGetItemId(page, i + 1);
if (ItemIdHasStorage(lp))
{
itemidptr->offsetindex = i;
itemidptr->itemoff = ItemIdGetOffset(lp);
if (itemidptr->itemoff < (int) pd_upper ||
itemidptr->itemoff >= (int) pd_special)
ereport(ERROR,
(errcode(ERRCODE_DATA_CORRUPTED),
errmsg("corrupted item pointer: %u",
itemidptr->itemoff)));
itemidptr->alignedlen = MAXALIGN(ItemIdGetLength(lp));
totallen += itemidptr->alignedlen;
itemidptr++;
}
}
if (totallen > (Size) (pd_special - pd_lower))
ereport(ERROR,
(errcode(ERRCODE_DATA_CORRUPTED),
errmsg("corrupted item lengths: total %u, available space %u",
(unsigned int) totallen, pd_special - pd_lower)));
compactify_tuples(itemidbase, nstorage, page);
}
/* Set hint bit for PageAddItem */
if (nunused > 0)
PageSetHasFreeLinePointers(page);
else
PageClearHasFreeLinePointers(page);
}
/*----------
* Add an item to a disk page from the sort output.
*
* We must be careful to observe the page layout conventions of nbtsearch.c:
* - rightmost pages start data items at P_HIKEY instead of at P_FIRSTKEY.
* - on non-leaf pages, the key portion of the first item need not be
* stored, we should store only the link.
*
* A leaf page being built looks like:
*
* +----------------+---------------------------------+
* | PageHeaderData | linp0 linp1 linp2 ... |
* +-----------+----+---------------------------------+
* | ... linpN | |
* +-----------+--------------------------------------+
* | ^ last |
* | |
* +-------------+------------------------------------+
* | | itemN ... |
* +-------------+------------------+-----------------+
* | ... item3 item2 item1 | "special space" |
* +--------------------------------+-----------------+
*
* Contrast this with the diagram in bufpage.h; note the mismatch
* between linps and items. This is because we reserve linp0 as a
* placeholder for the pointer to the "high key" item; when we have
* filled up the page, we will set linp0 to point to itemN and clear
* linpN. On the other hand, if we find this is the last (rightmost)
* page, we leave the items alone and slide the linp array over.
*
* 'last' pointer indicates the last offset added to the page.
*----------
*/
static void
_bt_buildadd(BTWriteState *wstate, BTPageState *state, IndexTuple itup)
{
Page npage;
BlockNumber nblkno;
OffsetNumber last_off;
Size pgspc;
Size itupsz;
/*
* This is a handy place to check for cancel interrupts during the btree
* load phase of index creation.
*/
CHECK_FOR_INTERRUPTS();
npage = state->btps_page;
nblkno = state->btps_blkno;
last_off = state->btps_lastoff;
pgspc = PageGetFreeSpace(npage);
itupsz = IndexTupleDSize(*itup);
itupsz = MAXALIGN(itupsz);
/*
* Check whether the item can fit on a btree page at all. (Eventually, we
* ought to try to apply TOAST methods if not.) We actually need to be
* able to fit three items on every page, so restrict any one item to 1/3
* the per-page available space. Note that at this point, itupsz doesn't
* include the ItemId.
*
* NOTE: similar code appears in _bt_insertonpg() to defend against
* oversize items being inserted into an already-existing index. But
* during creation of an index, we don't go through there.
*/
if (itupsz > BTMaxItemSize(npage))
ereport(ERROR,
(errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
errmsg("index row size %zu exceeds maximum %zu for index \"%s\"",
itupsz, BTMaxItemSize(npage),
RelationGetRelationName(wstate->index)),
errhint("Values larger than 1/3 of a buffer page cannot be indexed.\n"
"Consider a function index of an MD5 hash of the value, "
"or use full text indexing."),
errtableconstraint(wstate->heap,
RelationGetRelationName(wstate->index))));
/*
* Check to see if page is "full". It's definitely full if the item won't
* fit. Otherwise, compare to the target freespace derived from the
* fillfactor. However, we must put at least two items on each page, so
* disregard fillfactor if we don't have that many.
*/
if (pgspc < itupsz || (pgspc < state->btps_full && last_off > P_FIRSTKEY))
{
/*
* Finish off the page and write it out.
*/
Page opage = npage;
BlockNumber oblkno = nblkno;
ItemId ii;
ItemId hii;
IndexTuple oitup;
/* Create new page of same level */
npage = _bt_blnewpage(state->btps_level);
/* and assign it a page position */
nblkno = wstate->btws_pages_alloced++;
/*
* We copy the last item on the page into the new page, and then
* rearrange the old page so that the 'last item' becomes its high key
* rather than a true data item. There had better be at least two
* items on the page already, else the page would be empty of useful
* data.
