public inbox for [email protected]help / color / mirror / Atom feed
[PATCH 5/5] WIP unify handling of attributes and expressions 28+ messages / 3 participants [nested] [flat]
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* [PATCH 5/5] WIP unify handling of attributes and expressions @ 2021-03-04 03:53 Tomas Vondra <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Tomas Vondra @ 2021-03-04 03:53 UTC (permalink / raw) --- src/backend/statistics/dependencies.c | 82 ++------ src/backend/statistics/extended_stats.c | 180 +++++++++++------- src/backend/statistics/mcv.c | 89 ++------- src/backend/statistics/mvdistinct.c | 138 +++----------- .../statistics/extended_stats_internal.h | 40 ++-- 5 files changed, 183 insertions(+), 346 deletions(-) diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index 602301b724..14d6503e1a 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -70,10 +70,7 @@ static void generate_dependencies(DependencyGenerator state); static DependencyGenerator DependencyGenerator_init(int n, int k); static void DependencyGenerator_free(DependencyGenerator state); static AttrNumber *DependencyGenerator_next(DependencyGenerator state); -static double dependency_degree(int numrows, HeapTuple *rows, - ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs); +static double dependency_degree(StatBuildData *data, int k, AttrNumber *dependency); static bool dependency_is_fully_matched(MVDependency *dependency, Bitmapset *attnums); static bool dependency_is_compatible_clause(Node *clause, Index relid, @@ -222,17 +219,13 @@ DependencyGenerator_next(DependencyGenerator state) * the last one. */ static double -dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, - AttrNumber *dependency, VacAttrStats **stats, - Bitmapset *attrs) +dependency_degree(StatBuildData *data, int k, AttrNumber *dependency) { int i, nitems; MultiSortSupport mss; SortItem *items; - AttrNumber *attnums; AttrNumber *attnums_dep; - int numattrs; /* counters valid within a group */ int group_size = 0; @@ -247,16 +240,9 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* sort info for all attributes columns */ mss = multi_sort_init(k); - /* - * Transform the attrs from bitmap to an array to make accessing the i-th - * member easier, and then construct a filtered version with only attnums - * referenced by the dependency we validate. - */ - attnums = build_attnums_array(attrs, exprs->nexprs, &numattrs); - attnums_dep = (AttrNumber *) palloc(k * sizeof(AttrNumber)); for (i = 0; i < k; i++) - attnums_dep[i] = attnums[dependency[i]]; + attnums_dep[i] = data->attnums[dependency[i]]; /* * Verify the dependency (a,b,...)->z, using a rather simple algorithm: @@ -274,7 +260,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, /* prepare the sort function for the dimensions */ for (i = 0; i < k; i++) { - VacAttrStats *colstat = stats[dependency[i]]; + VacAttrStats *colstat = data->stats[dependency[i]]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -293,8 +279,7 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * descriptor. For now that assumption holds, but it might change in the * future for example if we support statistics on multiple tables. */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, k, attnums_dep); + items = build_sorted_items(data, &nitems, mss, k, attnums_dep); /* * Walk through the sorted array, split it into rows according to the @@ -340,11 +325,10 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, pfree(items); pfree(mss); - pfree(attnums); pfree(attnums_dep); /* Compute the 'degree of validity' as (supporting/total). */ - return (n_supporting_rows * 1.0 / numrows); + return (n_supporting_rows * 1.0 / data->numrows); } /* @@ -364,54 +348,15 @@ dependency_degree(int numrows, HeapTuple *rows, ExprInfo *exprs, int k, * (c) -> b */ MVDependencies * -statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_dependencies_build(StatBuildData *data) { int i, k; - int numattrs; - AttrNumber *attnums; - int nattnums; /* result */ MVDependencies *dependencies = NULL; - /* - * Transform the bms into an array of attnums, to make accessing i-th - * member easier. We add the expressions first, represented by negative - * attnums (this is OK, we don't allow statistics on system attributes), - * and then regular attributes. - */ - nattnums = bms_num_members(attrs) + exprs->nexprs; - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * nattnums); - - numattrs = 0; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - /* - * and then regular attributes - * - * XXX Maybe add this in the opposite order, just like in MCV? first - * regular attnums, then exressions. - */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - Assert(numattrs >= 2); - Assert(numattrs == nattnums); - - /* - * Build a new bitmapset of attnums, offset by number of expressions (this - * is needed, because bitmaps can store only non-negative values). - */ - attrs = NULL; - for (i = 0; i < numattrs; i++) - attrs = bms_add_member(attrs, attnums[i] + exprs->nexprs); + Assert(data->nattnums >= 2); /* * We'll try build functional dependencies starting from the smallest ones @@ -419,12 +364,12 @@ statext_dependencies_build(int numrows, HeapTuple *rows, * included in the statistics object. We start from the smallest ones * because we want to be able to skip already implied ones. */ - for (k = 2; k <= numattrs; k++) + for (k = 2; k <= data->nattnums; k++) { AttrNumber *dependency; /* array with k elements */ /* prepare a DependencyGenerator of variation */ - DependencyGenerator DependencyGenerator = DependencyGenerator_init(numattrs, k); + DependencyGenerator DependencyGenerator = DependencyGenerator_init(data->nattnums, k); /* generate all possible variations of k values (out of n) */ while ((dependency = DependencyGenerator_next(DependencyGenerator))) @@ -433,8 +378,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, MVDependency *d; /* compute how valid the dependency seems */ - degree = dependency_degree(numrows, rows, exprs, k, dependency, - stats, attrs); + degree = dependency_degree(data, k, dependency); /* * if the dependency seems entirely invalid, don't store it @@ -449,7 +393,7 @@ statext_dependencies_build(int numrows, HeapTuple *rows, d->degree = degree; d->nattributes = k; for (i = 0; i < k; i++) - d->attributes[i] = attnums[dependency[i]]; + d->attributes[i] = data->attnums[dependency[i]]; /* initialize the list of dependencies */ if (dependencies == NULL) @@ -477,8 +421,6 @@ statext_dependencies_build(int numrows, HeapTuple *rows, DependencyGenerator_free(DependencyGenerator); } - pfree(attrs); - return dependencies; } diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c index 95b2cc683e..5b3fa523e9 100644 --- a/src/backend/statistics/extended_stats.c +++ b/src/backend/statistics/extended_stats.c @@ -96,8 +96,11 @@ static Datum serialize_expr_stats(AnlExprData *exprdata, int nexprs); static Datum expr_fetch_func(VacAttrStatsP stats, int rownum, bool *isNull); static AnlExprData *build_expr_data(List *exprs); static VacAttrStats *examine_expression(Node *expr); -static ExprInfo *evaluate_expressions(Relation rel, List *exprs, - int numrows, HeapTuple *rows); + +static StatBuildData *make_build_data(Relation onerel, StatExtEntry *stat, + int numrows, HeapTuple *rows, + VacAttrStats **stats); + /* * Compute requested extended stats, using the rows sampled for the plain @@ -156,7 +159,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, VacAttrStats **stats; ListCell *lc2; int stattarget; - ExprInfo *exprs; + StatBuildData *data; /* * Check if we can build these stats based on the column analyzed. If @@ -191,7 +194,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, continue; /* evaluate expressions (if the statistics has any) */ - exprs = evaluate_expressions(onerel, stat->exprs, numrows, rows); + data = make_build_data(onerel, stat, numrows, rows, stats); /* compute statistic of each requested type */ foreach(lc2, stat->types) @@ -199,16 +202,11 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, char t = (char) lfirst_int(lc2); if (t == STATS_EXT_NDISTINCT) - ndistinct = statext_ndistinct_build(totalrows, numrows, rows, - exprs, stat->columns, - stats); + ndistinct = statext_ndistinct_build(totalrows, data); else if (t == STATS_EXT_DEPENDENCIES) - dependencies = statext_dependencies_build(numrows, rows, - exprs, stat->columns, - stats); + dependencies = statext_dependencies_build(data); else if (t == STATS_EXT_MCV) - mcv = statext_mcv_build(numrows, rows, exprs, stat->columns, - stats, totalrows, stattarget); + mcv = statext_mcv_build(data, totalrows, stattarget); else if (t == STATS_EXT_EXPRESSIONS) { AnlExprData *exprdata; @@ -236,7 +234,8 @@ BuildRelationExtStatistics(Relation onerel, double totalrows, pgstat_progress_update_param(PROGRESS_ANALYZE_EXT_STATS_COMPUTED, ++ext_cnt); - pfree(exprs); + /* free the build data (allocated as a single chunk) */ + pfree(data); } table_close(pg_stext, RowExclusiveLock); @@ -937,30 +936,31 @@ build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs) * can simply pfree the return value to release all of it. */ SortItem * -build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, - TupleDesc tdesc, MultiSortSupport mss, +build_sorted_items(StatBuildData *data, int *nitems, + MultiSortSupport mss, int numattrs, AttrNumber *attnums) { int i, j, len, - idx; - int nvalues = numrows * numattrs; + nrows; + int nvalues = data->numrows * numattrs; SortItem *items; Datum *values; bool *isnull; char *ptr; + int *typlen; /* Compute the total amount of memory we need (both items and values). */ - len = numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); + len = data->numrows * sizeof(SortItem) + nvalues * (sizeof(Datum) + sizeof(bool)); /* Allocate the memory and split it into the pieces. */ ptr = palloc0(len); /* items to sort */ items = (SortItem *) ptr; - ptr += numrows * sizeof(SortItem); + ptr += data->numrows * sizeof(SortItem); /* values and null flags */ values = (Datum *) ptr; @@ -973,13 +973,24 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, Assert((ptr - (char *) items) == len); /* fix the pointers to Datum and bool arrays */ - idx = 0; - for (i = 0; i < numrows; i++) + nrows = 0; + for (i = 0; i < data->numrows; i++) { - bool toowide = false; + items[nrows].values = &values[nrows * numattrs]; + items[nrows].isnull = &isnull[nrows * numattrs]; + + nrows++; + } + + /* build a local cache of typlen for all attributes */ + typlen = (int *) palloc(sizeof(int) * data->nattnums); + for (i = 0; i < data->nattnums; i++) + typlen[i] = get_typlen(data->stats[i]->attrtypid); - items[idx].values = &values[idx * numattrs]; - items[idx].isnull = &isnull[idx * numattrs]; + nrows = 0; + for (i = 0; i < data->numrows; i++) + { + bool toowide = false; /* load the values/null flags from sample rows */ for (j = 0; j < numattrs; j++) @@ -989,22 +1000,20 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, int attlen; AttrNumber attnum = attnums[j]; - if (AttrNumberIsForUserDefinedAttr(attnum)) + int idx; + + /* match attnum to the pre-calculated data */ + for (idx = 0; idx < data->nattnums; idx++) { - value = heap_getattr(rows[i], attnum, tdesc, &isnull); - attlen = TupleDescAttr(tdesc, attnum - 1)->attlen; + if (attnum == data->attnums[idx]) + break; } - else - { - int idx = -(attnums[j] + 1); - - Assert((idx >= 0) && (idx < exprs->nexprs)); - value = exprs->values[idx][i]; - isnull = exprs->nulls[idx][i]; + Assert(idx < data->nattnums); - attlen = get_typlen(exprs->types[idx]); - } + value = data->values[idx][i]; + isnull = data->nulls[idx][i]; + attlen = typlen[idx]; /* * If this is a varlena value, check if it's too wide and if yes @@ -1026,21 +1035,21 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, value = PointerGetDatum(PG_DETOAST_DATUM(value)); } - items[idx].values[j] = value; - items[idx].isnull[j] = isnull; + items[nrows].values[j] = value; + items[nrows].isnull[j] = isnull; } if (toowide) continue; - idx++; + nrows++; } /* store the actual number of items (ignoring the too-wide ones) */ - *nitems = idx; + *nitems = nrows; /* all items were too wide */ - if (idx == 0) + if (nrows == 0) { /* everything is allocated as a single chunk */ pfree(items); @@ -1048,7 +1057,7 @@ build_sorted_items(int numrows, int *nitems, HeapTuple *rows, ExprInfo *exprs, } /* do the sort, using the multi-sort */ - qsort_arg((void *) items, idx, sizeof(SortItem), + qsort_arg((void *) items, nrows, sizeof(SortItem), multi_sort_compare, mss); return items; @@ -2434,59 +2443,61 @@ statext_expressions_load(Oid stxoid, int idx) * all