PostgreSQL 소스 코드 해독 (64) - 검색 어 \ # 49 (make one rel 함수 \ # 14 - 연결 경로 \ # 3)

36624 단어
이 절 은 동적 계획 알고리즘 실현 (standard join search) 중의 join 을 대체적으로 소개 합 니 다.search_one_level->make_join_rel 함수, 이 함수 가 두 개의 rels 연결 로 생 성 된 RelOptInfo 를 만 들 고 RelOptInfo 에 접근 경로 정 보 를 추가 합 니 다.
데이터 구조
RelOptInfo
 typedef enum RelOptKind
 {
     RELOPT_BASEREL,//    (   /    )
     RELOPT_JOINREL,//       ,                        
     RELOPT_OTHER_MEMBER_REL,
     RELOPT_OTHER_JOINREL,
     RELOPT_UPPER_REL,//     
     RELOPT_OTHER_UPPER_REL,
     RELOPT_DEADREL
 } RelOptKind;
 
 /*
  * Is the given relation a simple relation i.e a base or "other" member
  * relation?
  */
 #define IS_SIMPLE_REL(rel) \
     ((rel)->reloptkind == RELOPT_BASEREL || \
      (rel)->reloptkind == RELOPT_OTHER_MEMBER_REL)
 
 /* Is the given relation a join relation? */
 #define IS_JOIN_REL(rel)    \
     ((rel)->reloptkind == RELOPT_JOINREL || \
      (rel)->reloptkind == RELOPT_OTHER_JOINREL)
 
 /* Is the given relation an upper relation? */
 #define IS_UPPER_REL(rel)   \
     ((rel)->reloptkind == RELOPT_UPPER_REL || \
      (rel)->reloptkind == RELOPT_OTHER_UPPER_REL)
 
 /* Is the given relation an "other" relation? */
 #define IS_OTHER_REL(rel) \
     ((rel)->reloptkind == RELOPT_OTHER_MEMBER_REL || \
      (rel)->reloptkind == RELOPT_OTHER_JOINREL || \
      (rel)->reloptkind == RELOPT_OTHER_UPPER_REL)
 
 typedef struct RelOptInfo
 {
     NodeTag     type;//    
 
     RelOptKind  reloptkind;//RelOpt  
 
     /* all relations included in this RelOptInfo */
     Relids      relids;         /*Relids(rtindex)   set of base relids (rangetable indexes) */
 
     /* size estimates generated by planner */
     double      rows;           /*          estimated number of result tuples */
 
     /* per-relation planner control flags */
     bool        consider_startup;   /*        ? ,             keep cheap-startup-cost paths? */
     bool        consider_param_startup; /*       ?    ditto, for parameterized paths? */
     bool        consider_parallel;  /*           consider parallel paths? */
 
     /* default result targetlist for Paths scanning this relation */
     struct PathTarget *reltarget;   /*   Relation       list of Vars/Exprs, cost, width */
 
     /* materialization information */
     List       *pathlist;       /*       Path structures */
     List       *ppilist;        /*               ParamPathInfos used in pathlist */
     List       *partial_pathlist;   /* partial Paths */
     struct Path *cheapest_startup_path;//         
     struct Path *cheapest_total_path;//         
     struct Path *cheapest_unique_path;//             
     List       *cheapest_parameterized_paths;//        ?    
 
     /* parameterization information needed for both base rels and join rels */
     /* (see also lateral_vars and lateral_referencers) */
     Relids      direct_lateral_relids;  /*  lateral  ,    Relids rels directly laterally referenced */
     Relids      lateral_relids; /* minimum parameterization of rel */
 
     /* information about a base rel (not set for join rels!) */
     //reloptkind=RELOPT_BASEREL        
     Index       relid;          /* Relation ID */
     Oid         reltablespace;  /*     containing tablespace */
     RTEKind     rtekind;        /*   ?   ?      ?RELATION, SUBQUERY, FUNCTION, etc */
     AttrNumber  min_attr;       /*         smallest attrno of rel (often <0) */
     AttrNumber  max_attr;       /*         largest attrno of rel */
     Relids     *attr_needed;    /*    array indexed [min_attr .. max_attr] */
     int32      *attr_widths;    /*      array indexed [min_attr .. max_attr] */
     List       *lateral_vars;   /*      Vars/PHVs LATERAL Vars and PHVs referenced by rel */
     Relids      lateral_referencers;    /*      Relids rels that reference me laterally */
     List       *indexlist;      /*     IndexOptInfo   list of IndexOptInfo */
     List       *statlist;       /*        list of StatisticExtInfo */
     BlockNumber pages;          /*    size estimates derived from pg_class */
     double      tuples;         /*     */
     double      allvisfrac;     /* ? */
     PlannerInfo *subroot;       /*      ,      root if subquery */
     List       *subplan_params; /*      ,         if subquery */
     int         rel_parallel_workers;   /*     ,     workers? wanted number of parallel workers */
 
