Som improve page split in multicolumn GiST index.

If user picksplit on n-th column generate equals
left and right unions then it calls picksplit on n+1-th
column.

diff --git a/src/backend/access/gist/gist.c b/src/backend/access/gist/gist.c
index 54ac45e..326e87e 100644 (file)
--- a/src/backend/access/gist/gist.c
+++ b/src/backend/access/gist/gist.c
@@ -8,7 +8,7 @@
  * Portions Copyright (c) 1994, Regents of the University of California
  *
  * IDENTIFICATION
- *       $PostgreSQL: pgsql/src/backend/access/gist/gist.c,v 1.137 2006/05/24 11:01:39 teodor Exp $
+ *       $PostgreSQL: pgsql/src/backend/access/gist/gist.c,v 1.138 2006/05/29 12:50:06 teodor Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -1033,7 +1033,13 @@ gistSplitByKey(Relation r, Page page, IndexTuple *itup, int len, GISTSTATE *gist
                /*
                 * all keys are not-null
                 */
-               gistUserPicksplit(r, entryvec, attno, v, itup, len, giststate);
+               if ( gistUserPicksplit(r, entryvec, attno, v, itup, len, giststate) && attno+1 != r->rd_att->natts )
+                       /*
+                        * Splitting on attno column is not optimized: unions of left and right
+                        * page are the same, we will try to split page by 
+                        * following columns
+                        */
+                       gistSplitByKey(r, page, itup, len, giststate, v, entryvec, attno+1);
        }
 }
 

diff --git a/src/backend/access/gist/gistutil.c b/src/backend/access/gist/gistutil.c
index 3db72aa..987aed8 100644 (file)
--- a/src/backend/access/gist/gistutil.c
+++ b/src/backend/access/gist/gistutil.c
@@ -8,7 +8,7 @@
  * Portions Copyright (c) 1994, Regents of the University of California
  *
  * IDENTIFICATION
- *                     $PostgreSQL: pgsql/src/backend/access/gist/gistutil.c,v 1.14 2006/05/24 11:01:39 teodor Exp $
+ *                     $PostgreSQL: pgsql/src/backend/access/gist/gistutil.c,v 1.15 2006/05/29 12:50:06 teodor Exp $
  *-------------------------------------------------------------------------
  */
 #include "postgres.h"
@@ -262,6 +262,16 @@ gistMakeUnionKey( GISTSTATE *giststate, int attno,
        }
 }
 
+static bool
+gistKeyIsEQ(GISTSTATE *giststate, int attno, Datum a, Datum b) {
+       bool result;
+
+       FunctionCall3(&giststate->equalFn[attno],
+                                                               a, b,
+                                                               PointerGetDatum(&result));
+       return result;
+}
+
 /*
  * Forms union of oldtup and addtup, if union == oldtup then return NULL
  */
@@ -300,19 +310,8 @@ gistgetadjusted(Relation r, IndexTuple oldtup, IndexTuple addtup, GISTSTATE *gis
                        continue;
 
                if ( !addisnull[i] ) {
-                       if ( oldisnull[i] )
+                       if ( oldisnull[i] || gistKeyIsEQ(giststate, i, oldentries[i].key, attrS[i])==false )
                                neednew = true;
-                       else {
-                               bool        result;
-
-                               FunctionCall3(&giststate->equalFn[i],
-                                                               oldentries[i].key,
-                                                               attrS[i],
-                                                               PointerGetDatum(&result));
-
-                               if (!result)
-                                       neednew = true;
-                       }
                }
        }
 
@@ -395,7 +394,6 @@ gistfindgroup(GISTSTATE *giststate, GISTENTRY *valvec, GIST_SPLITVEC *spl, int a
        {
                int                     j;
                int                     len;
-               bool            result;
 
                if (spl->spl_idgrp[spl->spl_left[i]])
                        continue;
@@ -405,11 +403,7 @@ gistfindgroup(GISTSTATE *giststate, GISTENTRY *valvec, GIST_SPLITVEC *spl, int a
                {
                        if (spl->spl_idgrp[spl->spl_right[j]])
                                continue;
-                       FunctionCall3(&giststate->equalFn[attno],
-                                                 valvec[spl->spl_left[i]].key,
-                                                 valvec[spl->spl_right[j]].key,
-                                                 PointerGetDatum(&result));
-                       if (result)
+                       if (gistKeyIsEQ(giststate, attno, valvec[spl->spl_left[i]].key, valvec[spl->spl_right[j]].key))
                        {
                                spl->spl_idgrp[spl->spl_right[j]] = curid;
                                len++;
@@ -425,11 +419,7 @@ gistfindgroup(GISTSTATE *giststate, GISTENTRY *valvec, GIST_SPLITVEC *spl, int a
                        {
                                if (spl->spl_idgrp[spl->spl_left[j]])
                                        continue;
-                               FunctionCall3(&giststate->equalFn[attno],
-                                                         valvec[spl->spl_left[i]].key,
-                                                         valvec[spl->spl_left[j]].key,
-                                                         PointerGetDatum(&result));
-                               if (result)
+                               if (gistKeyIsEQ(giststate, attno, valvec[spl->spl_left[i]].key, valvec[spl->spl_left[j]].key))
                                {
                                        spl->spl_idgrp[spl->spl_left[j]] = curid;
                                        len++;
@@ -758,7 +748,14 @@ gistpenalty(GISTSTATE *giststate, int attno,
        return penalty;
 }
 
