database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-11-10 17:42:29 +03:00
Code Activity

Reimplement hash index locking algorithms, per my recent proposal to

Browse Source 
pghackers.  This fixes the problem recently reported by Markus KrÌutner
(hash bucket split corrupts the state of scans being done concurrently),
and I believe it also fixes all the known problems with deadlocks in
hash index operations.  Hash indexes are still not really ready for prime
time (since they aren't WAL-logged), but this is a step forward.
This commit is contained in:
Tom Lane
2003-09-04 22:06:27 +00:00
parent ca43f71ca5
commit 7a3693716d
11 changed files with 1052 additions and 1036 deletions
Download Patch File Download Diff File Expand all files Collapse all files

254
src/backend/access/hash/hashinsert.c
View File

@@ -8,7 +8,7 @@
*
*
* IDENTIFICATION
* $Header: /cvsroot/pgsql/src/backend/access/hash/hashinsert.c,v 1.29 2003/09/02 18:13:30 tgl Exp $
* $Header: /cvsroot/pgsql/src/backend/access/hash/hashinsert.c,v 1.30 2003/09/04 22:06:27 tgl Exp $
*
*-------------------------------------------------------------------------
*/
@@ -16,136 +16,124 @@
#include "postgres.h"
#include "access/hash.h"
#include "storage/lmgr.h"
static OffsetNumber _hash_pgaddtup(Relation rel, Buffer buf,
Size itemsize, HashItem hitem);
static InsertIndexResult _hash_insertonpg(Relation rel, Buffer buf, int keysz, ScanKey scankey, HashItem hitem, Buffer metabuf);
static OffsetNumber _hash_pgaddtup(Relation rel, Buffer buf, int keysz, ScanKey itup_scankey, Size itemsize, HashItem hitem);
/*
* _hash_doinsert() -- Handle insertion of a single HashItem in the table.
*
* This routine is called by the public interface routines, hashbuild
* and hashinsert. By here, hashitem is filled in, and has a unique
* (xid, seqno) pair. The datum to be used as a "key" is in the
* hashitem.
* and hashinsert. By here, hashitem is completely filled in.
* The datum to be used as a "key" is in the hashitem.
*/
InsertIndexResult
_hash_doinsert(Relation rel, HashItem hitem)
{
Buffer buf;
Buffer metabuf;
BlockNumber blkno;
HashMetaPage metap;
IndexTuple itup;
BlockNumber itup_blkno;
OffsetNumber itup_off;
InsertIndexResult res;
ScanKey itup_scankey;
int natts;
BlockNumber blkno;
Page page;
HashPageOpaque pageopaque;
Size itemsz;
bool do_expand;
uint32 hashkey;
Bucket bucket;
Datum datum;
bool isnull;
/*
* Compute the hash key for the item. We do this first so as not to
* need to hold any locks while running the hash function.
*/
itup = &(hitem->hash_itup);
if (rel->rd_rel->relnatts != 1)
elog(ERROR, "hash indexes support only one index key");
datum = index_getattr(itup, 1, RelationGetDescr(rel), &isnull);
Assert(!isnull);
hashkey = _hash_datum2hashkey(rel, datum);
/* compute item size too */
itemsz = IndexTupleDSize(hitem->hash_itup)
+ (sizeof(HashItemData) - sizeof(IndexTupleData));
itemsz = MAXALIGN(itemsz); /* be safe, PageAddItem will do this but
* we need to be consistent */
/*
* Acquire shared split lock so we can compute the target bucket
* safely (see README).
*/
_hash_getlock(rel, 0, HASH_SHARE);
/* Read the metapage */
metabuf = _hash_getbuf(rel, HASH_METAPAGE, HASH_READ);
metap = (HashMetaPage) BufferGetPage(metabuf);
_hash_checkpage(rel, (Page) metap, LH_META_PAGE);
/* we need a scan key to do our search, so build one */
itup = &(hitem->hash_itup);
if ((natts = rel->rd_rel->relnatts) != 1)
elog(ERROR, "Hash indexes support only one index key");
itup_scankey = _hash_mkscankey(rel, itup);
/*
* Check whether the item can fit on a hash page at all. (Eventually,
* we ought to try to apply TOAST methods if not.) Note that at this
* point, itemsz doesn't include the ItemId.