     /* Information about foreign tables and foreign joins */
     //FWD    
     Oid         serverid;       /* identifies server for the table or join */
     Oid         userid;         /* identifies user to check access as */
     bool        useridiscurrent;    /* join is only valid for current user */
     /* use "struct FdwRoutine" to avoid including fdwapi.h here */
     struct FdwRoutine *fdwroutine;
     void       *fdw_private;
 
     /* cache space for remembering if we have proven this relation unique */
     //   ,      Relids  
     List       *unique_for_rels;    /* known unique for these other relid
                                      * set(s) */
     List       *non_unique_for_rels;    /*    ,    Relids   known not unique for these set(s) */
 
     /* used by various scans and joins: */
     List       *baserestrictinfo;   /*       ,       RestrictInfo structures (if base rel) */
     QualCost    baserestrictcost;   /*           ? cost of evaluating the above */
     Index       baserestrict_min_security;  /*        min security_level found in
                                              * baserestrictinfo */
     List       *joininfo;       /*             RestrictInfo structures for join clauses
                                  * involving this rel */
     bool        has_eclass_joins;   /*          ? T means joininfo is incomplete */
 
     /* used by partitionwise joins: */
     bool        consider_partitionwise_join;    /*   ? consider partitionwise
                                                  * join paths? (if
                                                  * partitioned rel) */
     Relids      top_parent_relids;  /* Relids of topmost parents (if "other"
                                      * rel) */
 
     /* used for partitioned relations */
     //     
     PartitionScheme part_scheme;    /*    schema Partitioning scheme. */
     int         nparts;         /*     number of partitions */
     struct PartitionBoundInfoData *boundinfo;   /*        Partition bounds */
     List       *partition_qual; /*      partition constraint */
     struct RelOptInfo **part_rels;  /*    RelOptInfo   Array of RelOptInfos of partitions,
                                      * stored in the same order of bounds */
     List      **partexprs;      /*          Non-nullable partition key expressions. */
     List      **nullable_partexprs; /*            Nullable partition key expressions. */
     List       *partitioned_child_rels; /* RT Indexes   List of RT indexes. */
 } RelOptInfo;


2. 소스 코드 해독
join_search_one_level - >... (예: make rels by clause joins) - > makejoin_rel 함수 가 두 개의 rels 연결 로 생 성 된 RelOptInfo 를 만 들 고 접근 경 로 를 만들어 RelOptInfo 의 pathlist 링크 에 추가 합 니 다.메 이 크join_rel 함수 중의 buildjoin_rel 함수, populatejoinrel_with_paths 함수 다음 소절 을 소개 합 니 다.

//---------------------------------------------------- make_join_rel
 /*
  * make_join_rel
  *     Find or create a join RelOptInfo that represents the join of
  *     the two given rels, and add to it path information for paths
  *     created with the two rels as outer and inner rel.
  *     (The join rel may already contain paths generated from other
  *     pairs of rels that add up to the same set of base rels.)
  *         rels      RelOptInfo,         .
  *     (    rel            rels        )
  *
  * NB: will return NULL if attempted join is not valid.  This can happen
  * when working with outer joins, or with IN or EXISTS clauses that have been
  * turned into joins.
  *   :        ,   NULL.                    IN/EXISTS   
  */
 RelOptInfo *
 make_join_rel(PlannerInfo *root, RelOptInfo *rel1, RelOptInfo *rel2)
 {
     Relids      joinrelids;
     SpecialJoinInfo *sjinfo;
     bool        reversed;
     SpecialJoinInfo sjinfo_data;
     RelOptInfo *joinrel;
     List       *restrictlist;
 
     /* We should never try to join two overlapping sets of rels. */
     Assert(!bms_overlap(rel1->relids, rel2->relids));//     
 
     /* Construct Relids set that identifies the joinrel. */
     joinrelids = bms_union(rel1->relids, rel2->relids);//  rel   rels
 
     /* Check validity and determine join type. */
     if (!join_is_legal(root, rel1, rel2, joinrelids,
                        &sjinfo, &reversed))//    
     {
         /* invalid join path */
         bms_free(joinrelids);
         return NULL;//  
     }
 