-void
+/*
+ * Calls user picksplit method for attno columns to split vector to
+ * two vectors. May use attno+n columns data to
+ * get better split.
+ * Returns TRUE if left and right unions of attno columns are the same,
+ * so caller may find better split
+ */
+bool
 gistUserPicksplit(Relation r, GistEntryVector *entryvec, int attno, GIST_SPLITVEC *v,
                                  IndexTuple *itup, int len, GISTSTATE *giststate)
 {
@@ -787,24 +784,36 @@ gistUserPicksplit(Relation r, GistEntryVector *entryvec, int attno, GIST_SPLITVE
         */
        if (giststate->tupdesc->natts > 1 && attno+1 != giststate->tupdesc->natts)
        {
-               int                     MaxGrpId;
+               if ( gistKeyIsEQ(giststate, attno, v->spl_ldatum, v->spl_rdatum) ) {
+                       /*
+                        * Left and right key's unions are equial, so
+                        * we can get better split by following columns. Note,
+                        * uninons for attno columns are already done.
+                        */
+
+                       return true;
+               } else {
+                       int                     MaxGrpId;
 
-               v->spl_idgrp = (int *) palloc0(sizeof(int) * entryvec->n);
-               v->spl_grpflag = (char *) palloc0(sizeof(char) * entryvec->n);
-               v->spl_ngrp = (int *) palloc(sizeof(int) * entryvec->n);
+                       v->spl_idgrp = (int *) palloc0(sizeof(int) * entryvec->n);
+                       v->spl_grpflag = (char *) palloc0(sizeof(char) * entryvec->n);
+                       v->spl_ngrp = (int *) palloc(sizeof(int) * entryvec->n);
 
-               MaxGrpId = gistfindgroup(giststate, entryvec->vector, v, attno);
+                       MaxGrpId = gistfindgroup(giststate, entryvec->vector, v, attno);
 
-               /* form union of sub keys for each page (l,p) */
-               gistunionsubkey(giststate, itup, v, attno + 1);
+                       /* form union of sub keys for each page (l,p) */
+                       gistunionsubkey(giststate, itup, v, attno + 1);
 
-               /*
-                * if possible, we insert equivalent tuples with control by penalty
-                * for a subkey(s)
-                */
-               if (MaxGrpId > 1)
-                       gistadjsubkey(r, itup, len, v, giststate, attno);
+                       /*
+                        * if possible, we insert equivalent tuples with control by penalty
+                       * for a subkey(s)
+                       */
+                       if (MaxGrpId > 1)
+                               gistadjsubkey(r, itup, len, v, giststate, attno);
+               }
        }
+
+       return false;
 }
 
 /*

diff --git a/src/include/access/gist_private.h b/src/include/access/gist_private.h
index 43a6f62..f756828 100644 (file)
--- a/src/include/access/gist_private.h
+++ b/src/include/access/gist_private.h
@@ -7,7 +7,7 @@
  * Portions Copyright (c) 1996-2006, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/access/gist_private.h,v 1.16 2006/05/24 11:01:39 teodor Exp $
+ * $PostgreSQL: pgsql/src/include/access/gist_private.h,v 1.17 2006/05/29 12:50:06 teodor Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -312,7 +312,7 @@ extern void GISTInitBuffer(Buffer b, uint32 f);
 extern void gistdentryinit(GISTSTATE *giststate, int nkey, GISTENTRY *e,
                           Datum k, Relation r, Page pg, OffsetNumber o,
                           int b, bool l, bool isNull);
-void gistUserPicksplit(Relation r, GistEntryVector *entryvec, int attno, GIST_SPLITVEC *v,
+bool gistUserPicksplit(Relation r, GistEntryVector *entryvec, int attno, GIST_SPLITVEC *v,
                                  IndexTuple *itup, int len, GISTSTATE *giststate);
 
 /* gistvacuum.c */