/*
* PageGetTempPage
* Get a temporary page in local memory for special processing
*/
Page
PageGetTempPage(Page page, Size specialSize)
{
Size pageSize;
Size size;
Page temp;
PageHeader thdr;
pageSize = PageGetPageSize(page);
temp = (Page) palloc(pageSize);
thdr = (PageHeader) temp;
/* copy old page in */
memcpy(temp, page, pageSize);
/* clear out the middle */
size = pageSize - SizeOfPageHeaderData;
size -= MAXALIGN(specialSize);
MemSet(PageGetContents(thdr), 0, size);
/* set high, low water marks */
thdr->pd_lower = SizeOfPageHeaderData;
thdr->pd_upper = pageSize - MAXALIGN(specialSize);
return temp;
}
/* -------------------------------------------------
* GetHashPageStatistics()
*
* Collect statistics of single hash page
* -------------------------------------------------
*/
static void
GetHashPageStatistics(Page page, HashPageStat * stat)
{
OffsetNumber maxoff = PageGetMaxOffsetNumber(page);
HashPageOpaque opaque = (HashPageOpaque) PageGetSpecialPointer(page);
int off;
stat->dead_items = stat->live_items = 0;
stat->page_size = PageGetPageSize(page);
/* hash page opaque data */
stat->hasho_prevblkno = opaque->hasho_prevblkno;
stat->hasho_nextblkno = opaque->hasho_nextblkno;
stat->hasho_bucket = opaque->hasho_bucket;
stat->hasho_flag = opaque->hasho_flag;
stat->hasho_page_id = opaque->hasho_page_id;
/* count live and dead tuples, and free space */
for (off = FirstOffsetNumber; off <= maxoff; off++)
{
ItemId id = PageGetItemId(page, off);
if (!ItemIdIsDead(id))
stat->live_items++;
else
stat->dead_items++;
}
stat->free_size = PageGetFreeSpace(page);
}
static bool
parse_page(const DataPage *page,
XLogRecPtr *lsn, uint16 *offset, uint16 *length)
{
const PageHeaderData *page_data = &page->page_data;
/* Get lsn from page header */
*lsn = PageXLogRecPtrGet(page_data->pd_lsn);
if (PageGetPageSize(page_data) == BLCKSZ &&
PageGetPageLayoutVersion(page_data) == PG_PAGE_LAYOUT_VERSION &&
(page_data->pd_flags & ~PD_VALID_FLAG_BITS) == 0 &&
page_data->pd_lower >= SizeOfPageHeaderData &&
page_data->pd_lower <= page_data->pd_upper &&
page_data->pd_upper <= page_data->pd_special &&
page_data->pd_special <= BLCKSZ &&
page_data->pd_special == MAXALIGN(page_data->pd_special) &&
!XLogRecPtrIsInvalid(*lsn))
{
*offset = page_data->pd_lower;
*length = page_data->pd_upper - page_data->pd_lower;
return true;
}
*offset = *length = 0;
return false;
}
/*
* PageHeaderIsValid
* Check that the header fields of a page appear valid.
*
* This is called when a page has just been read in from disk. The idea is
* to cheaply detect trashed pages before we go nuts following bogus item
* pointers, testing invalid transaction identifiers, etc.
*
* It turns out to be necessary to allow zeroed pages here too. Even though
* this routine is *not* called when deliberately adding a page to a relation,
* there are scenarios in which a zeroed page might be found in a table.
* (Example: a backend extends a relation, then crashes before it can write
* any WAL entry about the new page. The kernel will already have the
* zeroed page in the file, and it will stay that way after restart.) So we
* allow zeroed pages here, and are careful that the page access macros
* treat such a page as empty and without free space. Eventually, VACUUM
* will clean up such a page and make it usable.
*/
bool
PageHeaderIsValid(PageHeader page)
{
char *pagebytes;
int i;
/* Check normal case */
if (PageGetPageSize(page) == BLCKSZ &&
PageGetPageLayoutVersion(page) == PG_PAGE_LAYOUT_VERSION &&
(page->pd_flags & ~PD_VALID_FLAG_BITS) == 0 &&
page->pd_lower >= SizeOfPageHeaderData &&
page->pd_lower <= page->pd_upper &&
page->pd_upper <= page->pd_special &&
page->pd_special <= BLCKSZ &&
page->pd_special == MAXALIGN(page->pd_special))
return true;
/* Check all-zeroes case */
pagebytes = (char *) page;
for (i = 0; i < BLCKSZ; i++)
{
if (pagebytes[i] != 0)
return false;
}
return true;
}
/*
* PageRestoreTempPage
* Copy temporary page back to permanent page after special processing
* and release the temporary page.