the requested statistics types. This matters especially for * expensive expressions, of course. */ -static ExprInfo * -evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) +static StatBuildData * +make_build_data(Relation rel, StatExtEntry *stat, int numrows, HeapTuple *rows, + VacAttrStats **stats) { /* evaluated expressions */ - ExprInfo *result; + StatBuildData *result; char *ptr; Size len; int i; + int k; int idx; TupleTableSlot *slot; EState *estate; ExprContext *econtext; List *exprstates = NIL; - int nexprs = list_length(exprs); + int nkeys = bms_num_members(stat->columns) + list_length(stat->exprs); ListCell *lc; /* allocate everything as a single chunk, so we can free it easily */ - len = MAXALIGN(sizeof(ExprInfo)); - len += MAXALIGN(sizeof(Oid) * nexprs); /* types */ - len += MAXALIGN(sizeof(Oid) * nexprs); /* collations */ + len = MAXALIGN(sizeof(StatBuildData)); + len += MAXALIGN(sizeof(AttrNumber) * nkeys); /* attnums */ + len += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* stats */ /* values */ - len += MAXALIGN(sizeof(Datum *) * nexprs); - len += nexprs * MAXALIGN(sizeof(Datum) * numrows); + len += MAXALIGN(sizeof(Datum *) * nkeys); + len += nkeys * MAXALIGN(sizeof(Datum) * numrows); /* nulls */ - len += MAXALIGN(sizeof(bool *) * nexprs); - len += nexprs * MAXALIGN(sizeof(bool) * numrows); + len += MAXALIGN(sizeof(bool *) * nkeys); + len += nkeys * MAXALIGN(sizeof(bool) * numrows); ptr = palloc(len); /* set the pointers */ - result = (ExprInfo *) ptr; - ptr += MAXALIGN(sizeof(ExprInfo)); + result = (StatBuildData *) ptr; + ptr += MAXALIGN(sizeof(StatBuildData)); - /* types */ - result->types = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* attnums */ + result->attnums = (AttrNumber *) ptr; + ptr += MAXALIGN(sizeof(AttrNumber) * nkeys); - /* collations */ - result->collations = (Oid *) ptr; - ptr += MAXALIGN(sizeof(Oid) * nexprs); + /* stats */ + result->stats = (VacAttrStats **) ptr; + ptr += MAXALIGN(sizeof(VacAttrStats *) * nkeys); /* values */ result->values = (Datum **) ptr; - ptr += MAXALIGN(sizeof(Datum *) * nexprs); + ptr += MAXALIGN(sizeof(Datum *) * nkeys); /* nulls */ result->nulls = (bool **) ptr; - ptr += MAXALIGN(sizeof(bool *) * nexprs); + ptr += MAXALIGN(sizeof(bool *) * nkeys); - for (i = 0; i < nexprs; i++) + for (i = 0; i < nkeys; i++) { result->values[i] = (Datum *) ptr; ptr += MAXALIGN(sizeof(Datum) * numrows); @@ -2497,17 +2508,46 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) Assert((ptr - (char *) result) == len); - result->nexprs = list_length(exprs); + /* we have it allocated, so let's fill the values */ + result->nattnums = nkeys; + result->numrows = numrows; + /* fill the attribute info - first attributes, then expressions */ idx = 0; - foreach (lc, exprs) + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->attnums[idx] = k; + result->stats[idx] = stats[idx]; + + idx++; + } + + k = -1; + foreach (lc, stat->exprs) { Node *expr = (Node *) lfirst(lc); - result->types[idx] = exprType(expr); - result->collations[idx] = exprCollation(expr); + result->attnums[idx] = k; + result->stats[idx] = examine_expression(expr); idx++; + k--; + } + + /* first extract values for all the regular attributes */ + for (i = 0; i < numrows; i++) + { + idx = 0; + k = -1; + while ((k = bms_next_member(stat->columns, k)) >= 0) + { + result->values[idx][i] = heap_getattr(rows[i], k, + result->stats[idx]->tupDesc, + &result->nulls[idx][i]); + + idx++; + } } /* @@ -2526,7 +2566,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) econtext->ecxt_scantuple = slot; /* Set up expression evaluation state */ - exprstates = ExecPrepareExprList(exprs, estate); + exprstates = ExecPrepareExprList(stat->exprs, estate); for (i = 0; i < numrows; i++) { @@ -2539,7 +2579,7 @@ evaluate_expressions(Relation rel, List *exprs, int numrows, HeapTuple *rows) /* Set up for predicate or expression evaluation */ ExecStoreHeapTuple(rows[i], slot, false); - idx = 0; + idx = bms_num_members(stat->columns); foreach (lc, exprstates) { Datum datum; diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 323d476814..844ba6f71f 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -74,8 +74,7 @@ ((ndims) * sizeof(DimensionInfo)) + \ ((nitems) * ITEM_SIZE(ndims))) -static MultiSortSupport build_mss(VacAttrStats **stats, int numattrs, - ExprInfo *exprs); +static MultiSortSupport build_mss(StatBuildData *data); static SortItem *build_distinct_groups(int numrows, SortItem *items, MultiSortSupport mss, int *ndistinct); @@ -182,16 +181,11 @@ get_mincount_for_mcv_list(int samplerows, double totalrows) * */ MCVList * -statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, - Bitmapset *attrs, VacAttrStats **stats, - double totalrows, int stattarget) +statext_mcv_build(StatBuildData *data, double totalrows, int stattarget) { int i, - k, - numattrs, ngroups, nitems; - AttrNumber *attnums; double mincount; SortItem *items; SortItem *groups; @@ -199,38 +193,11 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, MultiSortSupport mss; /* comparator for all the columns */ - mss = build_mss(stats, bms_num_members(attrs), exprs); - - /* - * treat expressions as special attributes with high attnums - * - * XXX We do this after build_mss, because that expects the bitmapset - * to only contain simple attributes (with a matching VacAttrStats) - */ - - /* - * Transform the bms into an array, to make accessing i-th member easier. - */ - attnums = (AttrNumber *) palloc(sizeof(AttrNumber) * (bms_num_members(attrs) + exprs->nexprs)); - - numattrs = 0; - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - attnums[numattrs++] = k; - - /* treat expressions as attributes with negative attnums */ - for (i = 0; i < exprs->nexprs; i++) - attnums[numattrs++] = -(i+1); - - Assert(numattrs >= 2); - Assert(numattrs == (bms_num_members(attrs) + exprs->nexprs)); - + mss = build_mss(data); /* sort the rows */ - items = build_sorted_items(numrows, &nitems, rows, exprs, - stats[0]->tupDesc, mss, numattrs, attnums); + items = build_sorted_items(data, &nitems, mss, + data->nattnums, data->attnums); if (!items) return NULL; @@ -265,7 +232,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * using get_mincount_for_mcv_list() and then keep all items that seem to * be more common than that. */ - mincount = get_mincount_for_mcv_list(numrows, totalrows); + mincount = get_mincount_for_mcv_list(data->numrows, totalrows); /* * Walk the groups until we find the first group with a count below the @@ -301,7 +268,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, + sizeof(SortSupportData)); /* compute frequencies for values in each column */ - nfreqs = (int *) palloc0(sizeof(int) * numattrs); + nfreqs = (int *) palloc0(sizeof(int) * data->nattnums); freqs = build_column_frequencies(groups, ngroups, mss, nfreqs); /* @@ -312,12 +279,12 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, mcvlist->magic = STATS_MCV_MAGIC; mcvlist->type = STATS_MCV_TYPE_BASIC; - mcvlist->ndimensions = numattrs; + mcvlist->ndimensions = data->nattnums; mcvlist->nitems = nitems; /* store info about data type OIDs */ - for (i = 0; i < numattrs; i++) - mcvlist->types[i] = stats[i]->attrtypid; + for (i = 0; i < data->nattnums; i++) + mcvlist->types[i] = data->stats[i]->attrtypid; /* Copy the first chunk of groups into the result. */ for (i = 0; i < nitems; i++) @@ -325,22 +292,22 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, /* just pointer to the proper place in the list */ MCVItem *item = &mcvlist->items[i]; - item->values = (Datum *) palloc(sizeof(Datum) * numattrs); - item->isnull = (bool *) palloc(sizeof(bool) * numattrs); + item->values = (Datum *) palloc(sizeof(Datum) * data->nattnums); + item->isnull = (bool *) palloc(sizeof(bool) * data->nattnums); /* copy values for the group */ - memcpy(item->values, groups[i].values, sizeof(Datum) * numattrs); - memcpy(item->isnull, groups[i].isnull, sizeof(bool) * numattrs); + memcpy(item->values, groups[i].values, sizeof(Datum) * data->nattnums); + memcpy(item->isnull, groups[i].isnull, sizeof(bool) * data->nattnums); /* groups should be sorted by frequency in descending order */ Assert((i == 0) || (groups[i - 1].count >= groups[i].count)); /* group frequency */ - item->frequency = (double) groups[i].count / numrows; + item->frequency = (double) groups[i].count / data->numrows; /* base frequency, if the attributes were independent */ item->base_frequency = 1.0; - for (j = 0; j < numattrs; j++) + for (j = 0; j < data->nattnums; j++) { SortItem *freq; @@ -356,7 +323,7 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, sizeof(SortItem), multi_sort_compare, tmp); - item->base_frequency *= ((double) freq->count) / numrows; + item->base_frequency *= ((double) freq->count) / data->numrows; } } @@ -375,17 +342,17 @@ statext_mcv_build(int numrows, HeapTuple *rows, ExprInfo *exprs, * build MultiSortSupport for the attributes passed in attrs */ static MultiSortSupport -build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) +build_mss(StatBuildData *data) { int i; /* Sort by multiple columns (using array of SortSupport) */ - MultiSortSupport mss = multi_sort_init(numattrs + exprs->nexprs); + MultiSortSupport mss = multi_sort_init(data->nattnums); /* prepare the sort functions for all the attributes */ - for (i = 0; i < numattrs; i++) + for (i = 0; i < data->nattnums; i++) { - VacAttrStats *colstat = stats[i]; + VacAttrStats *colstat = data->stats[i]; TypeCacheEntry *type; type = lookup_type_cache(colstat->attrtypid, TYPECACHE_LT_OPR); @@ -396,20 +363,6 @@ build_mss(VacAttrStats **stats, int numattrs, ExprInfo *exprs) multi_sort_add_dimension(mss, i, type->lt_opr, colstat->attrcollid); } - /* prepare the sort functions for all the expressions */ - for (i = 0; i < exprs->nexprs; i++) - { - TypeCacheEntry *type; - - type = lookup_type_cache(exprs->types[i], TYPECACHE_LT_OPR); - if (type->lt_opr == InvalidOid) /* shouldn't happen */ - elog(ERROR, "cache lookup failed for ordering operator for type %u", - exprs->types[i]); - - multi_sort_add_dimension(mss, numattrs + i, type->lt_opr, - exprs->collations[i]); - } - return mss; } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 5e796e7123..7ca59d9785 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -36,9 +36,7 @@ #include "utils/syscache.h" #include "utils/typcache.h" -static double ndistinct_for_combination(double totalrows, int numrows, - HeapTuple *rows, ExprInfo *exprs, - int nattrs, VacAttrStats **stats, +static double ndistinct_for_combination(double totalrows, StatBuildData *data, int k, int *combination); static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); @@ -88,17 +86,12 @@ static void generate_combinations(CombinationGenerator *state); * allow using Bitmapsets. */ MVNDistinct * -statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats) +statext_ndistinct_build(double totalrows, StatBuildData *data) { MVNDistinct *result; - int i; int k; int itemcnt; - int numattrs = bms_num_members(attrs); - int numcombs = num_combinations(numattrs + exprs->nexprs); - Bitmapset *tmp = NULL; + int numcombs = num_combinations(data->nattnums); result = palloc(offsetof(MVNDistinct, items) + numcombs * sizeof(MVNDistinctItem)); @@ -106,38 +99,14 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, result->type = STATS_NDISTINCT_TYPE_BASIC; result->nitems = numcombs; - /* - * Treat expressions as system attributes with negative attnums, - * but offset everything by number of expressions. - */ - for (i = 0; i < exprs->nexprs; i++) - { - AttrNumber attnum = -(i + 1); - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* regular attributes */ - k = -1; - while ((k = bms_next_member(attrs, k)) >= 0) - { - AttrNumber attnum = k; - tmp = bms_add_member(tmp, attnum + exprs->nexprs); - } - - /* use the newly built bitmapset */ - attrs = tmp; - - /* make sure there were no clashes */ - Assert(bms_num_members(attrs) == numattrs + exprs->nexprs); - itemcnt = 0; - for (k = 2; k <= bms_num_members(attrs); k++) + for (k = 2; k <= data->nattnums; k++) { int *combination; CombinationGenerator *generator; /* generate combinations of K out of N elements */ - generator = generator_init(bms_num_members(attrs), k); + generator = generator_init(data->nattnums, k); while ((combination = generator_next(generator))) { @@ -147,36 +116,16 @@ statext_ndistinct_build(double totalrows, int numrows, HeapTuple *rows, item->attributes = palloc(sizeof(AttrNumber) * k); item->nattributes = k; + /* translate the indexes to attnums */ for (j = 0; j < k; j++) { - AttrNumber attnum = InvalidAttrNumber; - - /* - * The expressions