     /* Swap rels if needed to match the join info. */
     if (reversed)//      
     {
         RelOptInfo *trel = rel1;
 
         rel1 = rel2;
         rel2 = trel;
     }
 
     /*
      * If it's a plain inner join, then we won't have found anything in
      * join_info_list.  Make up a SpecialJoinInfo so that selectivity
      * estimation functions will know what's being joined.
      *       ,     join_info_list,
      *     SpecialJoinInfo              
      */
     if (sjinfo == NULL)
     {
         sjinfo = &sjinfo_data;
         sjinfo->type = T_SpecialJoinInfo;
         sjinfo->min_lefthand = rel1->relids;
         sjinfo->min_righthand = rel2->relids;
         sjinfo->syn_lefthand = rel1->relids;
         sjinfo->syn_righthand = rel2->relids;
         sjinfo->jointype = JOIN_INNER;
         /* we don't bother trying to make the remaining fields valid */
         sjinfo->lhs_strict = false;
         sjinfo->delay_upper_joins = false;
         sjinfo->semi_can_btree = false;
         sjinfo->semi_can_hash = false;
         sjinfo->semi_operators = NIL;
         sjinfo->semi_rhs_exprs = NIL;
     }
 
     /*
      * Find or build the join RelOptInfo, and compute the restrictlist that
      * goes with this particular joining.
      *           RelOptInfo,             
      */
     joinrel = build_join_rel(root, joinrelids, rel1, rel2, sjinfo,
                              &restrictlist);
 
     /*
      * If we've already proven this join is empty, we needn't consider any
      * more paths for it.
      */
     if (is_dummy_rel(joinrel))
     {
         bms_free(joinrelids);
         return joinrel;
     }
 
     /* Add paths to the join relation. */
     //               
     populate_joinrel_with_paths(root, rel1, rel2, joinrel, sjinfo,
                                 restrictlist);
 
     bms_free(joinrelids);//    
 
     return joinrel;//  joinrel
 }
 
//-------------------------------------------------------------------- build_join_rel
 /*
  * build_join_rel
  *    Returns relation entry corresponding to the union of two given rels,
  *    creating a new relation entry if none already exists.
  *        rels,          rels       Relation
  * 
  * 'joinrelids' is the Relids set that uniquely identifies the join
  * 'outer_rel' and 'inner_rel' are relation nodes for the relations to be
  *      joined
  * 'sjinfo': join context info
  * 'restrictlist_ptr': result variable.  If not NULL, *restrictlist_ptr
  *      receives the list of RestrictInfo nodes that apply to this
  *      particular pair of joinable relations.
  * joinrelids-         relids
  * outer_rel inner_rel-       (   )    
  * sjinfo-       
  * restrictlist_ptr-       ,   NULL ,     RestrictInfo(    )    
  *
  * restrictlist_ptr makes the routine's API a little grotty, but it saves
  * duplicated calculation of the restrictlist...
  */
 RelOptInfo *
 build_join_rel(PlannerInfo *root,
                Relids joinrelids,
                RelOptInfo *outer_rel,
                RelOptInfo *inner_rel,
                SpecialJoinInfo *sjinfo,
                List **restrictlist_ptr)
 {
     RelOptInfo *joinrel;
     List       *restrictlist;
 
     /* This function should be used only for join between parents. */
     Assert(!IS_OTHER_REL(outer_rel) && !IS_OTHER_REL(inner_rel));
 
     /*
      * See if we already have a joinrel for this set of base rels.
      *     rels           ?
      */
     joinrel = find_join_rel(root, joinrelids);
 
     if (joinrel)//   
     {
         /*
          * Yes, so we only need to figure the restrictlist for this particular
          * pair of component relations.
          */
         if (restrictlist_ptr)
             *restrictlist_ptr = build_joinrel_restrictlist(root,
                                                            joinrel,
                                                            outer_rel,
                                                            inner_rel);//        ,          
         return joinrel;//  
     }
 