*/
void
PageRestoreTempPage(Page tempPage, Page oldPage)
{
Size pageSize;
pageSize = PageGetPageSize(tempPage);
memcpy((char *) oldPage, (char *) tempPage, pageSize);
pfree(tempPage);
}
static Page
GistPageGetCopyPage(Page page)
{
Size pageSize = PageGetPageSize(page);
Page tmppage;
tmppage = (Page) palloc(pageSize);
memcpy(tmppage, page, pageSize);
return tmppage;
}
/*
* PageGetTempPage
* Get a temporary page in local memory for special processing.
* The returned page is not initialized at all; caller must do that.
*/
Page
PageGetTempPage(Page page)
{
Size pageSize;
Page temp;
pageSize = PageGetPageSize(page);
temp = (Page) palloc(pageSize);
return temp;
}
/*
* PageGetTempPageCopy
* Get a temporary page in local memory for special processing.
* The page is initialized by copying the contents of the given page.
*/
Page
PageGetTempPageCopy(Page page)
{
Size pageSize;
Page temp;
pageSize = PageGetPageSize(page);
temp = (Page) palloc(pageSize);
memcpy(temp, page, pageSize);
return temp;
}
/*
* PageGetTempPageCopySpecial
* Get a temporary page in local memory for special processing.
* The page is PageInit'd with the same special-space size as the
* given page, and the special space is copied from the given page.
*/
Page
PageGetTempPageCopySpecial(Page page)
{
Size pageSize;
Page temp;
pageSize = PageGetPageSize(page);
temp = (Page) palloc(pageSize);
PageInit(temp, pageSize, PageGetSpecialSize(page));
memcpy(PageGetSpecialPointer(temp),
PageGetSpecialPointer(page),
PageGetSpecialSize(page));
return temp;
}
/*
* _hash_checkpage -- sanity checks on the format of all hash pages
*/
void
_hash_checkpage(Relation rel, Page page, int flags)
{
Assert(page);
/*
* When checking the metapage, always verify magic number and version.
*/
if (flags == LH_META_PAGE)
{
HashMetaPage metap = (HashMetaPage) page;
if (metap->hashm_magic != HASH_MAGIC)
ereport(ERROR,
(errcode(ERRCODE_INDEX_CORRUPTED),
errmsg("index \"%s\" is not a hash index",
RelationGetRelationName(rel))));
if (metap->hashm_version != HASH_VERSION)
ereport(ERROR,
(errcode(ERRCODE_INDEX_CORRUPTED),
errmsg("index \"%s\" has wrong hash version", RelationGetRelationName(rel)),
errhint("Please REINDEX it.")));
}
/*
* These other checks are for debugging purposes only.
*/
#ifdef USE_ASSERT_CHECKING
Assert(((PageHeader) (page))->pd_lower >= SizeOfPageHeaderData);
Assert(((PageHeader) (page))->pd_upper <=
(BLCKSZ - MAXALIGN(sizeof(HashPageOpaqueData))));
Assert(((PageHeader) (page))->pd_special ==
(BLCKSZ - MAXALIGN(sizeof(HashPageOpaqueData))));
Assert(PageGetPageSize(page) == BLCKSZ);
if (flags)
{
HashPageOpaque opaque = (HashPageOpaque) PageGetSpecialPointer(page);
Assert(opaque->hasho_flag & flags);
}
#endif /* USE_ASSERT_CHECKING */
}
Datum
page_header(PG_FUNCTION_ARGS)
{
bytea *raw_page = PG_GETARG_BYTEA_P(0);
int raw_page_size;
TupleDesc tupdesc;
Datum result;
HeapTuple tuple;
Datum values[9];
bool nulls[9];
PageHeader page;
XLogRecPtr lsn;
char lsnchar[64];
if (!superuser())
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
(errmsg("must be superuser to use raw page functions"))));
raw_page_size = VARSIZE(raw_page) - VARHDRSZ;
/*
* Check that enough data was supplied, so that we don't try to access
* fields outside the supplied buffer.