have negative attnums, so even with the - * offset are before regular attributes. So the first chunk - * of indexes are for expressions. - */ - if (combination[j] >= exprs->nexprs) - attnum - = stats[combination[j] - exprs->nexprs]->attr->attnum; - else - { - /* make sure the expression index is valid */ - Assert(combination[j] >= 0); - Assert(combination[j] < exprs->nexprs); - - attnum = -(combination[j] + 1); - } - - Assert(attnum != InvalidAttrNumber); - - item->attributes[j] = attnum; + item->attributes[j] = data->attnums[combination[j]]; + + Assert(AttributeNumberIsValid(item->attributes[j])); } item->ndistinct = - ndistinct_for_combination(totalrows, numrows, rows, - exprs, numattrs, - stats, k, combination); + ndistinct_for_combination(totalrows, data, k, combination); itemcnt++; Assert(itemcnt <= result->nitems); @@ -471,9 +420,8 @@ pg_ndistinct_send(PG_FUNCTION_ARGS) * combination of multiple columns. */ static double -ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, - ExprInfo *exprs, int nattrs, - VacAttrStats **stats, int k, int *combination) +ndistinct_for_combination(double totalrows, StatBuildData *data, + int k, int *combination) { int i, j; @@ -493,11 +441,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, * using the specified column combination as dimensions. We could try to * sort in place, but it'd probably be more complex and bug-prone. */ - items = (SortItem *) palloc(numrows * sizeof(SortItem)); - values = (Datum *) palloc0(sizeof(Datum) * numrows * k); - isnull = (bool *) palloc0(sizeof(bool) * numrows * k); + items = (SortItem *) palloc(data->numrows * sizeof(SortItem)); + values = (Datum *) palloc0(sizeof(Datum) * data->numrows * k); + isnull = (bool *) palloc0(sizeof(bool) * data->numrows * k); - for (i = 0; i < numrows; i++) + for (i = 0; i < data->numrows; i++) { items[i].values = &values[i * k]; items[i].isnull = &isnull[i * k]; @@ -514,24 +462,11 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, { Oid typid; TypeCacheEntry *type; - AttrNumber attnum = InvalidAttrNumber; - TupleDesc tdesc = NULL; Oid collid = InvalidOid; + VacAttrStats *colstat = data->stats[combination[i]]; - /* first nexprs indexes are for expressions, then regular attributes */ - if (combination[i] >= exprs->nexprs) - { - VacAttrStats *colstat = stats[combination[i] - exprs->nexprs]; - typid = colstat->attrtypid; - attnum = colstat->attr->attnum; - collid = colstat->attrcollid; - tdesc = colstat->tupDesc; - } - else - { - typid = exprs->types[combination[i]]; - collid = exprs->collations[combination[i]]; - } + typid = colstat->attrtypid; + collid = colstat->attrcollid; type = lookup_type_cache(typid, TYPECACHE_LT_OPR); if (type->lt_opr == InvalidOid) /* shouldn't happen */ @@ -542,38 +477,15 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, multi_sort_add_dimension(mss, i, type->lt_opr, collid); /* accumulate all the data for this dimension into the arrays */ - for (j = 0; j < numrows; j++) + for (j = 0; j < data->numrows; j++) { - /* - * The first exprs indexes identify expressions, higher indexes - * are for plain attributes. - * - * XXX This seems a bit strange that we don't offset the (i) - * in any way? - */ - if (combination[i] >= exprs->nexprs) - items[j].values[i] = - heap_getattr(rows[j], - attnum, - tdesc, - &items[j].isnull[i]); - else - { - /* we know the first nexprs expressions are expressions, - * and the value is directly the expression index */ - int idx = combination[i]; - - /* make sure the expression index is valid */ - Assert((idx >= 0) && (idx < exprs->nexprs)); - - items[j].values[i] = exprs->values[idx][j]; - items[j].isnull[i] = exprs->nulls[idx][j]; - } + items[j].values[i] = data->values[combination[i]][j]; + items[j].isnull[i] = data->nulls[combination[i]][j]; } } /* We can sort the array now ... */ - qsort_arg((void *) items, numrows, sizeof(SortItem), + qsort_arg((void *) items, data->numrows, sizeof(SortItem), multi_sort_compare, mss); /* ... and count the number of distinct combinations */ @@ -581,7 +493,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, f1 = 0; cnt = 1; d = 1; - for (i = 1; i < numrows; i++) + for (i = 1; i < data->numrows; i++) { if (multi_sort_compare(&items[i], &items[i - 1], mss) != 0) { @@ -598,7 +510,7 @@ ndistinct_for_combination(double totalrows, int numrows, HeapTuple *rows, if (cnt == 1) f1 += 1; - return estimate_ndistinct(totalrows, numrows, d, f1); + return estimate_ndistinct(totalrows, data->numrows, d, f1); } /* The Duj1 estimator (already used in analyze.c). */ diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 1f09799deb..7acf82aa0e 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -57,35 +57,26 @@ typedef struct SortItem int count; } SortItem; -/* - * Used to pass pre-computed information about expressions the stats - * object is defined on. - */ -typedef struct ExprInfo -{ - int nexprs; /* number of expressions */ - Oid *collations; /* collation for each expression */ - Oid *types; /* type of each expression */ - Datum **values; /* values for each expression */ - bool **nulls; /* nulls for each expression */ -} ExprInfo; - -extern MVNDistinct *statext_ndistinct_build(double totalrows, - int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +/* a unified representation of the data the statistics is built on */ +typedef struct StatBuildData { + int numrows; + int nattnums; + AttrNumber *attnums; + VacAttrStats **stats; + Datum **values; + bool **nulls; +} StatBuildData; + + +extern MVNDistinct *statext_ndistinct_build(double totalrows, StatBuildData *data); extern bytea *statext_ndistinct_serialize(MVNDistinct *ndistinct); extern MVNDistinct *statext_ndistinct_deserialize(bytea *data); -extern MVDependencies *statext_dependencies_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats); +extern MVDependencies *statext_dependencies_build(StatBuildData *data); extern bytea *statext_dependencies_serialize(MVDependencies *dependencies); extern MVDependencies *statext_dependencies_deserialize(bytea *data); -extern MCVList *statext_mcv_build(int numrows, HeapTuple *rows, - ExprInfo *exprs, Bitmapset *attrs, - VacAttrStats **stats, +extern MCVList *statext_mcv_build(StatBuildData *data, double totalrows, int stattarget); extern bytea *statext_mcv_serialize(MCVList *mcv, VacAttrStats **stats); extern MCVList *statext_mcv_deserialize(bytea *data); @@ -108,8 +99,7 @@ extern void *bsearch_arg(const void *key, const void *base, extern AttrNumber *build_attnums_array(Bitmapset *attrs, int nexprs, int *numattrs); -extern SortItem *build_sorted_items(int numrows, int *nitems, HeapTuple *rows, - ExprInfo *exprs, TupleDesc tdesc, +extern SortItem *build_sorted_items(StatBuildData *data, int *nitems, MultiSortSupport mss, int numattrs, AttrNumber *attnums); -- 2.26.2 --------------614DDB87AFFED893713AC0E9-- ^ permalink raw reply [nested|flat] 28+ messages in thread