     /*
      * Nope, so make one.
      *   ,    
      */
     joinrel = makeNode(RelOptInfo);
     joinrel->reloptkind = RELOPT_JOINREL;
     joinrel->relids = bms_copy(joinrelids);
     joinrel->rows = 0;
     /* cheap startup cost is interesting iff not all tuples to be retrieved */
     joinrel->consider_startup = (root->tuple_fraction > 0);
     joinrel->consider_param_startup = false;
     joinrel->consider_parallel = false;
     joinrel->reltarget = create_empty_pathtarget();
     joinrel->pathlist = NIL;
     joinrel->ppilist = NIL;
     joinrel->partial_pathlist = NIL;
     joinrel->cheapest_startup_path = NULL;
     joinrel->cheapest_total_path = NULL;
     joinrel->cheapest_unique_path = NULL;
     joinrel->cheapest_parameterized_paths = NIL;
     /* init direct_lateral_relids from children; we'll finish it up below */
     joinrel->direct_lateral_relids =
         bms_union(outer_rel->direct_lateral_relids,
                   inner_rel->direct_lateral_relids);
     joinrel->lateral_relids = min_join_parameterization(root, joinrel->relids,
                                                         outer_rel, inner_rel);
     joinrel->relid = 0;         /* indicates not a baserel */
     joinrel->rtekind = RTE_JOIN;//RTE_JOIN
     joinrel->min_attr = 0;
     joinrel->max_attr = 0;
     joinrel->attr_needed = NULL;
     joinrel->attr_widths = NULL;
     joinrel->lateral_vars = NIL;
     joinrel->lateral_referencers = NULL;
     joinrel->indexlist = NIL;
     joinrel->statlist = NIL;
     joinrel->pages = 0;
     joinrel->tuples = 0;
     joinrel->allvisfrac = 0;
     joinrel->subroot = NULL;
     joinrel->subplan_params = NIL;
     joinrel->rel_parallel_workers = -1;
     joinrel->serverid = InvalidOid;
     joinrel->userid = InvalidOid;
     joinrel->useridiscurrent = false;
     joinrel->fdwroutine = NULL;
     joinrel->fdw_private = NULL;
     joinrel->unique_for_rels = NIL;
     joinrel->non_unique_for_rels = NIL;
     joinrel->baserestrictinfo = NIL;
     joinrel->baserestrictcost.startup = 0;
     joinrel->baserestrictcost.per_tuple = 0;
     joinrel->baserestrict_min_security = UINT_MAX;
     joinrel->joininfo = NIL;
     joinrel->has_eclass_joins = false;
     joinrel->consider_partitionwise_join = false; /* might get changed later */
     joinrel->top_parent_relids = NULL;
     joinrel->part_scheme = NULL;
     joinrel->nparts = 0;
     joinrel->boundinfo = NULL;
     joinrel->partition_qual = NIL;
     joinrel->part_rels = NULL;
     joinrel->partexprs = NULL;
     joinrel->nullable_partexprs = NULL;
     joinrel->partitioned_child_rels = NIL;
 
     /*   FDW     ,Compute information relevant to the foreign relations. */
     set_foreign_rel_properties(joinrel, outer_rel, inner_rel);
 
     /*
      * Create a new tlist containing just the vars that need to be output from
      * this join (ie, are needed for higher joinclauses or final output).
      *
      *            ,         
      * NOTE: the tlist order for a join rel will depend on which pair of outer
      * and inner rels we first try to build it from.  But the contents should
      * be the same regardless.
      */
     build_joinrel_tlist(root, joinrel, outer_rel);//    
     build_joinrel_tlist(root, joinrel, inner_rel);//    
     add_placeholders_to_joinrel(root, joinrel, outer_rel, inner_rel);//  PHV
 
     /*
      * add_placeholders_to_joinrel also took care of adding the ph_lateral
      * sets of any PlaceHolderVars computed here to direct_lateral_relids, so
      * now we can finish computing that.  This is much like the computation of
      * the transitively-closed lateral_relids in min_join_parameterization,
      * except that here we *do* have to consider the added PHVs.
      * add_placeholders_to_joinrel        PlaceHolderVars ph_lateral  
      *    direct_lateral_relids         。
      *       min_join_parameterization transtivelyclosed lateral al_relid   ,
      *            phv。
      */
     joinrel->direct_lateral_relids =
         bms_del_members(joinrel->direct_lateral_relids, joinrel->relids);
     if (bms_is_empty(joinrel->direct_lateral_relids))
         joinrel->direct_lateral_relids = NULL;
 
     /*
      * Construct restrict and join clause lists for the new joinrel. (The
      * caller might or might not need the restrictlist, but I need it anyway
      * for set_joinrel_size_estimates().)
      *      joinrel           
      */
     restrictlist = build_joinrel_restrictlist(root, joinrel,
                                               outer_rel, inner_rel);//        
     if (restrictlist_ptr)
         *restrictlist_ptr = restrictlist;
     build_joinrel_joinlist(joinrel, outer_rel, inner_rel);//        
 
     /*
      * This is also the right place to check whether the joinrel has any
      * pending EquivalenceClass joins.
      *          EC
      */
     joinrel->has_eclass_joins = has_relevant_eclass_joinclause(root, joinrel);
 
     /* S      ,tore the partition information. */
     build_joinrel_partition_info(joinrel, outer_rel, inner_rel, restrictlist,
                                  sjinfo->jointype);
 
     /*
      *   joinrel   ,Set estimates of the joinrel's size.
      */
     set_joinrel_size_estimates(root, joinrel, outer_rel, inner_rel,
                                sjinfo, restrictlist);
 