*/
if (raw_page_size < sizeof(PageHeaderData))
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("input page too small (%d bytes)", raw_page_size)));
page = (PageHeader) VARDATA(raw_page);
/* Build a tuple descriptor for our result type */
if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
elog(ERROR, "return type must be a row type");
/* Extract information from the page header */
lsn = PageGetLSN(page);
snprintf(lsnchar, sizeof(lsnchar), "%X/%X", lsn.xlogid, lsn.xrecoff);
values[0] = CStringGetTextDatum(lsnchar);
values[1] = UInt16GetDatum(PageGetTLI(page));
values[2] = UInt16GetDatum(page->pd_flags);
values[3] = UInt16GetDatum(page->pd_lower);
values[4] = UInt16GetDatum(page->pd_upper);
values[5] = UInt16GetDatum(page->pd_special);
values[6] = UInt16GetDatum(PageGetPageSize(page));
values[7] = UInt16GetDatum(PageGetPageLayoutVersion(page));
values[8] = TransactionIdGetDatum(page->pd_prune_xid);
/* Build and return the tuple. */
memset(nulls, 0, sizeof(nulls));
tuple = heap_form_tuple(tupdesc, values, nulls);
result = HeapTupleGetDatum(tuple);
PG_RETURN_DATUM(result);
}
/*
* split page and fills WAL record. original buffer(lbuf) leaves untouched,
* returns shadow page of lbuf filled new data. In leaf page and build mode puts all
* ItemPointers to pages. Also, in build mode splits data by way to full fulled
* left page
*/
static Page
dataSplitPage(GinBtree btree, Buffer lbuf, Buffer rbuf, OffsetNumber off,
void *insertdata, BlockNumber updateblkno, XLogRecData **prdata)
{
char *ptr;
OffsetNumber separator;
ItemPointer bound;
Page lpage = PageGetTempPageCopy(BufferGetPage(lbuf));
bool isleaf = GinPageIsLeaf(lpage);
ItemPointerData oldbound = *GinDataPageGetRightBound(lpage);
int sizeofitem = GinSizeOfDataPageItem(lpage);
OffsetNumber maxoff = GinPageGetOpaque(lpage)->maxoff;
Page rpage = BufferGetPage(rbuf);
Size pageSize = PageGetPageSize(lpage);
Size freeSpace;
/* these must be static so they can be returned to caller */
static ginxlogSplitData data;
static XLogRecData rdata[2];
static char vector[2 * BLCKSZ];
GinInitPage(rpage, GinPageGetOpaque(lpage)->flags, pageSize);
freeSpace = GinDataPageGetFreeSpace(rpage);
*prdata = rdata;
/* Update existing downlink to point to next page (on internal page) */
if (!isleaf)
{
PostingItem *pitem = GinDataPageGetPostingItem(lpage, off);
PostingItemSetBlockNumber(pitem, updateblkno);
}
if (isleaf)
{
memcpy(vector,
GinDataPageGetItemPointer(lpage, FirstOffsetNumber),
maxoff * sizeof(ItemPointerData));
}
else
{
memcpy(vector,
GinDataPageGetPostingItem(lpage, FirstOffsetNumber),
maxoff * sizeof(PostingItem));
}
if (isleaf && GinPageRightMost(lpage) && off > GinPageGetOpaque(lpage)->maxoff)
{
/* append new items to the end */
GinBtreeDataLeafInsertData *items = insertdata;
while (items->curitem < items->nitem &&
maxoff * sizeof(ItemPointerData) < 2 * (freeSpace - sizeof(ItemPointerData)))
{
memcpy(vector + maxoff * sizeof(ItemPointerData),
items->items + items->curitem,
sizeof(ItemPointerData));
maxoff++;
items->curitem++;
}
}
else
{
ptr = vector + (off - 1) * sizeofitem;
if (maxoff + 1 - off != 0)
memmove(ptr + sizeofitem, ptr, (maxoff - off + 1) * sizeofitem);
if (isleaf)
{
GinBtreeDataLeafInsertData *items = insertdata;
memcpy(ptr, items->items + items->curitem, sizeofitem);
items->curitem++;
}
else
{
PostingItem *pitem = insertdata;
memcpy(ptr, pitem, sizeofitem);
}
maxoff++;
}
/*
* we assume that during index creation the table scanned from beginning
* to end, so ItemPointers are in monotonically increasing order.
*/
if (btree->isBuild && GinPageRightMost(lpage))
separator = freeSpace / sizeofitem;
else
separator = maxoff / 2;
GinInitPage(rpage, GinPageGetOpaque(lpage)->flags, pageSize);
GinInitPage(lpage, GinPageGetOpaque(rpage)->flags, pageSize);
if (isleaf)
memcpy(GinDataPageGetItemPointer(lpage, FirstOffsetNumber),
vector, separator * sizeof(ItemPointerData));
else
memcpy(GinDataPageGetPostingItem(lpage, FirstOffsetNumber),
vector, separator * sizeof(PostingItem));
GinPageGetOpaque(lpage)->maxoff = separator;
if (isleaf)
memcpy(GinDataPageGetItemPointer(rpage, FirstOffsetNumber),
vector + separator * sizeof(ItemPointerData),
(maxoff - separator) * sizeof(ItemPointerData));
else
memcpy(GinDataPageGetPostingItem(rpage, FirstOffsetNumber),
vector + separator * sizeof(PostingItem),
(maxoff - separator) * sizeof(PostingItem));
GinPageGetOpaque(rpage)->maxoff = maxoff - separator;
/* set up right bound for left page */
bound = GinDataPageGetRightBound(lpage);
if (GinPageIsLeaf(lpage))
*bound = *GinDataPageGetItemPointer(lpage,
GinPageGetOpaque(lpage)->maxoff);
else
*bound = GinDataPageGetPostingItem(lpage,
GinPageGetOpaque(lpage)->maxoff)->key;
/* set up right bound for right page */
bound = GinDataPageGetRightBound(rpage);
*bound = oldbound;
data.separator = separator;
data.nitem = maxoff;
data.rightbound = oldbound;
rdata[0].buffer = InvalidBuffer;
rdata[0].data = (char *) &data;
rdata[0].len = sizeof(ginxlogSplitData);
rdata[0].next = &rdata[1];
rdata[1].buffer = InvalidBuffer;
rdata[1].data = vector;
rdata[1].len = maxoff * sizeofitem;
rdata[1].next = NULL;
return lpage;
}
/*
* Split entry page and insert new data.
