rdelaage
/
postgresql
mirror of https://git.postgresql.org/git/postgresql.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Refactor planning of projection steps that don't need a Result plan node. 

The original upper-planner-pathification design (commit 3fc6e2d7f5)
assumed that we could always determine during Path formation whether or not
we would need a Result plan node to perform projection of a targetlist.
That turns out not to work very well, though, because createplan.c still
has some responsibilities for choosing the specific target list associated
with sorting/grouping nodes (in particular it might choose to add resjunk
columns for sorting).  We might not ever refactor that --- doing so would
push more work into Path formation, which isn't attractive --- and we
certainly won't do so for 9.6.  So, while create_projection_path and
apply_projection_to_path can tell for sure what will happen if the subpath
is projection-capable, they can't tell for sure when it isn't.  This is at
least a latent bug in apply_projection_to_path, which might think it can
apply a target to a non-projecting node when the node will end up computing
something different.

Also, I'd tied the creation of a ProjectionPath node to whether or not a
Result is needed, but it turns out that we sometimes need a ProjectionPath
node anyway to avoid modifying a possibly-shared subpath node.  Callers had
to use create_projection_path for such cases, and we added code to them
that knew about the potential omission of a Result node and attempted to
adjust the cost estimates for that.  That was uncertainly correct and
definitely ugly/unmaintainable.

To fix, have create_projection_path explicitly check whether a Result
is needed and adjust its cost estimate accordingly, though it creates
a ProjectionPath in either case.  apply_projection_to_path is now mostly
just an optimized version that can avoid creating an extra Path node when
the input is known to not be shared with any other live path.  (There
is one case that create_projection_path doesn't handle, which is pushing
parallel-safe expressions below a Gather node.  We could make it do that
by duplicating the GatherPath, but there seems no need as yet.)

create_projection_plan still has to recheck the tlist-match condition,
which means that if the matching situation does get changed by createplan.c
then we'll have made a slightly incorrect cost estimate.  But there seems
no help for that in the near term, and I doubt it occurs often enough,
let alone would change planning decisions often enough, to be worth
stressing about.

I added a "dummypp" field to ProjectionPath to track whether
create_projection_path thinks a Result is needed.  This is not really
necessary as-committed because create_projection_plan doesn't look at the
flag; but it seems like a good idea to remember what we thought when
forming the cost estimate, if only for debugging purposes.

In passing, get rid of the target_parallel parameter added to
apply_projection_to_path by commit 54f5c5150.  I don't think that's a good
idea because it involves callers in what should be an internal decision,
and opens us up to missing optimization opportunities if callers think they
don't need to provide a valid flag, as most don't.  For the moment, this
just costs us an extra has_parallel_hazard call when planning a Gather.
If that starts to look expensive, I think a better solution would be to
teach PathTarget to carry/cache knowledge of parallel-safety of its
contents.
This commit is contained in:
Tom Lane 2016-06-21 18:38:20 -04:00
parent 936b62ddf2
commit 8b9d323cb9
8 changed files with 110 additions and 101 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
src/backend/nodes/outfuncs.c
View File

@ -1809,6 +1809,7 @@ _outProjectionPath(StringInfo str, const ProjectionPath *node)
_outPathInfo(str, (const Path *) node);
WRITE_NODE_FIELD(subpath);
WRITE_BOOL_FIELD(dummypp);
}
static void