* Re: Add an option to skip loading missing publication to avoid logical replication failure @ 2025-02-18 11:23 Amit Kapila <[email protected]> 0 siblings, 1 reply; 28+ messages in thread From: Amit Kapila @ 2025-02-18 11:23 UTC (permalink / raw) To: vignesh C <[email protected]>; +Cc: PostgreSQL Hackers <[email protected]> On Tue, Feb 18, 2025 at 2:24 PM vignesh C <[email protected]> wrote: > > On Fri, 14 Feb 2025 at 15:36, Amit Kapila <[email protected]> wrote: > > > > Did you try to measure the performance impact of this change? We can > > try a few cases where DDL and DMLs are involved, missing publication > > (drop publication and recreate after a varying number of records to > > check the impact). > > Since we don't have an exact scenario to compare with the patch > (because, in the current HEAD, when the publication is missing, an > error is thrown and the walsender/worker restarts), I compared the > positive case, where records are successfully replicated to the > subscriber, as shown below. For the scenario with the patch, I ran the > same test, where the publication is dropped before the insert, > allowing the walsender to check whether the publication is present. > The test results, which represent the median of 7 runs and the > execution run is in milliseconds, are provided below: > > Brach/records | 100 | 1000 | 10000 | 100000 | 1000000 > Head | 1.214 | 2.548 | 10.823 | 90.3 | 951.833 > Patch | 1.215 | 2.5485 | 10.8545 | 90.94 | 955.134 > % diff | 0.082 | 0.020 | 0.291 | 0.704 | 0.347 > > I noticed that the test run with patches is very negligible. The > scripts used for execution are attached. > You have used the synchronous_standby_name to evaluate the performance which covers other parts of replication than the logical decoding. It would be better to test using pg_recvlogical. -- With Regards, Amit Kapila. ^ permalink raw reply [nested|flat] 28+ messages in thread
* Re: Add an option to skip loading missing publication to avoid logical replication failure @ 2025-02-25 10:02 vignesh C <[email protected]> parent: Amit Kapila <[email protected]> 0 siblings, 1 reply; 28+ messages in thread From: vignesh C @ 2025-02-25 10:02 UTC (permalink / raw) To: Amit Kapila <[email protected]>; +Cc: PostgreSQL Hackers <[email protected]> On Tue, 18 Feb 2025 at 16:53, Amit Kapila <[email protected]> wrote: > > On Tue, Feb 18, 2025 at 2:24 PM vignesh C <[email protected]> wrote: > > > > On Fri, 14 Feb 2025 at 15:36, Amit Kapila <[email protected]> wrote: > > > > > > Did you try to measure the performance impact of this change? We can > > > try a few cases where DDL and DMLs are involved, missing publication > > > (drop publication and recreate after a varying number of records to > > > check the impact). > > > > Since we don't have an exact scenario to compare with the patch > > (because, in the current HEAD, when the publication is missing, an > > error is thrown and the walsender/worker restarts), I compared the > > positive case, where records are successfully replicated to the > > subscriber, as shown below. For the scenario with the patch, I ran the > > same test, where the publication is dropped before the insert, > > allowing the walsender to check whether the publication is present. > > The test results, which represent the median of 7 runs and the > > execution run is in milliseconds, are provided below: > > > > Brach/records | 100 | 1000 | 10000 | 100000 | 1000000 > > Head | 1.214 | 2.548 | 10.823 | 90.3 | 951.833 > > Patch | 1.215 | 2.5485 | 10.8545 | 90.94 | 955.134 > > % diff | 0.082 | 0.020 | 0.291 | 0.704 | 0.347 > > > > I noticed that the test run with patches is very negligible. The > > scripts used for execution are attached. > > > > You have used the synchronous_standby_name to evaluate the performance > which covers other parts of replication than the logical decoding. It > would be better to test using pg_recvlogical. Here are the test runs with pg_recvlogical, the test results, which represent the median of 10 runs and the execution run is in milliseconds, are provided below: Brach/records | 100 | 1000 | 10000 | 100000 | 1000000 Head | 9.95 | 15.26 | 62.62 | 536.57 | 8480.83 Patch | 9.218 | 10.32 | 23.05 | 143.83 | 4852.43 % diff | 7.356 | 32.38 | 63.19 | 73.193| 42.783 We observe that test execution with the patch performs better between 7.35 percent to 73.19 percent. This is because, in HEAD, after loading and verifying that the publication is valid, it must continue processing to output the change. In contrast, with the patch, outputting the change is skipped since the publication does not exist. The attached script has the script that was used for testing. Here the NUM_RECORDS count should be changed accordingly for each of the tests and while running the test with the patch change uncomment the drop publication command. Regards, Vignesh Attachments: [text/x-sh] test_pg_recvlogical.sh (1.2K, ../../CALDaNm13A+KyNu47YMyu+tL4XeombTHzZt3fpD1QLxh_Mtw+WQ@mail.gmail.com/2-test_pg_recvlogical.sh) download | inline: #!/bin/bash ##### SLOT_NAME=test PLUGIN_NAME=pgoutput NUM_RECORDS=100 LOOP=10 ##### for i in `seq 1 $LOOP` do # Cleanup previous result pg_ctl stop -D data rm -rf data logfile # Initialize an instance initdb -D data -U postgres -c wal_level=logical # Start the instance pg_ctl -D data -l logfile start # Create a table psql -U postgres -c "CREATE TABLE foo (id int);" psql -U postgres -c "CREATE publication pub1 for table foo;" #psql -U postgres -c "drop publication pub1;" # Create a replication slot psql -U postgres -c "SELECT * FROM pg_create_logical_replication_slot('$SLOT_NAME', '$PLUGIN_NAME')" # Insert tuples (this transaction will be decoded) psql -U postgres -c "INSERT INTO foo VALUES (generate_series(1, $NUM_RECORDS))" # Confirm current WAL location WAL_POS=$(psql -qtAX -U postgres -c "SELECT * FROM pg_current_wal_lsn()") t1=$(($(date +%s%N)/1000)) echo $t1 > run_${i}.dat # Run pg_recvlogical till the current WAL location (time pg_recvlogical -d postgres -U postgres --start -S $SLOT_NAME -E $WAL_POS -f - -o publication_names='pub1' -o proto_version=4 ) &>> run_${i}.dat t2=$(($(date +%s%N)/1000)) echo $t2 >> run_${i}.dat t3=$((t2-t1)) echo "execution time=$t3" >> run_${i}.dat done ^ permalink raw reply [nested|flat] 28+ messages in thread