*
* Returns new temp pages to *newlpage and *newrpage.
* The original buffer is left untouched.
*/
static void
entrySplitPage(GinBtree btree, Buffer origbuf,
GinBtreeStack *stack,
GinBtreeEntryInsertData *insertData,
BlockNumber updateblkno,
Page *newlpage, Page *newrpage)
{
OffsetNumber off = stack->off;
OffsetNumber i,
maxoff,
separator = InvalidOffsetNumber;
Size totalsize = 0;
Size lsize = 0,
size;
char *ptr;
IndexTuple itup;
Page page;
Page lpage = PageGetTempPageCopy(BufferGetPage(origbuf));
Page rpage = PageGetTempPageCopy(BufferGetPage(origbuf));
Size pageSize = PageGetPageSize(lpage);
char tupstore[2 * BLCKSZ];
entryPreparePage(btree, lpage, off, insertData, updateblkno);
/*
* First, append all the existing tuples and the new tuple we're inserting
* one after another in a temporary workspace.
*/
maxoff = PageGetMaxOffsetNumber(lpage);
ptr = tupstore;
for (i = FirstOffsetNumber; i <= maxoff; i++)
{
if (i == off)
{
size = MAXALIGN(IndexTupleSize(insertData->entry));
memcpy(ptr, insertData->entry, size);
ptr += size;
totalsize += size + sizeof(ItemIdData);
}
itup = (IndexTuple) PageGetItem(lpage, PageGetItemId(lpage, i));
size = MAXALIGN(IndexTupleSize(itup));
memcpy(ptr, itup, size);
ptr += size;
totalsize += size + sizeof(ItemIdData);
}
if (off == maxoff + 1)
{
size = MAXALIGN(IndexTupleSize(insertData->entry));
memcpy(ptr, insertData->entry, size);
ptr += size;
totalsize += size + sizeof(ItemIdData);
}
/*
* Initialize the left and right pages, and copy all the tuples back to
* them.
*/
GinInitPage(rpage, GinPageGetOpaque(lpage)->flags, pageSize);
GinInitPage(lpage, GinPageGetOpaque(rpage)->flags, pageSize);
ptr = tupstore;
maxoff++;
lsize = 0;
page = lpage;
for (i = FirstOffsetNumber; i <= maxoff; i++)
{
itup = (IndexTuple) ptr;
/*
* Decide where to split. We try to equalize the pages' total data
* size, not number of tuples.
*/
if (lsize > totalsize / 2)
{
if (separator == InvalidOffsetNumber)
separator = i - 1;
page = rpage;
}
else
{
lsize += MAXALIGN(IndexTupleSize(itup)) + sizeof(ItemIdData);
}
if (PageAddItem(page, (Item) itup, IndexTupleSize(itup), InvalidOffsetNumber, false, false) == InvalidOffsetNumber)
elog(ERROR, "failed to add item to index page in \"%s\"",
RelationGetRelationName(btree->index));
ptr += MAXALIGN(IndexTupleSize(itup));
}
/* return temp pages to caller */
*newlpage = lpage;
*newrpage = rpage;
}
/*
* Place tuple and split page, original buffer(lbuf) leaves untouched,
* returns shadow pages filled with new data.
* Tuples are distributed between pages by equal size on its, not
* an equal number!
*/
static void
entrySplitPage(GinBtree btree, Buffer origbuf,
GinBtreeStack *stack,
void *insertPayload,
BlockNumber updateblkno, XLogRecData **prdata,
Page *newlpage, Page *newrpage)
{
GinBtreeEntryInsertData *insertData = insertPayload;
OffsetNumber off = stack->off;
OffsetNumber i,
maxoff,
separator = InvalidOffsetNumber;
Size totalsize = 0;
Size tupstoresize;
Size lsize = 0,
size;
char *ptr;
IndexTuple itup;
Page page;
Page lpage = PageGetTempPageCopy(BufferGetPage(origbuf));
Page rpage = PageGetTempPageCopy(BufferGetPage(origbuf));
Size pageSize = PageGetPageSize(lpage);
/* these must be static so they can be returned to caller */
static XLogRecData rdata[2];
static ginxlogSplitEntry data;
static char tupstore[2 * BLCKSZ];
*prdata = rdata;
entryPreparePage(btree, lpage, off, insertData, updateblkno);
/*
* First, append all the existing tuples and the new tuple we're inserting
* one after another in a temporary workspace.
