Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings

Commit97930cf

Browse files
committed
Improve eqjoinsel's ndistinct clamping to work for multiple levels of join.
This patch fixes an oversight in my commit7f3eba3 of 2008-10-23. That patchaccounted for baserel restriction clauses that reduced the number of rowscoming out of a table (and hence the number of possibly-distinct values ofa join variable), but not for join restriction clauses that might have beenapplied at a lower level of join. To account for the latter, look up thesizes of the min_lefthand and min_righthand inputs of the current join,and clamp with those in the same way as for the base relations.Noted while investigating a complaint from Ben Chobot, although this initself doesn't seem to explain his report.Back-patch to 8.4; previous versions used different estimation methodsfor which this heuristic isn't relevant.
1 parent5cfe33f commit97930cf

File tree

1 file changed

+73
-8
lines changed

1 file changed

+73
-8
lines changed

‎src/backend/utils/adt/selfuncs.c

Lines changed: 73 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -142,9 +142,11 @@ static double ineq_histogram_selectivity(PlannerInfo *root,
142142
FmgrInfo*opproc,boolisgt,
143143
Datumconstval,Oidconsttype);
144144
staticdoubleeqjoinsel_inner(Oidoperator,
145-
VariableStatData*vardata1,VariableStatData*vardata2);
145+
VariableStatData*vardata1,VariableStatData*vardata2,
146+
RelOptInfo*rel1,RelOptInfo*rel2);
146147
staticdoubleeqjoinsel_semi(Oidoperator,
147-
VariableStatData*vardata1,VariableStatData*vardata2);
148+
VariableStatData*vardata1,VariableStatData*vardata2,
149+
RelOptInfo*rel1,RelOptInfo*rel2);
148150
staticboolconvert_to_scalar(Datumvalue,Oidvaluetypid,double*scaledvalue,
149151
Datumlobound,Datumhibound,Oidboundstypid,
150152
double*scaledlobound,double*scaledhibound);
@@ -173,6 +175,7 @@ static bool get_actual_variable_range(PlannerInfo *root,
173175
VariableStatData*vardata,
174176
Oidsortop,
175177
Datum*min,Datum*max);
178+
staticRelOptInfo*find_join_input_rel(PlannerInfo*root,Relidsrelids);
176179
staticSelectivityprefix_selectivity(PlannerInfo*root,
177180
VariableStatData*vardata,
178181
Oidvartype,Oidopfamily,Const*prefixcon);
@@ -2008,24 +2011,47 @@ eqjoinsel(PG_FUNCTION_ARGS)
20082011
VariableStatDatavardata1;
20092012
VariableStatDatavardata2;
20102013
booljoin_is_reversed;
2014+
RelOptInfo*rel1;
2015+
RelOptInfo*rel2;
20112016

20122017
get_join_variables(root,args,sjinfo,
20132018
&vardata1,&vardata2,&join_is_reversed);
20142019