35
src/backend/optimizer/plan/createplan.c
View File

@ -1409,8 +1409,9 @@ create_gather_plan(PlannerInfo *root, GatherPath *best_path)
/*
* create_projection_plan
*
* Create a Result node to do a projection step and (recursively) plans
* for its subpaths.
* Create a plan tree to do a projection step and (recursively) plans
* for its subpaths. We may need a Result node for the projection,
* but sometimes we can just let the subplan do the work.
*/
static Plan *
create_projection_plan(PlannerInfo *root, ProjectionPath *best_path)
@ -1425,31 +1426,37 @@ create_projection_plan(PlannerInfo *root, ProjectionPath *best_path)
tlist = build_path_tlist(root, &best_path->path);
/*
* We might not really need a Result node here. There are several ways
* that this can happen. For example, MergeAppend doesn't project, so we
* would have thought that we needed a projection to attach resjunk sort
* columns to its output ... but create_merge_append_plan might have added
* those same resjunk sort columns to both MergeAppend and its children.
* Alternatively, apply_projection_to_path might have created a projection
* path as the subpath of a Gather node even though the subpath was
* projection-capable. So, if the subpath is capable of projection or the
* desired tlist is the same expression-wise as the subplan's, just jam it
* in there. We'll have charged for a Result that doesn't actually appear
* in the plan, but that's better than having a Result we don't need.
* We might not really need a Result node here, either because the subplan
* can project or because it's returning the right list of expressions
* anyway. Usually create_projection_path will have detected that and set
* dummypp if we don't need a Result; but its decision can't be final,
* because some createplan.c routines change the tlists of their nodes.
* (An example is that create_merge_append_plan might add resjunk sort
* columns to a MergeAppend.) So we have to recheck here. If we do
* arrive at a different answer than create_projection_path did, we'll
* have made slightly wrong cost estimates; but label the plan with the
* cost estimates we actually used, not "corrected" ones. (XXX this could
* be cleaned up if we moved more of the sortcolumn setup logic into Path
* creation, but that would add expense to creating Paths we might end up
* not using.)
*/
if (is_projection_capable_path(best_path->subpath) ||
tlist_same_exprs(tlist, subplan->targetlist))
{
/* Don't need a separate Result, just assign tlist to subplan */
plan = subplan;
plan->targetlist = tlist;
/* Adjust cost to match what we thought during planning */
/* Label plan with the estimated costs we actually used */
plan->startup_cost = best_path->path.startup_cost;
plan->total_cost = best_path->path.total_cost;
plan->plan_rows = best_path->path.rows;
plan->plan_width = best_path->path.pathtarget->width;
/* ... but be careful not to munge subplan's parallel-aware flag */
}
else
{
/* We need a Result node */
plan = (Plan *) make_result(tlist, NULL, subplan);
copy_generic_path_info(plan, (Path *) best_path);

3
src/backend/optimizer/plan/planagg.c
View File

@ -465,8 +465,7 @@ build_minmax_path(PlannerInfo *root, MinMaxAggInfo *mminfo,
* cheapest path.)
*/
sorted_path = apply_projection_to_path(subroot, final_rel, sorted_path,
create_pathtarget(subroot, tlist),
false);
create_pathtarget(subroot, tlist));
/*
* Determine cost to get just the first row of the presorted path.

54
src/backend/optimizer/plan/planner.c
View File

@ -1500,7 +1500,6 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
PathTarget *grouping_target;
PathTarget *scanjoin_target;
bool have_grouping;
bool scanjoin_target_parallel_safe = false;
WindowFuncLists *wflists = NULL;
List *activeWindows = NIL;
List *rollup_lists = NIL;
@ -1730,14 +1729,6 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
else
scanjoin_target = grouping_target;
/*
* Check whether scan/join target is parallel safe ... unless there
* are no partial paths, in which case we don't care.
*/
if (current_rel->partial_pathlist &&
!has_parallel_hazard((Node *) scanjoin_target->exprs, false))
scanjoin_target_parallel_safe = true;
/*
* Forcibly apply scan/join target to all the Paths for the scan/join
* rel.
@ -1756,8 +1747,7 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
Assert(subpath->param_info == NULL);
path = apply_projection_to_path(root, current_rel,
subpath, scanjoin_target,
scanjoin_target_parallel_safe);
subpath, scanjoin_target);
/* If we had to add a Result, path is different from subpath */
if (path != subpath)
{
@ -1774,15 +1764,13 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
* partial pathlist will expect partial paths for that rel to produce
* the same output as complete paths ... and we just changed the
* output for the complete paths, so we'll need to do the same thing
* for partial paths.
* for partial paths. But only parallel-safe expressions can be
* computed by partial paths.
*/
if (scanjoin_target_parallel_safe)
if (current_rel->partial_pathlist &&
!has_parallel_hazard((Node *) scanjoin_target->exprs, false))
{
/*
* Apply the scan/join target to each partial path. Otherwise,
* anything that attempts to use the partial paths for further
* upper planning may go wrong.
*/
/* Apply the scan/join target to each partial path */
foreach(lc, current_rel->partial_pathlist)
{
Path *subpath = (Path *) lfirst(lc);
@ -1792,36 +1780,14 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
Assert(subpath->param_info == NULL);
/*
* We can't use apply_projection_to_path() here, because there
* could already be pointers to these paths, and therefore we
* dare not modify them in place. Instead, we must use
* create_projection_path() unconditionally.
* Don't use apply_projection_to_path() here, because there
* could be other pointers to these paths, and therefore we
* mustn't modify them in place.
*/
newpath = (Path *) create_projection_path(root,
current_rel,
subpath,
scanjoin_target);
/*
* Although create_projection_path() inserts a ProjectionPath
* unconditionally, create_projection_plan() will only insert
* a Result node if the subpath is not projection-capable, so
* we should discount the cost of that node if it will not
* actually get inserted. (This is pretty grotty but we can
* improve it later if it seems important.)
*/
if (equal(scanjoin_target->exprs, subpath->pathtarget->exprs))
{
/* at most we need a relabeling of the subpath */
newpath->startup_cost = subpath->startup_cost;
newpath->total_cost = subpath->total_cost;
}
else if (is_projection_capable_path(subpath))
{
/* need to project, but we don't need a Result */
newpath->total_cost -= cpu_tuple_cost * subpath->rows;
}
lfirst(lc) = newpath;
}
}
@ -4231,7 +4197,7 @@ create_ordered_paths(PlannerInfo *root,
/* Add projection step if needed */
if (path->pathtarget != target)
path = apply_projection_to_path(root, ordered_rel,
path, target, false);
path, target);
add_path(ordered_rel, path);
}