*/
if (itemsz > HashMaxItemSize((Page) metap))
ereport(ERROR,
(errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
errmsg("index tuple size %lu exceeds hash maximum, %lu",
(unsigned long) itemsz,
(unsigned long) HashMaxItemSize((Page) metap))));
/*
* find the first page in the bucket chain containing this key and
* place it in buf. _hash_search obtains a read lock for us.
* Compute the target bucket number, and convert to block number.
*/
_hash_search(rel, natts, itup_scankey, &buf, metap);
bucket = _hash_hashkey2bucket(hashkey,
metap->hashm_maxbucket,
metap->hashm_highmask,
metap->hashm_lowmask);
blkno = BUCKET_TO_BLKNO(metap, bucket);
/* release lock on metapage, but keep pin since we'll need it again */
_hash_chgbufaccess(rel, metabuf, HASH_READ, HASH_NOLOCK);
/*
* Acquire share lock on target bucket; then we can release split lock.
*/
_hash_getlock(rel, blkno, HASH_SHARE);
_hash_droplock(rel, 0, HASH_SHARE);
/* Fetch the primary bucket page for the bucket */
buf = _hash_getbuf(rel, blkno, HASH_WRITE);
page = BufferGetPage(buf);
_hash_checkpage(rel, page, LH_BUCKET_PAGE);
/*
* trade in our read lock for a write lock so that we can do the
* insertion.
*/
blkno = BufferGetBlockNumber(buf);
_hash_relbuf(rel, buf, HASH_READ);
buf = _hash_getbuf(rel, blkno, HASH_WRITE);
/*
* XXX btree comment (haven't decided what to do in hash): don't think
* the bucket can be split while we're reading the metapage.
*
* If the page was split between the time that we surrendered our read
* lock and acquired our write lock, then this page may no longer be
* the right place for the key we want to insert.
*/
/* do the insertion */
res = _hash_insertonpg(rel, buf, natts, itup_scankey,
hitem, metabuf);
/* be tidy */
_hash_freeskey(itup_scankey);
return res;
}
/*
* _hash_insertonpg() -- Insert a tuple on a particular page in the table.
*
* This recursive procedure does the following things:
*
* + if necessary, splits the target page.
* + inserts the tuple.
*
* On entry, we must have the right buffer on which to do the
* insertion, and the buffer must be pinned and locked. On return,
* we will have dropped both the pin and the write lock on the buffer.
*
*/
static InsertIndexResult
_hash_insertonpg(Relation rel,
Buffer buf,
int keysz,
ScanKey scankey,
HashItem hitem,
Buffer metabuf)
{
InsertIndexResult res;
Page page;
BlockNumber itup_blkno;
OffsetNumber itup_off;
Size itemsz;
HashPageOpaque pageopaque;
bool do_expand = false;
Buffer ovflbuf;
HashMetaPage metap;
Bucket bucket;
metap = (HashMetaPage) BufferGetPage(metabuf);
_hash_checkpage(rel, (Page) metap, LH_META_PAGE);
page = BufferGetPage(buf);
_hash_checkpage(rel, page, LH_BUCKET_PAGE | LH_OVERFLOW_PAGE);
pageopaque = (HashPageOpaque) PageGetSpecialPointer(page);
bucket = pageopaque->hasho_bucket;
itemsz = IndexTupleDSize(hitem->hash_itup)
+ (sizeof(HashItemData) - sizeof(IndexTupleData));
itemsz = MAXALIGN(itemsz);
Assert(pageopaque->hasho_bucket == bucket);
/* Do the insertion */
while (PageGetFreeSpace(page) < itemsz)
{
/*
* no space on this page; check for an overflow page
*/
if (BlockNumberIsValid(pageopaque->hasho_nextblkno))
BlockNumber nextblkno = pageopaque->hasho_nextblkno;
if (BlockNumberIsValid(nextblkno))
{
/*
* ovfl page exists; go get it. if it doesn't have room,
* we'll find out next pass through the loop test above.