     /*
      * Set the consider_parallel flag if this joinrel could potentially be
      * scanned within a parallel worker.  If this flag is false for either
      * inner_rel or outer_rel, then it must be false for the joinrel also.
      * Even if both are true, there might be parallel-restricted expressions
      * in the targetlist or quals.
      *   consider_parallel  , joinrel        
      *
      * Note that if there are more than two rels in this relation, they could
      * be divided between inner_rel and outer_rel in any arbitrary way.  We
      * assume this doesn't matter, because we should hit all the same baserels
      * and joinclauses while building up to this joinrel no matter which we
      * take; therefore, we should make the same decision here however we get
      * here.
      */
     if (inner_rel->consider_parallel && outer_rel->consider_parallel &&
         is_parallel_safe(root, (Node *) restrictlist) &&
         is_parallel_safe(root, (Node *) joinrel->reltarget->exprs))
         joinrel->consider_parallel = true;
 
     /* Add the joinrel to the PlannerInfo. */
     add_join_rel(root, joinrel);//         
 
     /*
      * Also, if dynamic-programming join search is active, add the new joinrel
      * to the appropriate sublist.  Note: you might think the Assert on number
      * of members should be for equality, but some of the level 1 rels might
      * have been joinrels already, so we can only assert <=.
      *          root->join_rel_levep[j]
      */
     if (root->join_rel_level)
     {
         Assert(root->join_cur_level > 0);
         Assert(root->join_cur_level <= bms_num_members(joinrel->relids));
         root->join_rel_level[root->join_cur_level] =
             lappend(root->join_rel_level[root->join_cur_level], joinrel);//      
     }
 
     return joinrel;
 }


//----------------------------------------------- find_join_rel

 /*
  * find_join_rel
  *    Returns relation entry corresponding to 'relids' (a set of RT indexes),
  *    or NULL if none exists.  This is for join relations.
  *        relids(RT indexes   ) RelOptInfo,     NULL.
  */
 RelOptInfo *
 find_join_rel(PlannerInfo *root, Relids relids)
 {
     /*
      * Switch to using hash lookup when list grows "too long".  The threshold
      * is arbitrary and is known only here.
      *      ,   hash  
      */
     if (!root->join_rel_hash && list_length(root->join_rel_list) > 32)
         build_join_rel_hash(root);
 
     /*
      * Use either hashtable lookup or linear search, as appropriate.
      *   hash         
      *
      * Note: the seemingly redundant hashkey variable is used to avoid taking
      * the address of relids; unless the compiler is exceedingly smart, doing
      * so would force relids out of a register and thus probably slow down the
      * list-search case.
      */
     if (root->join_rel_hash)//hash
     {
         Relids      hashkey = relids;
         JoinHashEntry *hentry;
 
         hentry = (JoinHashEntry *) hash_search(root->join_rel_hash,
                                                &hashkey,
                                                HASH_FIND,
                                                NULL);
         if (hentry)
             return hentry->join_rel;
     }
     else//  
     {
         ListCell   *l;
 
         foreach(l, root->join_rel_list)
         {
             RelOptInfo *rel = (RelOptInfo *) lfirst(l);
 
             if (bms_equal(rel->relids, relids))
                 return rel;
         }
     }
 
     return NULL;
 }
 

//----------------------------------------------- build_joinrel_restrictlist

 /*
  * build_joinrel_restrictlist
  * build_joinrel_joinlist
  *    These routines build lists of restriction and join clauses for a
  *    join relation from the joininfo lists of the relations it joins.
  *        joininfo                
  *    
  *    These routines are separate because the restriction list must be
  *    built afresh for each pair of input sub-relations we consider, whereas
  *    the join list need only be computed once for any join RelOptInfo.
  *    The join list is fully determined by the set of rels making up the
  *    joinrel, so we should get the same results (up to ordering) from any
  *    candidate pair of sub-relations.  But the restriction list is whatever
  *    is not handled in the sub-relations, so it depends on which
  *    sub-relations are considered.
  *              ,                           ,
  *                   RelOptInfo      。
  *             joinrel   rels  ,
  *                           (      )。
  *                        ,            。   
  *
  *    If a join clause from an input relation refers to base rels still not
  *    present in the joinrel, then it is still a join clause for the joinrel;
  *    we put it into the joininfo list for the joinrel.  Otherwise,
  *    the clause is now a restrict clause for the joined relation, and we
  *    return it to the caller of build_joinrel_restrictlist() to be stored in
  *    join paths made from this pair of sub-relations.  (It will not need to
  *    be considered further up the join tree.)
  *                       base rels  joinrel ,
  *          joinrel       ;       joinrel joininfo   。
  *      ,                  ,
  *           build_joinrel_restrictlist()    ,                   。
  *    (                。)
  *
  *    In many case we will find the same RestrictInfos in both input
  *    relations' joinlists, so be careful to eliminate duplicates.
  *    Pointer equality should be a sufficient test for dups, since all
  *    the various joinlist entries ultimately refer to RestrictInfos
  *    pushed into them by distribute_restrictinfo_to_rels().
  *          ,                  RestrictInfos,         。
  *                       ,     joinlist    
  *       distribute_restrictinfo_to_rels()   RestrictInfos。
  *
  * 'joinrel' is a join relation node,        
  * 'outer_rel' and 'inner_rel' are a pair of relations that can be joined
  *      to form joinrel.        
  *
  * build_joinrel_restrictlist() returns a list of relevant restrictinfos,
  * whereas build_joinrel_joinlist() stores its results in the joinrel's
  * joininfo list.  One or the other must accept each given clause!
  * build_joinrel_restrictlist()           ,
  *  build_joinrel_joinlist()      joinrel joininfo   
  *
  * NB: Formerly, we made deep(!) copies of each input RestrictInfo to pass
  * up to the join relation.  I believe this is no longer necessary, because
  * RestrictInfo nodes are no longer context-dependent.  Instead, just include
  * the original nodes in the lists made for the join relation.
  */
 static List *
 build_joinrel_restrictlist(PlannerInfo *root,
                            RelOptInfo *joinrel,
                            RelOptInfo *outer_rel,
                            RelOptInfo *inner_rel)
 {
     List       *result;
 