6
src/backend/optimizer/prep/prepunion.c
View File

@ -325,8 +325,7 @@ recurse_set_operations(Node *setOp, PlannerInfo *root,
refnames_tlist);
path = apply_projection_to_path(root, rel, path,
create_pathtarget(root, tlist),
false);
create_pathtarget(root, tlist));
/* Return the fully-fledged tlist to caller, too */
*pTargetList = tlist;
@ -395,8 +394,7 @@ recurse_set_operations(Node *setOp, PlannerInfo *root,
path->parent,
path,
create_pathtarget(root,
*pTargetList),
false);
*pTargetList));
}
return path;
}

94
src/backend/optimizer/util/pathnode.c
View File

@ -2168,6 +2168,7 @@ create_projection_path(PlannerInfo *root,
PathTarget *target)
{
ProjectionPath *pathnode = makeNode(ProjectionPath);
PathTarget *oldtarget = subpath->pathtarget;
pathnode->path.pathtype = T_Result;
pathnode->path.parent = rel;
@ -2184,13 +2185,46 @@ create_projection_path(PlannerInfo *root,
pathnode->subpath = subpath;
/*
* The Result node's cost is cpu_tuple_cost per row, plus the cost of
* evaluating the tlist. There is no qual to worry about.
* We might not need a separate Result node. If the input plan node type
* can project, we can just tell it to project something else. Or, if it
* can't project but the desired target has the same expression list as
* what the input will produce anyway, we can still give it the desired
* tlist (possibly changing its ressortgroupref labels, but nothing else).
* Note: in the latter case, create_projection_plan has to recheck our
* conclusion; see comments therein.
*/
pathnode->path.rows = subpath->rows;
pathnode->path.startup_cost = subpath->startup_cost + target->cost.startup;
pathnode->path.total_cost = subpath->total_cost + target->cost.startup +
(cpu_tuple_cost + target->cost.per_tuple) * subpath->rows;
if (is_projection_capable_path(subpath) ||
equal(oldtarget->exprs, target->exprs))
{
/* No separate Result node needed */
pathnode->dummypp = true;
/*
* Set cost of plan as subpath's cost, adjusted for tlist replacement.
*/
pathnode->path.rows = subpath->rows;
pathnode->path.startup_cost = subpath->startup_cost +
(target->cost.startup - oldtarget->cost.startup);
pathnode->path.total_cost = subpath->total_cost +
(target->cost.startup - oldtarget->cost.startup) +
(target->cost.per_tuple - oldtarget->cost.per_tuple) * subpath->rows;
}
else
{
/* We really do need the Result node */
pathnode->dummypp = false;
/*
* The Result node's cost is cpu_tuple_cost per row, plus the cost of
* evaluating the tlist. There is no qual to worry about.
*/
pathnode->path.rows = subpath->rows;
pathnode->path.startup_cost = subpath->startup_cost +
target->cost.startup;
pathnode->path.total_cost = subpath->total_cost +
target->cost.startup +
(cpu_tuple_cost + target->cost.per_tuple) * subpath->rows;
}
return pathnode;
}
@ -2199,38 +2233,37 @@ create_projection_path(PlannerInfo *root,
* apply_projection_to_path
* Add a projection step, or just apply the target directly to given path.
*
* Most plan types include ExecProject, so we can implement a new projection
* without an extra plan node: just replace the given path's pathtarget with
* the desired one. If the given path can't project, add a ProjectionPath.
* This has the same net effect as create_projection_path(), except that if
* a separate Result plan node isn't needed, we just replace the given path's
* pathtarget with the desired one. This must be used only when the caller
* knows that the given path isn't referenced elsewhere and so can be modified
* in-place.
*
* We can also short-circuit cases where the targetlist expressions are
* actually equal; this is not an uncommon case, since it may arise from