*/
maxoff = PageGetMaxOffsetNumber(lpage);
ptr = tupstore;
for (i = FirstOffsetNumber; i <= maxoff; i++)
{
if (i == off)
{
size = MAXALIGN(IndexTupleSize(insertData->entry));
memcpy(ptr, insertData->entry, size);
ptr += size;
totalsize += size + sizeof(ItemIdData);
}
itup = (IndexTuple) PageGetItem(lpage, PageGetItemId(lpage, i));
size = MAXALIGN(IndexTupleSize(itup));
memcpy(ptr, itup, size);
ptr += size;
totalsize += size + sizeof(ItemIdData);
}
if (off == maxoff + 1)
{
size = MAXALIGN(IndexTupleSize(insertData->entry));
memcpy(ptr, insertData->entry, size);
ptr += size;
totalsize += size + sizeof(ItemIdData);
}
tupstoresize = ptr - tupstore;
/*
* Initialize the left and right pages, and copy all the tuples back to
* them.
*/
GinInitPage(rpage, GinPageGetOpaque(lpage)->flags, pageSize);
GinInitPage(lpage, GinPageGetOpaque(rpage)->flags, pageSize);
ptr = tupstore;
maxoff++;
lsize = 0;
page = lpage;
for (i = FirstOffsetNumber; i <= maxoff; i++)
{
itup = (IndexTuple) ptr;
if (lsize > totalsize / 2)
{
if (separator == InvalidOffsetNumber)
separator = i - 1;
page = rpage;
}
else
{
lsize += MAXALIGN(IndexTupleSize(itup)) + sizeof(ItemIdData);
}
if (PageAddItem(page, (Item) itup, IndexTupleSize(itup), InvalidOffsetNumber, false, false) == InvalidOffsetNumber)
elog(ERROR, "failed to add item to index page in \"%s\"",
RelationGetRelationName(btree->index));
ptr += MAXALIGN(IndexTupleSize(itup));
}
data.separator = separator;
data.nitem = maxoff;
rdata[0].buffer = InvalidBuffer;
rdata[0].data = (char *) &data;
rdata[0].len = sizeof(ginxlogSplitEntry);
rdata[0].next = &rdata[1];
rdata[1].buffer = InvalidBuffer;
rdata[1].data = tupstore;
rdata[1].len = tupstoresize;
rdata[1].next = NULL;
*newlpage = lpage;
*newrpage = rpage;
}
/* -------------------------------------------------
* GetBTPageStatistics()
*
* Collect statistics of single b-tree page
* -------------------------------------------------
*/
static void
GetBTPageStatistics(BlockNumber blkno, Buffer buffer, BTPageStat *stat)
{
Page page = BufferGetPage(buffer);
PageHeader phdr = (PageHeader) page;
OffsetNumber maxoff = PageGetMaxOffsetNumber(page);
BTPageOpaque opaque = (BTPageOpaque) PageGetSpecialPointer(page);
int item_size = 0;
int off;
stat->blkno = blkno;
stat->max_avail = BLCKSZ - (BLCKSZ - phdr->pd_special + SizeOfPageHeaderData);
stat->dead_items = stat->live_items = 0;
stat->page_size = PageGetPageSize(page);
/* page type (flags) */
if (P_ISDELETED(opaque))
{
stat->type = 'd';
stat->btpo.xact = opaque->btpo.xact;
return;
}
else if (P_IGNORE(opaque))
stat->type = 'e';
else if (P_ISLEAF(opaque))
stat->type = 'l';
else if (P_ISROOT(opaque))
stat->type = 'r';
else
stat->type = 'i';
/* btpage opaque data */
stat->btpo_prev = opaque->btpo_prev;
stat->btpo_next = opaque->btpo_next;
stat->btpo.level = opaque->btpo.level;
stat->btpo_flags = opaque->btpo_flags;
stat->btpo_cycleid = opaque->btpo_cycleid;
/* count live and dead tuples, and free space */
for (off = FirstOffsetNumber; off <= maxoff; off++)
{
IndexTuple itup;
ItemId id = PageGetItemId(page, off);
itup = (IndexTuple) PageGetItem(page, id);
item_size += IndexTupleSize(itup);
if (!ItemIdIsDead(id))
stat->live_items++;
else
stat->dead_items++;
}
stat->free_size = PageGetFreeSpace(page);
if ((stat->live_items + stat->dead_items) > 0)
stat->avg_item_size = item_size / (stat->live_items + stat->dead_items);
else
stat->avg_item_size = 0;
}
static const char *
convert_gpdb4_heap_page(char *page)
{
VERSION4_PageHeaderData *oldhdr;
PageHeader newhdr;
OffsetNumber off;
OffsetNumber maxoff;
if (PageGetPageSize(page) != BLCKSZ)
return "invalid block size on page";
/* Can only convert from GPDB4 format */
if (PageGetPageLayoutVersion(page) != 4)
return "invalid page version";
oldhdr = (VERSION4_PageHeaderData *) page;
/* Other checks that PageHeaderIsValid() normally performs */
if (!((oldhdr->pd_flags & ~PD_VALID_FLAG_BITS) == 0 &&
oldhdr->pd_lower >= SizeOfPageHeaderData &&
oldhdr->pd_lower <= oldhdr->pd_upper &&
oldhdr->pd_upper <= oldhdr->pd_special &&
oldhdr->pd_special <= BLCKSZ &&
oldhdr->pd_special == MAXALIGN(oldhdr->pd_special)))
return "invalid header";
/*
* Ok, it's a valid heap page, from GPDB4. We know how to convert that!
