git.postgresql.org Git - postgresql.git/commitdiff

git projects / postgresql.git / commitdiff
? search:
summary | shortlog | log | commit | commitdiff | tree
raw | patch | inline | side by side (parent: d59ff6c)
Save a few cycles in advance_transition_function().
Wed, 8 Jan 2014 18:58:15 +0000 (13:58 -0500)
Wed, 8 Jan 2014 18:58:37 +0000 (13:58 -0500)
Keep a pre-initialized FunctionCallInfoData in AggStatePerAggData, and
re-use that at each row instead of doing InitFunctionCallInfoData each
time. This saves only half a dozen assignments and maybe some stack
manipulation, and yet that seems to be good for a percent or two of the
overall query run time for simple aggregates such as count(*). The cost
is that the FunctionCallInfoData (which is about a kilobyte, on 64-bit
machines) stays allocated for the duration of the query instead of being
short-lived stack data. But we're already paying an equivalent space cost
for each regular FuncExpr or OpExpr node, so I don't feel bad about paying
it for aggregate functions. The code seems a little cleaner this way too,
since the number of things passed to advance_transition_function decreases.


diff --git a/src/backend/executor/nodeAgg.c b/src/backend/executor/nodeAgg.c
index 827b009c90cd3fa3feadd7730a42bfbc7cb234fb..7e4bca5b4d85af3e557697a1c8391f71e0ba28f8 100644 (file)
--- a/src/backend/executor/nodeAgg.c
+++ b/src/backend/executor/nodeAgg.c
@@ -235,6 +235,14 @@ typedef struct AggStatePerAggData
*/
Tuplesortstate *sortstate; /* sort object, if DISTINCT or ORDER BY */
+
+ /*
+ * This field is a pre-initialized FunctionCallInfo struct used for
+ * calling this aggregate's transfn. We save a few cycles per row by not
+ * re-initializing the unchanging fields; which isn't much, but it seems
+ * worth the extra space consumption.
+ */
+ FunctionCallInfoData transfn_fcinfo;
} AggStatePerAggData;
/*
@@ -290,8 +298,7 @@ static void initialize_aggregates(AggState *aggstate,
AggStatePerGroup pergroup);
static void advance_transition_function(AggState *aggstate,
AggStatePerAgg peraggstate,
- AggStatePerGroup pergroupstate,
- FunctionCallInfoData *fcinfo);
+ AggStatePerGroup pergroupstate);
static void advance_aggregates(AggState *aggstate, AggStatePerGroup pergroup);
static void process_ordered_aggregate_single(AggState *aggstate,
AggStatePerAgg peraggstate,
@@ -399,21 +406,20 @@ initialize_aggregates(AggState *aggstate,
* Given new input value(s), advance the transition function of an aggregate.
*
* The new values (and null flags) have been preloaded into argument positions
- * 1 and up in fcinfo, so that we needn't copy them again to pass to the
- * transition function. No other fields of fcinfo are assumed valid.
+ * 1 and up in peraggstate->transfn_fcinfo, so that we needn't copy them again
+ * to pass to the transition function. We also expect that the static fields
+ * of the fcinfo are already initialized; that was done by ExecInitAgg().
*
* It doesn't matter which memory context this is called in.
*/
static void
advance_transition_function(AggState *aggstate,
AggStatePerAgg peraggstate,
- AggStatePerGroup pergroupstate,
- FunctionCallInfoData *fcinfo)
+ AggStatePerGroup pergroupstate)
{
- int numTransInputs = peraggstate->numTransInputs;
+ FunctionCallInfo fcinfo = &peraggstate->transfn_fcinfo;
MemoryContext oldContext;
Datum newVal;
- int i;
if (peraggstate->transfn.fn_strict)
{
@@ -421,6 +427,9 @@ advance_transition_function(AggState *aggstate,
* For a strict transfn, nothing happens when there's a NULL input; we
* just keep the prior transValue.
*/
+ int numTransInputs = peraggstate->numTransInputs;
+ int i;
+
for (i = 1; i <= numTransInputs; i++)
{
if (fcinfo->argnull[i])
@@ -467,12 +476,9 @@ advance_transition_function(AggState *aggstate,
/*
* OK to call the transition function
*/
- InitFunctionCallInfoData(*fcinfo, &(peraggstate->transfn),