* Re: Add an option to skip loading missing publication to avoid logical replication failure @ 2025-03-03 09:00 vignesh C <[email protected]> parent: vignesh C <[email protected]> 0 siblings, 1 reply; 28+ messages in thread From: vignesh C @ 2025-03-03 09:00 UTC (permalink / raw) To: Amit Kapila <[email protected]>; +Cc: PostgreSQL Hackers <[email protected]> On Tue, 25 Feb 2025 at 15:32, vignesh C <[email protected]> wrote: > > The attached script has the script that was used for testing. Here the > NUM_RECORDS count should be changed accordingly for each of the tests > and while running the test with the patch change uncomment the drop > publication command. I have done further analysis on the test and changed the test to compare it better with HEAD. The execution time is in milliseconds. Brach/records | 100 | 1000 | 10000 | 100000 | 1000000 Head | 10.43 | 15.86 | 64.44 | 550.56 | 8991.04 Patch | 11.35 | 17.26 | 73.50 | 640.21 | 10104.72 % diff | -8.82 | -8.85 | -14.08 | -16.28 | -12.38 There is a performance degradation in the range of 8.8 to 16.2 percent. Test Details (With Head): a) Create two publications for the same table. b) Insert the records listed in the table below. c) Use pg_recvlogical to capture the changes. Test Details (With Patch): a) Create two publications for the same table.b) Drop one publication(to check the impact of skip missing publication), ensuring that changes from the remaining publication continue to be captured. c) Insert the records listed in the table below.d) Use pg_recvlogical to capture the changes. The performance degradation is in the range of 8.8 to 16.2 percentage.The script used for the testing is attached, while running with patch the drop publication command in script should be uncommented and the record count should be changed for each of the run. Also I changed the patch so that we need not execute the get_rel_sync_entry code flow for every record in case of missing publication case and to do so only in case when the publications have changed. The attached v4 version patch has the changes for the same. Regards, Vignesh Attachments: [application/octet-stream] v4-0001-Fix-logical-replication-breakage-after.patch (5.2K, ../../CALDaNm1Lg1YOtY+Sy2his=ikHsQYYQ1pWN620QiBZDBdbsjMHw@mail.gmail.com/2-v4-0001-Fix-logical-replication-breakage-after.patch) download | inline diff: From f2d31bdf99240288349a8a3acebc78391c8275f4 Mon Sep 17 00:00:00 2001 From: Vignesh <[email protected]> Date: Mon, 3 Mar 2025 11:54:27 +0530 Subject: [PATCH v4] Fix logical replication breakage after ALTER SUBSCRIPTION ... SET PUBLICATION Altering a subscription with `ALTER SUBSCRIPTION ... SET PUBLICATION` could cause logical replication to break under certain conditions. When the apply worker restarts after executing SET PUBLICATION, it continues using the existing replication slot and origin. If the replication origin was not updated before the restart, the WAL start location could point to a position prior to the existence of the specified publication, leading to persistent start of apply worker and reporting errors. This patch skips loading the publication if the publication does not exist and loads the publication and updates the relation entry when the publication gets created. Discussion: https://www.postgresql.org/message-id/flat/CALDaNm0-n8FGAorM%2BbTxkzn%2BAOUyx5%3DL_XmnvOP6T24%2B-NcBKg%40mail.gmail.com Discussion: https://www.postgresql.org/message-id/CAA4eK1+T-ETXeRM4DHWzGxBpKafLCp__5bPA_QZfFQp7-0wj4Q@mail.gmail.com --- src/backend/replication/pgoutput/pgoutput.c | 74 ++++++++++++++++++--- 1 file changed, 66 insertions(+), 8 deletions(-) diff --git a/src/backend/replication/pgoutput/pgoutput.c b/src/backend/replication/pgoutput/pgoutput.c index 7d464f656aa..64c31cc2758 100644 --- a/src/backend/replication/pgoutput/pgoutput.c +++ b/src/backend/replication/pgoutput/pgoutput.c @@ -82,7 +82,7 @@ static void pgoutput_stream_prepare_txn(LogicalDecodingContext *ctx, static bool publications_valid; -static List *LoadPublications(List *pubnames); +static List *LoadPublications(List *pubnames, bool *skipped); static void publication_invalidation_cb(Datum arg, int cacheid, uint32 hashvalue); static void send_repl_origin(LogicalDecodingContext *ctx, @@ -1762,9 +1762,13 @@ pgoutput_shutdown(LogicalDecodingContext *ctx) /* * Load publications from the list of publication names. + * + * Here, we just skip the publications that don't exist yet. 'skipped' + * will be true if we find any publication from the given list that doesn't + * exist. */ static List * -LoadPublications(List *pubnames) +LoadPublications(List *pubnames, bool *skipped) { List *result = NIL; ListCell *lc; @@ -1772,9 +1776,15 @@ LoadPublications(List *pubnames) foreach(lc, pubnames) { char *pubname = (char *) lfirst(lc); - Publication *pub = GetPublicationByName(pubname, false); + Publication *pub = GetPublicationByName(pubname, true); - result = lappend(result, pub); + if (pub) + result = lappend(result, pub); + else + { + elog(DEBUG1, "skipped loading publication: %s", pubname); + *skipped = true; + } } return result; @@ -2026,6 +2036,7 @@ get_rel_sync_entry(PGOutputData *data, Relation relation) bool found; MemoryContext oldctx; Oid relid = RelationGetRelid(relation); + bool publications_updated = false; Assert(RelationSyncCache != NULL); @@ -2053,8 +2064,48 @@ get_rel_sync_entry(PGOutputData *data, Relation relation) entry->attrmap = NULL; } - /* Validate the entry */ - if (!entry->replicate_valid) + /* + * If the publication is invalid, check for updates. + * This optimization ensures that the next block, which queries the system + * tables and builds the relation entry, runs only if a new publication was + * created. + */ + if (!publications_valid && data->publications) + { + bool skipped_pub = false; + List *publications; + + publications = LoadPublications(data->publication_names, &skipped_pub); + + /* Check if any new publications have been created. */ + foreach_ptr(Publication, pub1, publications) + { + bool match = false; + foreach_ptr(Publication, pub2, data->publications) + { + if (strcmp(pub1->name, pub2->name) == 0) + { + match = true; + break; + } + } + + if (!match) + { + publications_updated = true; + break; + } + } + + list_free(publications); + } + + /* + * Validate the entry only if the entry is not valid or in case a new + * publication has been added. + */ + if (!entry->replicate_valid || + (!publications_valid && publications_updated)) { Oid schemaId = get_rel_namespace(relid); List *pubids = GetRelationPublications(relid); @@ -2071,6 +2122,7 @@ get_rel_sync_entry(PGOutputData *data, Relation relation) bool am_partition = get_rel_relispartition(relid); char relkind = get_rel_relkind(relid); List *rel_publications = NIL; + bool skipped_pub = false; /* Reload publications if needed before use. */ if (!publications_valid) @@ -2078,9 +2130,15 @@ get_rel_sync_entry(PGOutputData *data, Relation relation) MemoryContextReset(data->pubctx); oldctx = MemoryContextSwitchTo(data->pubctx); - data->publications = LoadPublications(data->publication_names); + data->publications = LoadPublications(data->publication_names, &skipped_pub); MemoryContextSwitchTo(oldctx); - publications_valid = true; + + /* + * We don't consider the publications to be valid till we have + * information of all the publications. + */ + if (!skipped_pub) + publications_valid = true; } /* -- 2.43.0 [text/x-sh] test_pg_recvlogical.sh (1.3K, ../../CALDaNm1Lg1YOtY+Sy2his=ikHsQYYQ1pWN620QiBZDBdbsjMHw@mail.gmail.com/3-test_pg_recvlogical.sh) download | inline: #!/bin/bash ##### SLOT_NAME=test PLUGIN_NAME=pgoutput NUM_RECORDS=100000 LOOP=1 ##### for i in `seq 1 $LOOP` do # Cleanup previous result pg_ctl stop -D data rm -rf data logfile # Initialize an instance initdb -D data -U postgres -c wal_level=logical # Start the instance pg_ctl -D data -l logfile start # Create a table psql -U postgres -c "CREATE TABLE foo (id int);" psql -U postgres -c "CREATE publication pub1 for table foo;" psql -U postgres -c "CREATE publication pub2 for table foo;" #psql -U postgres -c "drop publication pub1;" # Create a replication slot psql -U postgres -c "SELECT * FROM pg_create_logical_replication_slot('$SLOT_NAME', '$PLUGIN_NAME')" # Insert tuples (this transaction will be decoded) psql -U postgres -c "INSERT INTO foo VALUES (generate_series(1, $NUM_RECORDS))" # Confirm current WAL location WAL_POS=$(psql -qtAX -U postgres -c "SELECT * FROM pg_current_wal_lsn()") t1=$(($(date +%s%N)/1000)) echo $t1 > run_${i}.dat # Run pg_recvlogical till the current WAL location time pg_recvlogical -d postgres -U postgres --start -S $SLOT_NAME -E $WAL_POS -f - -o publication_names='pub1,pub2' -o proto_version=4 ) &>> run_${i}.dat t2=$(($(date +%s%N)/1000)) echo $t2 >> run_${i}.dat t3=$((t2-t1)) echo "execution time=$t3" >> run_${i}.dat done ^ permalink raw reply [nested|flat] 28+ messages in thread
* Re: Add an option to skip loading missing publication to avoid logical replication failure @ 2025-03-03 11:11 Amit Kapila <[email protected]> parent: vignesh C <[email protected]> 0 siblings, 0 replies; 28+ messages in thread From: Amit Kapila @ 2025-03-03 11:11 UTC (permalink / raw) To: vignesh C <[email protected]>; +Cc: PostgreSQL Hackers <[email protected]> On Mon, Mar 3, 2025 at 2:30 PM vignesh C <[email protected]> wrote: > > On Tue, 25 Feb 2025 at 15:32, vignesh C <[email protected]> wrote: > > > > The attached script has the script that was used for testing. Here the > > NUM_RECORDS count should be changed accordingly for each of the tests > > and while running the test with the patch change uncomment the drop > > publication command. > > I have done further analysis on the test and changed the test to > compare it better with HEAD. The execution time is in milliseconds. > Brach/records | 100 | 1000 | 10000 | 100000 | 1000000 > Head | 10.43 | 15.86 | 64.44 | 550.56 | 8991.04 > Patch | 11.35 | 17.26 | 73.50 | 640.21 | 10104.72 > % diff | -8.82 | -8.85 | -14.08 | -16.28 | -12.38 > > There is a performance degradation in the range of 8.8 to 16.2 percent. > - /* Validate the entry */ - if (!entry->replicate_valid) + /* + * If the publication is invalid, check for updates. + * This optimization ensures that the next block, which queries the system + * tables and builds the relation entry, runs only if a new publication was + * created. + */ + if (!publications_valid && data->publications) + { + bool skipped_pub = false; + List *publications; + + publications = LoadPublications(data->publication_names, &skipped_pub); The publications_valid flag indicates whether the publications cache is valid or not; the flag is set to false for any invalidation in the pg_publication catalog. I wonder that instead of using the same flag what if we use a separate publications_skipped flag? If that works, you don't even need to change the current location where we LoadPublications. -- With Regards, Amit Kapila. ^ permalink raw reply [nested|flat] 28+ messages in thread
end of thread, other threads:[~2025-03-03 11:11 UTC | newest] Thread overview: 28+ messages (download: mbox mbox.gz follow: Atom feed) -- links below jump to the message on this page -- 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2021-03-04 03:53 [PATCH 5/5] WIP unify handling of attributes and expressions Tomas Vondra <[email protected]> 2025-02-18 11:23 Re: Add an option to skip loading missing publication to avoid logical replication failure Amit Kapila <[email protected]> 2025-02-25 10:02 ` Re: Add an option to skip loading missing publication to avoid logical replication failure vignesh C <[email protected]> 2025-03-03 09:00 ` Re: Add an option to skip loading missing publication to avoid logical replication failure vignesh C <[email protected]> 2025-03-03 11:11 ` Re: Add an option to skip loading missing publication to avoid logical replication failure Amit Kapila <[email protected]>
This inbox is served by agora; see mirroring instructions for how to clone and mirror all data and code used for this inbox