*/
/* 1. pd_prune_xid was added to page header */
/* 2. Line pointer flags were changed */
/* 3. HEAP_COMPRESSED flag was removed */
/* 4. On-disk representation of numerics was changed */
/*
* Also, money datatype was widened from 32 to 64 bits. (pg_upgrade
* should've refused the upgrade)
*/
/*
* First, check if there is enough space on the page, after we expand the header.
*/
oldhdr = (VERSION4_PageHeaderData *) page;
newhdr = (PageHeader) page;
/*
* If there isn't enough space on this page for the new header field, relocate a tuple
*/
make_room(page);
/*
* There is space. Move the line pointers. We also convert the line pointer flags
* while we're at it. Begin from end to beginning, so that we don't overwrite items
* we haven't processed yet.
*/
maxoff = (oldhdr->pd_lower - VERSION4_SizeOfPageHeaderData) / sizeof(ItemIdData); /* PageGetMaxOffsetNumber */
for (off = maxoff; off >= 1; off--)
{
ItemIdData iid = oldhdr->pd_linp[off - 1]; /* PageGetItemId */
if (iid.lp_flags == VERSION4_LP_UNUSED)
iid.lp_flags = LP_UNUSED;
else if (iid.lp_flags == VERSION4_LP_USED)
iid.lp_flags = LP_NORMAL;
else
{
/* LP_DELETE and LP_USED were never used on heap pages. */
return "unexpected LP_DELETE or LP_DEAD line pointer on old-format heap page";
}
newhdr->pd_linp[off - 1] = iid;
}
newhdr->pd_lower = (char *) &newhdr->pd_linp[maxoff] - (char *) page;
/* Initialize the field that was added after version 4 format */
newhdr->pd_prune_xid = 0;
/*
* Ok, the page header and line pointers are in the new format now. Mangle
* the tuples themselves
*/
for (off = 1; off <= maxoff; off++)
{
ItemId iid = PageGetItemId(page, off); /* we can use PageGetItemId now */
HeapTupleHeader htup;
if (!ItemIdIsNormal(iid))
continue;
htup = (HeapTupleHeader) PageGetItem(page, iid);
convert_heaptuple(htup);
}
/*
* Finally, change the version number.
*/
PageSetPageSizeAndVersion(page, BLCKSZ, TARGET_PAGE_VERSION);
return NULL;
}
/*
* Place tuple and split page, original buffer(lbuf) leaves untouched,
* returns shadow page of lbuf filled new data.
* Tuples are distributed between pages by equal size on its, not
* an equal number!
*/
static Page
entrySplitPage(GinBtree btree, Buffer lbuf, Buffer rbuf, OffsetNumber off, XLogRecData **prdata)
{
static XLogRecData rdata[2];
OffsetNumber i,
maxoff,
separator = InvalidOffsetNumber;
Size totalsize = 0;
Size lsize = 0,
size;
static char tupstore[2 * BLCKSZ];
char *ptr;
IndexTuple itup,
leftrightmost = NULL;
static ginxlogSplit data;
Page page;
Page lpage = PageGetTempPageCopy(BufferGetPage(lbuf));
Page rpage = BufferGetPage(rbuf);
Size pageSize = PageGetPageSize(lpage);
*prdata = rdata;
data.leftChildBlkno = (GinPageIsLeaf(lpage)) ?