     /*
      * Collect all the clauses that syntactically belong at this level,
      * eliminating any duplicates (important since we will see many of the
      * same clauses arriving from both input relations).
      *                    ,      (    ,               )。
      */
     result = subbuild_joinrel_restrictlist(joinrel, outer_rel->joininfo, NIL);
     result = subbuild_joinrel_restrictlist(joinrel, inner_rel->joininfo, result);
 
     /*
      * Add on any clauses derived from EquivalenceClasses.  These cannot be
      * redundant with the clauses in the joininfo lists, so don't bother
      * checking.
      *     EC     .
      */
     result = list_concat(result,
                          generate_join_implied_equalities(root,
                                                           joinrel->relids,
                                                           outer_rel->relids,
                                                           inner_rel));
 
     return result;
 }
 
 
 static void
 build_joinrel_joinlist(RelOptInfo *joinrel,
                        RelOptInfo *outer_rel,
                        RelOptInfo *inner_rel)
 {
     List       *result;
 
     /*
      * Collect all the clauses that syntactically belong above this level,
      * eliminating any duplicates (important since we will see many of the
      * same clauses arriving from both input relations).
      *                    ,      (    ,               )。
      */
     result = subbuild_joinrel_joinlist(joinrel, outer_rel->joininfo, NIL);
     result = subbuild_joinrel_joinlist(joinrel, inner_rel->joininfo, result);
 
     joinrel->joininfo = result;
 }
 

3. 추적 분석
테스트 표 와 데 이 터 는 이전 절 에 만 든 표 와 데 이 터 를 계속 사용 하고 사용 하 는 SQL 문 구 는 다음 과 같 습 니 다.
testdb=# explain verbose select a.*,b.c1,c.c2,d.c2,e.c1,f.c2
from a inner join b on a.c1=b.c1,c,d,e inner join f on e.c1 = f.c1 and e.c1 < 100
where a.c1=f.c1 and b.c1=c.c1 and c.c1 = d.c1 and d.c1 = e.c1;
                                                QUERY PLAN                                                
----------------------------------------------------------------------------------------------------------
 Nested Loop  (cost=101.17..2218.24 rows=2 width=42)
   Output: a.c1, a.c2, b.c1, c.c2, d.c2, e.c1, f.c2
   Join Filter: (a.c1 = b.c1)
   ->  Hash Join  (cost=3.25..196.75 rows=100 width=22)
         Output: a.c1, a.c2, c.c2, c.c1
         Hash Cond: (c.c1 = a.c1)
         ->  Seq Scan on public.c  (cost=0.00..155.00 rows=10000 width=12)
               Output: c.c1, c.c2
         ->  Hash  (cost=2.00..2.00 rows=100 width=10)
               Output: a.c1, a.c2
               ->  Seq Scan on public.a  (cost=0.00..2.00 rows=100 width=10)
                     Output: a.c1, a.c2
   ->  Materialize  (cost=97.92..2014.00 rows=5 width=32)
         Output: b.c1, d.c2, d.c1, e.c1, f.c2, f.c1
         ->  Hash Join  (cost=97.92..2013.97 rows=5 width=32)
               Output: b.c1, d.c2, d.c1, e.c1, f.c2, f.c1
               Hash Cond: (f.c1 = b.c1)
               ->  Seq Scan on public.f  (cost=0.00..1541.00 rows=100000 width=13)
                     Output: f.c1, f.c2
               ->  Hash  (cost=97.86..97.86 rows=5 width=19)
                     Output: b.c1, d.c2, d.c1, e.c1
                     ->  Hash Join  (cost=78.10..97.86 rows=5 width=19)
                           Output: b.c1, d.c2, d.c1, e.c1
                           Hash Cond: (b.c1 = e.c1)
                           ->  Seq Scan on public.b  (cost=0.00..16.00 rows=1000 width=4)
                                 Output: b.c1, b.c2
                           ->  Hash  (cost=78.04..78.04 rows=5 width=15)
                                 Output: d.c2, d.c1, e.c1
                                 ->  Hash Join  (cost=73.24..78.04 rows=5 width=15)
                                       Output: d.c2, d.c1, e.c1
                                       Hash Cond: (d.c1 = e.c1)
                                       ->  Seq Scan on public.d  (cost=0.00..4.00 rows=200 width=11)
                                             Output: d.c1, d.c2
                                       ->  Hash  (cost=72.00..72.00 rows=99 width=4)
                                             Output: e.c1
                                             ->  Seq Scan on public.e  (cost=0.00..72.00 rows=99 width=4)
                                                   Output: e.c1
                                                   Filter: (e.c1 < 100)
(38 rows)