*/
Assert(last_off > P_FIRSTKEY);
ii = PageGetItemId(opage, last_off);
oitup = (IndexTuple) PageGetItem(opage, ii);
_bt_sortaddtup(npage, ItemIdGetLength(ii), oitup, P_FIRSTKEY);
/*
* Move 'last' into the high key position on opage
*/
hii = PageGetItemId(opage, P_HIKEY);
*hii = *ii;
ItemIdSetUnused(ii); /* redundant */
((PageHeader) opage)->pd_lower -= sizeof(ItemIdData);
/*
* Link the old page into its parent, using its minimum key. If we
* don't have a parent, we have to create one; this adds a new btree
* level.
*/
if (state->btps_next == NULL)
state->btps_next = _bt_pagestate(wstate, state->btps_level + 1);
Assert(state->btps_minkey != NULL);
ItemPointerSet(&(state->btps_minkey->t_tid), oblkno, P_HIKEY);
_bt_buildadd(wstate, state->btps_next, state->btps_minkey);
pfree(state->btps_minkey);
/*
* Save a copy of the minimum key for the new page. We have to copy
* it off the old page, not the new one, in case we are not at leaf
* level.
*/
state->btps_minkey = CopyIndexTuple(oitup);
/*
* Set the sibling links for both pages.
*/
{
BTPageOpaque oopaque = (BTPageOpaque) PageGetSpecialPointer(opage);
BTPageOpaque nopaque = (BTPageOpaque) PageGetSpecialPointer(npage);
oopaque->btpo_next = nblkno;
nopaque->btpo_prev = oblkno;
nopaque->btpo_next = P_NONE; /* redundant */
}
/*
* Write out the old page. We never need to touch it again, so we can
* free the opage workspace too.
*/
_bt_blwritepage(wstate, opage, oblkno);
/*
* Reset last_off to point to new page
*/
last_off = P_FIRSTKEY;
}
/*
* If the new item is the first for its page, stash a copy for later. Note
* this will only happen for the first item on a level; on later pages,
* the first item for a page is copied from the prior page in the code
* above.
*/
if (last_off == P_HIKEY)
{
Assert(state->btps_minkey == NULL);
state->btps_minkey = CopyIndexTuple(itup);
}
/*
* Add the new item into the current page.
*/
last_off = OffsetNumberNext(last_off);
_bt_sortaddtup(npage, itupsz, itup, last_off);
state->btps_page = npage;
state->btps_blkno = nblkno;
state->btps_lastoff = last_off;
}
/*
* Perform the actual page changes needed by heap_page_prune.
* It is expected that the caller has suitable pin and lock on the
* buffer, and is inside a critical section.
*
* This is split out because it is also used by heap_xlog_clean()
* to replay the WAL record when needed after a crash. Note that the
* arguments are identical to those of log_heap_clean().
*/
void
heap_page_prune_execute(Buffer buffer,
OffsetNumber *redirected, int nredirected,
OffsetNumber *nowdead, int ndead,
OffsetNumber *nowunused, int nunused,
bool redirect_move)
{
Page page = (Page) BufferGetPage(buffer);
OffsetNumber *offnum;
int i;
/* Update all redirected or moved line pointers */
offnum = redirected;
for (i = 0; i < nredirected; i++)
{
OffsetNumber fromoff = *offnum++;
OffsetNumber tooff = *offnum++;
ItemId fromlp = PageGetItemId(page, fromoff);
if (redirect_move)
{
/* Physically move the "to" item to the "from" slot */
ItemId tolp = PageGetItemId(page, tooff);
HeapTupleHeader htup;
*fromlp = *tolp;
ItemIdSetUnused(tolp);
/*
* Change heap-only status of the tuple because after the line
* pointer manipulation, it's no longer a heap-only tuple, but is
* directly pointed to by index entries.
*/
Assert(ItemIdIsNormal(fromlp));
htup = (HeapTupleHeader) PageGetItem(page, fromlp);
Assert(HeapTupleHeaderIsHeapOnly(htup));
HeapTupleHeaderClearHeapOnly(htup);
}
else
{
/* Just insert a REDIRECT link at fromoff */
ItemIdSetRedirect(fromlp, tooff);
}
}
/* Update all now-dead line pointers */
offnum = nowdead;
for (i = 0; i < ndead; i++)
{
OffsetNumber off = *offnum++;
ItemId lp = PageGetItemId(page, off);
ItemIdSetDead(lp);
}
/* Update all now-unused line pointers */
offnum = nowunused;
for (i = 0; i < nunused; i++)
{
OffsetNumber off = *offnum++;
ItemId lp = PageGetItemId(page, off);
ItemIdSetUnused(lp);
}
/*
* Finally, repair any fragmentation, and update the page's hint bit about
* whether it has free pointers.