InvalidOffsetNumber : GinItemPointerGetBlockNumber(&(btree->entry->t_tid));
data.updateBlkno = entryPreparePage(btree, lpage, off);
maxoff = PageGetMaxOffsetNumber(lpage);
ptr = tupstore;
for (i = FirstOffsetNumber; i <= maxoff; i++)
{
if (i == off)
{
size = MAXALIGN(IndexTupleSize(btree->entry));
memcpy(ptr, btree->entry, size);
ptr += size;
totalsize += size + sizeof(ItemIdData);
}
itup = (IndexTuple) PageGetItem(lpage, PageGetItemId(lpage, i));
size = MAXALIGN(IndexTupleSize(itup));
memcpy(ptr, itup, size);
ptr += size;
totalsize += size + sizeof(ItemIdData);
}
if (off == maxoff + 1)
{
size = MAXALIGN(IndexTupleSize(btree->entry));
memcpy(ptr, btree->entry, size);
ptr += size;
totalsize += size + sizeof(ItemIdData);
}
GinInitPage(rpage, GinPageGetOpaque(lpage)->flags, pageSize);
GinInitPage(lpage, GinPageGetOpaque(rpage)->flags, pageSize);
ptr = tupstore;
maxoff++;
lsize = 0;
page = lpage;
for (i = FirstOffsetNumber; i <= maxoff; i++)
{
itup = (IndexTuple) ptr;
if (lsize > totalsize / 2)
{
if (separator == InvalidOffsetNumber)
separator = i - 1;
page = rpage;
}
else
{
leftrightmost = itup;
lsize += MAXALIGN(IndexTupleSize(itup)) + sizeof(ItemIdData);
}
if (PageAddItem(page, (Item) itup, IndexTupleSize(itup), InvalidOffsetNumber, false, false) == InvalidOffsetNumber)
elog(ERROR, "failed to add item to index page in \"%s\"",
RelationGetRelationName(btree->index));
ptr += MAXALIGN(IndexTupleSize(itup));
}
btree->entry = copyIndexTuple(leftrightmost, lpage);
ItemPointerSet(&(btree->entry)->t_tid, BufferGetBlockNumber(lbuf), InvalidOffsetNumber);
btree->rightblkno = BufferGetBlockNumber(rbuf);
data.node = btree->index->rd_node;
data.rootBlkno = InvalidBlockNumber;
data.lblkno = BufferGetBlockNumber(lbuf);
data.rblkno = BufferGetBlockNumber(rbuf);
data.separator = separator;
data.nitem = maxoff;
data.isData = FALSE;
data.isLeaf = GinPageIsLeaf(lpage) ? TRUE : FALSE;
data.isRootSplit = FALSE;
rdata[0].buffer = InvalidBuffer;
rdata[0].data = (char *) &data;
rdata[0].len = sizeof(ginxlogSplit);
rdata[0].next = &rdata[1];
rdata[1].buffer = InvalidBuffer;
rdata[1].data = tupstore;
rdata[1].len = MAXALIGN(totalsize);
rdata[1].next = NULL;
return lpage;
}
/*
* split page and fills WAL record. original buffer(lbuf) leaves untouched,
* returns shadow page of lbuf filled new data. In leaf page and build mode puts all
* ItemPointers to pages. Also, in build mode splits data by way to full fulled
* left page
*/
static Page
dataSplitPage(GinBtree btree, Buffer lbuf, Buffer rbuf, OffsetNumber off, XLogRecData **prdata)
{
char *ptr;
OffsetNumber separator;
ItemPointer bound;
Page lpage = PageGetTempPageCopy(BufferGetPage(lbuf));
ItemPointerData oldbound = *GinDataPageGetRightBound(lpage);
int sizeofitem = GinSizeOfDataPageItem(lpage);
OffsetNumber maxoff = GinPageGetOpaque(lpage)->maxoff;
Page rpage = BufferGetPage(rbuf);
Size pageSize = PageGetPageSize(lpage);
Size freeSpace;
uint32 nCopied = 1;
/* these must be static so they can be returned to caller */
static ginxlogSplit data;
static XLogRecData rdata[4];
static char vector[2 * BLCKSZ];
GinInitPage(rpage, GinPageGetOpaque(lpage)->flags, pageSize);
freeSpace = GinDataPageGetFreeSpace(rpage);
*prdata = rdata;
data.leftChildBlkno = (GinPageIsLeaf(lpage)) ?
InvalidOffsetNumber : PostingItemGetBlockNumber(&(btree->pitem));
data.updateBlkno = dataPrepareData(btree, lpage, off);
memcpy(vector, GinDataPageGetItem(lpage, FirstOffsetNumber),
maxoff * sizeofitem);
if (GinPageIsLeaf(lpage) && GinPageRightMost(lpage) && off > GinPageGetOpaque(lpage)->maxoff)
{
nCopied = 0;
while (btree->curitem < btree->nitem &&
maxoff * sizeof(ItemPointerData) < 2 * (freeSpace - sizeof(ItemPointerData)))
{
memcpy(vector + maxoff * sizeof(ItemPointerData),
btree->items + btree->curitem,
sizeof(ItemPointerData));
maxoff++;
nCopied++;
btree->curitem++;
}
}
else
{
ptr = vector + (off - 1) * sizeofitem;
if (maxoff + 1 - off != 0)
memmove(ptr + sizeofitem, ptr, (maxoff - off + 1) * sizeofitem);
if (GinPageIsLeaf(lpage))
{
memcpy(ptr, btree->items + btree->curitem, sizeofitem);
btree->curitem++;
}
else
memcpy(ptr, &(btree->pitem), sizeofitem);
maxoff++;
}
/*
* we suppose that during index creation table scaned from begin to end,
* so ItemPointers are monotonically increased..