*/
ovflbuf = _hash_getbuf(rel, pageopaque->hasho_nextblkno,
HASH_WRITE);
_hash_relbuf(rel, buf, HASH_WRITE);
buf = ovflbuf;
_hash_relbuf(rel, buf);
buf = _hash_getbuf(rel, nextblkno, HASH_WRITE);
page = BufferGetPage(buf);
}
else
@@ -154,65 +142,72 @@ _hash_insertonpg(Relation rel,
* we're at the end of the bucket chain and we haven't found a
* page with enough room. allocate a new overflow page.
*/
do_expand = true;
ovflbuf = _hash_addovflpage(rel, metabuf, buf);
_hash_relbuf(rel, buf, HASH_WRITE);
buf = ovflbuf;
/* release our write lock without modifying buffer */
_hash_chgbufaccess(rel, buf, HASH_READ, HASH_NOLOCK);
/* chain to a new overflow page */
buf = _hash_addovflpage(rel, metabuf, buf);
page = BufferGetPage(buf);
if (PageGetFreeSpace(page) < itemsz)
{
/* it doesn't fit on an empty page -- give up */
elog(ERROR, "hash item too large");
}
/* should fit now, given test above */
Assert(PageGetFreeSpace(page) >= itemsz);
}
_hash_checkpage(rel, page, LH_OVERFLOW_PAGE);
pageopaque = (HashPageOpaque) PageGetSpecialPointer(page);
Assert(pageopaque->hasho_bucket == bucket);
}
itup_off = _hash_pgaddtup(rel, buf, keysz, scankey, itemsz, hitem);
/* found page with enough space, so add the item here */
itup_off = _hash_pgaddtup(rel, buf, itemsz, hitem);
itup_blkno = BufferGetBlockNumber(buf);
/* by here, the new tuple is inserted */
/* write and release the modified page */
_hash_wrtbuf(rel, buf);
/* We can drop the bucket lock now */
_hash_droplock(rel, blkno, HASH_SHARE);
/*
* Write-lock the metapage so we can increment the tuple count.
* After incrementing it, check to see if it's time for a split.
*/
_hash_chgbufaccess(rel, metabuf, HASH_NOLOCK, HASH_WRITE);
metap->hashm_ntuples += 1;
/* Make sure this stays in sync with _hash_expandtable() */
do_expand = metap->hashm_ntuples >
(double) metap->hashm_ffactor * (metap->hashm_maxbucket + 1);
/* Write out the metapage and drop lock, but keep pin */
_hash_chgbufaccess(rel, metabuf, HASH_WRITE, HASH_NOLOCK);
/* Attempt to split if a split is needed */
if (do_expand)
_hash_expandtable(rel, metabuf);
/* Finally drop our pin on the metapage */
_hash_dropbuf(rel, metabuf);
/* Create the return data structure */
res = (InsertIndexResult) palloc(sizeof(InsertIndexResultData));
ItemPointerSet(&(res->pointerData), itup_blkno, itup_off);
if (res != NULL)
{
/*
* Increment the number of keys in the table. We switch lock
* access type just for a moment to allow greater accessibility to
* the metapage.
*/
_hash_chgbufaccess(rel, metabuf, HASH_READ, HASH_WRITE);
metap->hashm_ntuples += 1;
_hash_chgbufaccess(rel, metabuf, HASH_WRITE, HASH_READ);
}
_hash_wrtbuf(rel, buf);
if (do_expand ||
(metap->hashm_ntuples / (metap->hashm_maxbucket + 1))
> (double) metap->hashm_ffactor)
_hash_expandtable(rel, metabuf);
_hash_relbuf(rel, metabuf, HASH_READ);
return res;
}
/*
* _hash_pgaddtup() -- add a tuple to a particular page in the index.
*
* This routine adds the tuple to the page as requested, and keeps the
* write lock and reference associated with the page's buffer. It is
* an error to call pgaddtup() without a write lock and reference.
* This routine adds the tuple to the page as requested; it does
* not write out the page. It is an error to call pgaddtup() without
* a write lock and pin.
*/
static OffsetNumber
_hash_pgaddtup(Relation rel,
Buffer buf,
int keysz,
ScanKey itup_scankey,
Size itemsize,
HashItem hitem)
{
@@ -228,8 +223,5 @@ _hash_pgaddtup(Relation rel,
elog(ERROR, "failed to add index item to \"%s\"",
RelationGetRelationName(rel));
/* write the buffer, but hold our lock */
_hash_wrtnorelbuf(buf);
return itup_off;
}