최적화 기 는 2 rels + 4 rels 의 연결 모델 을 선택 하여 bushy plans 의 실행 상황 을 중점적으로 고찰 합 니 다.
gdb 를 시작 하고 정지점 을 설정 하 며 level = 6 의 상황 만 고찰 합 니 다.
(gdb) b join_search_one_level
Breakpoint 2 at 0x7b0289: file joinrels.c, line 67.
(gdb) c
Continuing.
...
(gdb) c
Continuing.

Breakpoint 2, join_search_one_level (root=0x241ca38, level=6) at joinrels.c:67
67      List      **joinrels = root->join_rel_level;

5 (rels) + 1 (rels) 호출 완료
(gdb) b joinrels.c:142
Breakpoint 3 at 0x7b03c4: file joinrels.c, line 142.
(gdb) c
Continuing.

Breakpoint 3, join_search_one_level (root=0x241ca38, level=6) at joinrels.c:142
142     for (k = 2;; k++)

root - > join 보기rel_level[6]
(gdb) p *root->join_rel_level[6]
$1 = {type = T_List, length = 1, head = 0x24c8468, tail = 0x24c8468}

이 링크 의 RelOptInfo 보기
(gdb) set $roi=(RelOptInfo *)root->join_rel_level[6]->head->data.ptr_value
(gdb) p *$roi
$3 = {type = T_RelOptInfo, reloptkind = RELOPT_JOINREL, relids = 0x1eb8330, rows = 2, consider_startup = false, 
  consider_param_startup = false, consider_parallel = true, reltarget = 0x1f25ac8, pathlist = 0x1f25f80, ppilist = 0x0, 
  partial_pathlist = 0x0, cheapest_startup_path = 0x0, cheapest_total_path = 0x0, cheapest_unique_path = 0x0, 
  cheapest_parameterized_paths = 0x0, direct_lateral_relids = 0x0, lateral_relids = 0x0, relid = 0, reltablespace = 0, 
  rtekind = RTE_JOIN, min_attr = 0, max_attr = 0, attr_needed = 0x0, attr_widths = 0x0, lateral_vars = 0x0, 
  lateral_referencers = 0x0, indexlist = 0x0, statlist = 0x0, pages = 0, tuples = 0, allvisfrac = 0, subroot = 0x0, 
  subplan_params = 0x0, rel_parallel_workers = -1, serverid = 0, userid = 0, useridiscurrent = false, fdwroutine = 0x0, 
  fdw_private = 0x0, unique_for_rels = 0x0, non_unique_for_rels = 0x0, baserestrictinfo = 0x0, baserestrictcost = {
    startup = 0, per_tuple = 0}, baserestrict_min_security = 4294967295, joininfo = 0x0, has_eclass_joins = false, 
  top_parent_relids = 0x0, part_scheme = 0x0, nparts = 0, boundinfo = 0x0, partition_qual = 0x0, part_rels = 0x0, 
  partexprs = 0x0, nullable_partexprs = 0x0, partitioned_child_rels = 0x0}

이 RelOptInfo 의 pathlist 보기
(gdb) p *$roi->pathlist
$4 = {type = T_List, length = 1, head = 0x1f25f60, tail = 0x1f25f60}
(gdb) p *(Node *)$roi->pathlist->head->data.ptr_value
$5 = {type = T_NestPath}
(gdb) set $np=(NestPath *)$roi->pathlist->head->data.ptr_value
(gdb) p *(NestPath *)$np
$5 = {path = {type = T_NestPath, pathtype = T_NestLoop, parent = 0x1f258b8, pathtarget = 0x1f25ac8, param_info = 0x0, 
    parallel_aware = false, parallel_safe = true, parallel_workers = 0, rows = 2, startup_cost = 290.57499999999999, 
    total_cost = 2216.1374999999998, pathkeys = 0x0}, jointype = JOIN_INNER, inner_unique = false, 
  outerjoinpath = 0x1f07c00, innerjoinpath = 0x1f27c40, joinrestrictinfo = 0x1f27e60}