*/
if (btree->isBuild && GinPageRightMost(lpage))
separator = freeSpace / sizeofitem;
else
separator = maxoff / 2;
GinInitPage(rpage, GinPageGetOpaque(lpage)->flags, pageSize);
GinInitPage(lpage, GinPageGetOpaque(rpage)->flags, pageSize);
memcpy(GinDataPageGetItem(lpage, FirstOffsetNumber), vector, separator * sizeofitem);
GinPageGetOpaque(lpage)->maxoff = separator;
memcpy(GinDataPageGetItem(rpage, FirstOffsetNumber),
vector + separator * sizeofitem, (maxoff - separator) * sizeofitem);
GinPageGetOpaque(rpage)->maxoff = maxoff - separator;
PostingItemSetBlockNumber(&(btree->pitem), BufferGetBlockNumber(lbuf));
if (GinPageIsLeaf(lpage))
btree->pitem.key = *(ItemPointerData *) GinDataPageGetItem(lpage,
GinPageGetOpaque(lpage)->maxoff);
else
btree->pitem.key = ((PostingItem *) GinDataPageGetItem(lpage,
GinPageGetOpaque(lpage)->maxoff))->key;
btree->rightblkno = BufferGetBlockNumber(rbuf);
/* set up right bound for left page */
bound = GinDataPageGetRightBound(lpage);
*bound = btree->pitem.key;
/* set up right bound for right page */
bound = GinDataPageGetRightBound(rpage);
*bound = oldbound;
data.node = btree->index->rd_node;
data.rootBlkno = InvalidBlockNumber;
data.lblkno = BufferGetBlockNumber(lbuf);
data.rblkno = BufferGetBlockNumber(rbuf);
data.separator = separator;
data.nitem = maxoff;
data.isData = TRUE;
data.isLeaf = GinPageIsLeaf(lpage) ? TRUE : FALSE;
data.isRootSplit = FALSE;
data.rightbound = oldbound;
rdata[0].buffer = InvalidBuffer;
rdata[0].data = (char *) &data;
rdata[0].len = sizeof(ginxlogSplit);
rdata[0].next = &rdata[1];
rdata[1].buffer = InvalidBuffer;
rdata[1].data = vector;
rdata[1].len = MAXALIGN(maxoff * sizeofitem);
rdata[1].next = NULL;
return lpage;
}
/*
* Place tuple and split page, original buffer(lbuf) leaves untouched,
* returns shadow page of lbuf filled new data.
* Tuples are distributed between pages by equal size on its, not
* an equal number!
*/
static Page
entrySplitPage(RumBtree btree, Buffer lbuf, Buffer rbuf,
Page lPage, Page rPage, OffsetNumber off)
{
OffsetNumber i,
maxoff,
separator = InvalidOffsetNumber;
Size totalsize = 0;
Size lsize = 0,
size;
char *ptr;
IndexTuple itup,
leftrightmost = NULL;
Page page;
Page newlPage = PageGetTempPageCopy(lPage);
Size pageSize = PageGetPageSize(newlPage);
static char tupstore[2 * BLCKSZ];
entryPreparePage(btree, newlPage, off);
maxoff = PageGetMaxOffsetNumber(newlPage);
ptr = tupstore;
for (i = FirstOffsetNumber; i <= maxoff; i++)
{
if (i == off)
{
size = MAXALIGN(IndexTupleSize(btree->entry));
memcpy(ptr, btree->entry, size);
ptr += size;
totalsize += size + sizeof(ItemIdData);
}
itup = (IndexTuple) PageGetItem(newlPage, PageGetItemId(newlPage, i));
size = MAXALIGN(IndexTupleSize(itup));
memcpy(ptr, itup, size);
ptr += size;
totalsize += size + sizeof(ItemIdData);
}
if (off == maxoff + 1)
{
size = MAXALIGN(IndexTupleSize(btree->entry));
memcpy(ptr, btree->entry, size);
totalsize += size + sizeof(ItemIdData);
}
RumInitPage(rPage, RumPageGetOpaque(newlPage)->flags, pageSize);
RumInitPage(newlPage, RumPageGetOpaque(rPage)->flags, pageSize);
ptr = tupstore;
maxoff++;
lsize = 0;
page = newlPage;
for (i = FirstOffsetNumber; i <= maxoff; i++)
{
itup = (IndexTuple) ptr;
if (lsize > totalsize / 2)
{
if (separator == InvalidOffsetNumber)
separator = i - 1;
page = rPage;
}
else
{
leftrightmost = itup;
lsize += MAXALIGN(IndexTupleSize(itup)) + sizeof(ItemIdData);
}
if (PageAddItem(page, (Item) itup, IndexTupleSize(itup), InvalidOffsetNumber, false, false) == InvalidOffsetNumber)
elog(ERROR, "failed to add item to index page in \"%s\"",
RelationGetRelationName(btree->index));
ptr += MAXALIGN(IndexTupleSize(itup));
}
btree->entry = RumFormInteriorTuple(btree, leftrightmost, newlPage,
BufferGetBlockNumber(lbuf));
btree->rightblkno = BufferGetBlockNumber(rbuf);
return newlPage;
}