- numTransInputs + 1,
- peraggstate->aggCollation,
- (void *) aggstate, NULL);
fcinfo->arg[0] = pergroupstate->transValue;
fcinfo->argnull[0] = pergroupstate->transValueIsNull;
+ fcinfo->isnull = false; /* just in case transfn doesn't set it */
newVal = FunctionCallInvoke(fcinfo);
@@ -574,19 +580,18 @@ advance_aggregates(AggState *aggstate, AggStatePerGroup pergroup)
else
{
/* We can apply the transition function immediately */
- FunctionCallInfoData fcinfo;
+ FunctionCallInfo fcinfo = &peraggstate->transfn_fcinfo;
/* Load values into fcinfo */
/* Start from 1, since the 0th arg will be the transition value */
Assert(slot->tts_nvalid >= numTransInputs);
for (i = 0; i < numTransInputs; i++)
{
- fcinfo.arg[i + 1] = slot->tts_values[i];
- fcinfo.argnull[i + 1] = slot->tts_isnull[i];
+ fcinfo->arg[i + 1] = slot->tts_values[i];
+ fcinfo->argnull[i + 1] = slot->tts_isnull[i];
}
- advance_transition_function(aggstate, peraggstate, pergroupstate,
- &fcinfo);
+ advance_transition_function(aggstate, peraggstate, pergroupstate);
}
}
}
@@ -622,17 +627,17 @@ process_ordered_aggregate_single(AggState *aggstate,
MemoryContext workcontext = aggstate->tmpcontext->ecxt_per_tuple_memory;
MemoryContext oldContext;
bool isDistinct = (peraggstate->numDistinctCols > 0);
+ FunctionCallInfo fcinfo = &peraggstate->transfn_fcinfo;
Datum *newVal;
bool *isNull;
- FunctionCallInfoData fcinfo;
Assert(peraggstate->numDistinctCols < 2);
tuplesort_performsort(peraggstate->sortstate);
/* Load the column into argument 1 (arg 0 will be transition value) */
- newVal = fcinfo.arg + 1;
- isNull = fcinfo.argnull + 1;
+ newVal = fcinfo->arg + 1;
+ isNull = fcinfo->argnull + 1;
/*
* Note: if input type is pass-by-ref, the datums returned by the sort are
@@ -668,8 +673,7 @@ process_ordered_aggregate_single(AggState *aggstate,
}
else
{
- advance_transition_function(aggstate, peraggstate, pergroupstate,
- &fcinfo);
+ advance_transition_function(aggstate, peraggstate, pergroupstate);
/* forget the old value, if any */
if (!oldIsNull && !peraggstate->inputtypeByVal)
pfree(DatumGetPointer(oldVal));
@@ -704,7 +708,7 @@ process_ordered_aggregate_multi(AggState *aggstate,
AggStatePerGroup pergroupstate)
{
MemoryContext workcontext = aggstate->tmpcontext->ecxt_per_tuple_memory;
- FunctionCallInfoData fcinfo;
+ FunctionCallInfo fcinfo = &peraggstate->transfn_fcinfo;
TupleTableSlot *slot1 = peraggstate->evalslot;
TupleTableSlot *slot2 = peraggstate->uniqslot;
int numTransInputs = peraggstate->numTransInputs;
@@ -739,12 +743,11 @@ process_ordered_aggregate_multi(AggState *aggstate,
/* Start from 1, since the 0th arg will be the transition value */
for (i = 0; i < numTransInputs; i++)
{
- fcinfo.arg[i + 1] = slot1->tts_values[i];
- fcinfo.argnull[i + 1] = slot1->tts_isnull[i];
+ fcinfo->arg[i + 1] = slot1->tts_values[i];
+ fcinfo->argnull[i + 1] = slot1->tts_isnull[i];
}
- advance_transition_function(aggstate, peraggstate, pergroupstate,
- &fcinfo);
+ advance_transition_function(aggstate, peraggstate, pergroupstate);
if (numDistinctCols > 0)
{
@@ -1799,6 +1802,7 @@ ExecInitAgg(Agg *node, EState *estate, int eflags)
&transfnexpr,
&finalfnexpr);
+ /* set up infrastructure for calling the transfn and finalfn */
fmgr_info(transfn_oid, &peraggstate->transfn);
fmgr_info_set_expr((Node *) transfnexpr, &peraggstate->transfn);
@@ -1810,6 +1814,13 @@ ExecInitAgg(Agg *node, EState *estate, int eflags)
peraggstate->aggCollation = aggref->inputcollid;
+ InitFunctionCallInfoData(peraggstate->transfn_fcinfo,
+ &peraggstate->transfn,
+ peraggstate->numTransInputs + 1,
+ peraggstate->aggCollation,
+ (void *) aggstate, NULL);
+
+ /* get info about relevant datatypes */
get_typlenbyval(aggref->aggtype,
&peraggstate->resulttypeLen,
&peraggstate->resulttypeByVal);
This is the main PostgreSQL git repository.
RSS Atom

AltStyle によって変換されたページ (->オリジナル) /