*/
PageRepairFragmentation(page);
}
/*
* PageIndexDeleteNoCompact
* Delete the given items for an index page, and defragment the resulting
* free space, but do not compact the item pointers array.
*
* itemnos is the array of tuples to delete; nitems is its size. maxIdxTuples
* is the maximum number of tuples that can exist in a page.
*
* Unused items at the end of the array are removed.
*
* This is used for index AMs that require that existing TIDs of live tuples
* remain unchanged.
*/
void
PageIndexDeleteNoCompact(Page page, OffsetNumber *itemnos, int nitems)
{
PageHeader phdr = (PageHeader) page;
LocationIndex pd_lower = phdr->pd_lower;
LocationIndex pd_upper = phdr->pd_upper;
LocationIndex pd_special = phdr->pd_special;
int nline;
bool empty;
OffsetNumber offnum;
int nextitm;
/*
* As with PageRepairFragmentation, paranoia seems justified.
*/
if (pd_lower < SizeOfPageHeaderData ||
pd_lower > pd_upper ||
pd_upper > pd_special ||
pd_special > BLCKSZ ||
pd_special != MAXALIGN(pd_special))
ereport(ERROR,
(errcode(ERRCODE_DATA_CORRUPTED),
errmsg("corrupted page pointers: lower = %u, upper = %u, special = %u",
pd_lower, pd_upper, pd_special)));
/*
* Scan the existing item pointer array and mark as unused those that are
* in our kill-list; make sure any non-interesting ones are marked unused
* as well.
*/
nline = PageGetMaxOffsetNumber(page);
empty = true;
nextitm = 0;
for (offnum = FirstOffsetNumber; offnum <= nline; offnum = OffsetNumberNext(offnum))
{
ItemId lp;
ItemLength itemlen;
ItemOffset offset;
lp = PageGetItemId(page, offnum);
itemlen = ItemIdGetLength(lp);
offset = ItemIdGetOffset(lp);
if (ItemIdIsUsed(lp))
{
if (offset < pd_upper ||
(offset + itemlen) > pd_special ||
offset != MAXALIGN(offset))
ereport(ERROR,
(errcode(ERRCODE_DATA_CORRUPTED),
errmsg("corrupted item pointer: offset = %u, length = %u",
offset, (unsigned int) itemlen)));
if (nextitm < nitems && offnum == itemnos[nextitm])
{
/* this one is on our list to delete, so mark it unused */
ItemIdSetUnused(lp);
nextitm++;
}
else if (ItemIdHasStorage(lp))
{
/* This one's live -- must do the compaction dance */
empty = false;
}
else
{
/* get rid of this one too */
ItemIdSetUnused(lp);
}
}
}
/* this will catch invalid or out-of-order itemnos[] */
if (nextitm != nitems)
elog(ERROR, "incorrect index offsets supplied");
if (empty)
{
/* Page is completely empty, so just reset it quickly */
phdr->pd_lower = SizeOfPageHeaderData;
phdr->pd_upper = pd_special;
}
else
{
/* There are live items: need to compact the page the hard way */
itemIdSortData itemidbase[MaxOffsetNumber];
itemIdSort itemidptr;
int i;
Size totallen;
/*
* Scan the page taking note of each item that we need to preserve.
* This includes both live items (those that contain data) and
* interspersed unused ones. It's critical to preserve these unused
* items, because otherwise the offset numbers for later live items
* would change, which is not acceptable. Unused items might get used
* again later; that is fine.
*/
itemidptr = itemidbase;
totallen = 0;
PageClearHasFreeLinePointers(page);
for (i = 0; i < nline; i++)
{
ItemId lp;
itemidptr->offsetindex = i;
lp = PageGetItemId(page, i + 1);
if (ItemIdHasStorage(lp))
{
itemidptr->itemoff = ItemIdGetOffset(lp);
itemidptr->alignedlen = MAXALIGN(ItemIdGetLength(lp));
totallen += itemidptr->alignedlen;
itemidptr++;
}
else
{
PageSetHasFreeLinePointers(page);
ItemIdSetUnused(lp);
}
}
nline = itemidptr - itemidbase;
/* By here, there are exactly nline elements in itemidbase array */
if (totallen > (Size) (pd_special - pd_lower))
ereport(ERROR,
(errcode(ERRCODE_DATA_CORRUPTED),
errmsg("corrupted item lengths: total %u, available space %u",
(unsigned int) totallen, pd_special - pd_lower)));
/*
* Defragment the data areas of each tuple, being careful to preserve
* each item's position in the linp array.
*/
compactify_tuples(itemidbase, nline, page);
}
}