이 연결 의 겉모습 과 내부 접근 경 로 를 보십시오
(gdb) p *$np->outerjoinpath
$6 = {type = T_Path, pathtype = T_SeqScan, parent = 0x1e228e8, pathtarget = 0x1f04bc0, param_info = 0x0, 
  parallel_aware = false, parallel_safe = true, parallel_workers = 0, rows = 100, startup_cost = 0, total_cost = 2, 
  pathkeys = 0x0}
(gdb) p *$np->innerjoinpath
$7 = {type = T_MaterialPath, pathtype = T_Material, parent = 0x1ebb538, pathtarget = 0x1ebb748, param_info = 0x0, 
  parallel_aware = false, parallel_safe = true, parallel_workers = 0, rows = 5, startup_cost = 290.57499999999999, 
  total_cost = 2206.6500000000001, pathkeys = 0x0}

다음은 bushy plans, 즉 (2 / 4 rels + 4 / 2 rels) 또는 (3 rels + 3 rels) 모델 을 시도 하여 ac + bdef 라 는 조합 을 중점적으로 고찰 합 니 다.
(gdb) b joinrels.c:156
Breakpoint 3 at 0x7557df: file joinrels.c, line 156.
(gdb) c
Continuing.

Breakpoint 3, join_search_one_level (root=0x1e214b8, level=6) at joinrels.c:164
164             if (old_rel->joininfo == NIL && !old_rel->has_eclass_joins &&
(gdb) p *old_rel->relids->words
$13 = 18

Make join rel 함수 들 어가 기
173             for_each_cell(r2, other_rels)
(gdb) 
175                 RelOptInfo *new_rel = (RelOptInfo *) lfirst(r2);
(gdb) 
177                 if (!bms_overlap(old_rel->relids, new_rel->relids))
(gdb) 
184                     if (have_relevant_joinclause(root, old_rel, new_rel) ||
(gdb) 
187                         (void) make_join_rel(root, old_rel, new_rel);
(gdb) step
make_join_rel (root=0x1e214b8, rel1=0x1f079f0, rel2=0x1e96520) at joinrels.c:681
681     joinrelids = bms_union(rel1->relids, rel2->relids);

build join rel 함수 에 들 어가 면 해당 RelOptInfo 가 이미 존재 합 니 다. 되 돌려 줍 니 다.
(gdb) 
728     joinrel = build_join_rel(root, joinrelids, rel1, rel2, sjinfo,
(gdb) step
build_join_rel (root=0x1e214b8, joinrelids=0x1e401d8, outer_rel=0x1f079f0, inner_rel=0x1e96520, sjinfo=0x7fff247e18a0, 
    restrictlist_ptr=0x7fff247e1898) at relnode.c:498
498     joinrel = find_join_rel(root, joinrelids);
500     if (joinrel)
(gdb) n
506         if (restrictlist_ptr)
(gdb) 
507             *restrictlist_ptr = build_joinrel_restrictlist(root,
(gdb) 
511         return joinrel;

populate joinrel with paths 를 실행 합 니 다. 이 함수 가 실 행 된 후에 외모 와 내부 방문 경 로 를 다시 보고 HashPath + MaterialPath 의 조합 이 되 었 습 니 다. 구체 적 인 변 화 는 다음 절 에 소개 합 니 다.
...
(gdb) 
742     populate_joinrel_with_paths(root, rel1, rel2, joinrel, sjinfo,
(gdb) n
745     bms_free(joinrelids);
(gdb) set $roi=(RelOptInfo *)root->join_rel_level[6]->head->data.ptr_value
(gdb) set $np=(NestPath *)$roi->pathlist->head->data.ptr_value
(gdb) p *$np->outerjoinpath
$30 = {type = T_HashPath, pathtype = T_HashJoin, parent = 0x1f079f0, pathtarget = 0x1e41128, param_info = 0x0, 
  parallel_aware = false, parallel_safe = true, parallel_workers = 0, rows = 100, startup_cost = 3.25, total_cost = 196.75, 
  pathkeys = 0x0}
(gdb) p *$np->innerjoinpath
$31 = {type = T_MaterialPath, pathtype = T_Material, parent = 0x1e96520, pathtarget = 0x1e96730, param_info = 0x0, 
  parallel_aware = false, parallel_safe = true, parallel_workers = 0, rows = 5, startup_cost = 97.962499999999991, 
  total_cost = 2014.0375000000001, pathkeys = 0x0}

DONE!
참고 자료
allpaths.c cost.h costsize.c PG Document:Query Planning

좋은 웹페이지 즐겨찾기