2020+
/*
2021+
* Identify the join's direct input relations. We use the min lefthand
2022+
* and min righthand as the inputs, even though the join might actually
2023+
* get done with larger input relations. The min inputs are guaranteed to
2024+
* have been formed by now, though, and always using them ensures
2025+
* consistency of estimates.
2026+
*/
2027+
if (!join_is_reversed)
2028+
{
2029+
rel1=find_join_input_rel(root,sjinfo->min_lefthand);
2030+
rel2=find_join_input_rel(root,sjinfo->min_righthand);
2031+
}
2032+
else
2033+
{
2034+
rel1=find_join_input_rel(root,sjinfo->min_righthand);
2035+
rel2=find_join_input_rel(root,sjinfo->min_lefthand);
2036+
}
2037+
20152038
switch (sjinfo->jointype)
20162039
{
20172040
caseJOIN_INNER:
20182041
caseJOIN_LEFT:
20192042
caseJOIN_FULL:
2020-
selec=eqjoinsel_inner(operator,&vardata1,&vardata2);
2043+
selec=eqjoinsel_inner(operator,&vardata1,&vardata2,
2044+
rel1,rel2);
20212045
break;
20222046
caseJOIN_SEMI:
20232047
caseJOIN_ANTI:
20242048
if (!join_is_reversed)
2025-
selec=eqjoinsel_semi(operator,&vardata1,&vardata2);
2049+
selec=eqjoinsel_semi(operator,&vardata1,&vardata2,
2050+
rel1,rel2);
20262051
else
20272052
selec=eqjoinsel_semi(get_commutator(operator),
2028-
&vardata2,&vardata1);
2053+
&vardata2,&vardata1,
2054+
rel2,rel1);
20292055
break;
20302056
default:
20312057
/* other values not expected here */
@@ -2051,7 +2077,8 @@ eqjoinsel(PG_FUNCTION_ARGS)
20512077
*/
20522078
staticdouble
20532079
eqjoinsel_inner(Oidoperator,
2054-
VariableStatData*vardata1,VariableStatData*vardata2)
2080+
VariableStatData*vardata1,VariableStatData*vardata2,
2081+
RelOptInfo*rel1,RelOptInfo*rel2)
20552082
{
20562083
doubleselec;
20572084
doublend1;
@@ -2252,15 +2279,19 @@ eqjoinsel_inner(Oid operator,
22522279
* be, providing a crude correction for the selectivity of restriction
22532280
* clauses on those relations.(We don't do that in the other path
22542281
* since there we are comparing the nd values to stats for the whole
2255-
* relations.)
2282+
* relations.) We can apply this clamp both with respect to the base
2283+
* relations from which the join variables come, and to the immediate
2284+
* input relations of the current join.
22562285
*/
22572286
doublenullfrac1=stats1 ?stats1->stanullfrac :0.0;
22582287
doublenullfrac2=stats2 ?stats2->stanullfrac :0.0;
22592288

22602289
if (vardata1->rel)
22612290
nd1=Min(nd1,vardata1->rel->rows);
2291+
nd1=Min(nd1,rel1->rows);
22622292
if (vardata2->rel)
22632293
nd2=Min(nd2,vardata2->rel->rows);
2294+
nd2=Min(nd2,rel2->rows);
22642295

22652296
selec= (1.0-nullfrac1)* (1.0-nullfrac2);
22662297
if (nd1>nd2)
@@ -2287,7 +2318,8 @@ eqjoinsel_inner(Oid operator,
22872318
*/
22882319
staticdouble
22892320
eqjoinsel_semi(Oidoperator,
2290-
VariableStatData*vardata1,VariableStatData*vardata2)
2321+
VariableStatData*vardata1,VariableStatData*vardata2,
2322+
RelOptInfo*rel1,RelOptInfo*rel2)
22912323
{
22922324
doubleselec;
22932325
doublend1;
@@ -2435,8 +2467,10 @@ eqjoinsel_semi(Oid operator,
24352467
{
24362468
if (vardata1->rel)
24372469
nd1=Min(nd1,vardata1->rel->rows);
2470+
nd1=Min(nd1,rel1->rows);
24382471
if (vardata2->rel)
24392472
nd2=Min(nd2,vardata2->rel->rows);
2473+
nd2=Min(nd2,rel2->rows);
24402474

24412475
if (nd1 <=nd2||nd2 <=0)
24422476
selec=1.0-nullfrac1;
@@ -4759,6 +4793,37 @@ get_actual_variable_range(PlannerInfo *root, VariableStatData *vardata,
47594793
returnhave_data;
47604794
}
47614795

4796+
/*
4797+
* find_join_input_rel
4798+
*Look up the input relation for a join.
4799+
*
4800+
* We assume that the input relation's RelOptInfo must have been constructed
4801+
* already.
4802+
*/
4803+
staticRelOptInfo*
4804+
find_join_input_rel(PlannerInfo*root,Relidsrelids)
4805+
{
4806+
RelOptInfo*rel=NULL;
4807+
4808+
switch (bms_membership(relids))
4809+
{
4810+
caseBMS_EMPTY_SET:
4811+
/* should not happen */
4812+
break;
4813+
caseBMS_SINGLETON:
4814+
rel=find_base_rel(root,bms_singleton_member(relids));
4815+
break;
4816+
caseBMS_MULTIPLE:
4817+
rel=find_join_rel(root,relids);
4818+
break;
4819+
}
4820+
4821+
if (rel==NULL)
4822+
elog(ERROR,"could not find RelOptInfo for given relids");
4823+
4824+
returnrel;
4825+
}
4826+
47624827

47634828
/*-------------------------------------------------------------------------
47644829
*

0 commit comments

Comments
 (0)
[8]ページ先頭
©2009-2025 Movatter.jp