* trying to apply a PathTarget with sortgroupref labeling to a derived
* path without such labeling.
* If the input path is a GatherPath, we try to push the new target down to
* its input as well; this is a yet more invasive modification of the input
* path, which create_projection_path() can't do.
*
* This requires knowing that the source path won't be referenced for other
* purposes (e.g., other possible paths), since we modify it in-place. Note
* also that we mustn't change the source path's parent link; so when it is
* Note that we mustn't change the source path's parent link; so when it is
* add_path'd to "rel" things will be a bit inconsistent. So far that has
* not caused any trouble.
*
* 'rel' is the parent relation associated with the result
* 'path' is the path representing the source of data
* 'target' is the PathTarget to be computed
* 'target_parallel' indicates that target expressions are all parallel-safe
*/
Path *
apply_projection_to_path(PlannerInfo *root,
RelOptInfo *rel,
Path *path,
PathTarget *target,
bool target_parallel)
PathTarget *target)
{
QualCost oldcost;
/* Make a separate ProjectionPath if needed */
if (!is_projection_capable_path(path) &&
!equal(path->pathtarget->exprs, target->exprs))
/*
* If given path can't project, we might need a Result node, so make a
* separate ProjectionPath.
*/
if (!is_projection_capable_path(path))
return (Path *) create_projection_path(root, rel, path, target);
/*
@ -2247,10 +2280,11 @@ apply_projection_to_path(PlannerInfo *root,
/*
* If the path happens to be a Gather path, we'd like to arrange for the
* subpath to return the required target list so that workers can help
* project. But if there is something that is not parallel-safe in the
* project. But if there is something that is not parallel-safe in the
* target expressions, then we can't.
*/
if (IsA(path, GatherPath) &&target_parallel)
if (IsA(path, GatherPath) &&
!has_parallel_hazard((Node *) target->exprs, false))
{
GatherPath *gpath = (GatherPath *) path;
@ -2258,14 +2292,12 @@ apply_projection_to_path(PlannerInfo *root,
* We always use create_projection_path here, even if the subpath is
* projection-capable, so as to avoid modifying the subpath in place.
* It seems unlikely at present that there could be any other
* references to the subpath anyway, but better safe than sorry.
* (create_projection_plan will only insert a Result node if the
* subpath is not projection-capable, so we only include the cost of
* that node if it will actually be inserted. This is a bit grotty
* but we can improve it later if it seems important.)
* references to the subpath, but better safe than sorry.
*
* Note that we don't change the GatherPath's cost estimates; it might
* be appropriate to do so, to reflect the fact that the bulk of the
* target evaluation will happen in workers.
*/
if (!is_projection_capable_path(gpath->subpath))
gpath->path.total_cost += cpu_tuple_cost * gpath->subpath->rows;
gpath->subpath = (Path *)
create_projection_path(root,
gpath->subpath->parent,

15
src/include/nodes/relation.h
View File

@ -1274,15 +1274,22 @@ typedef struct HashPath
/*
* ProjectionPath represents a projection (that is, targetlist computation)
*
* This path node represents using a Result plan node to do a projection.
* It's only needed atop a node that doesn't support projection (such as
* Sort); otherwise we just jam the new desired PathTarget into the lower
* path node, and adjust that node's estimated cost accordingly.
* Nominally, this path node represents using a Result plan node to do a
* projection step. However, if the input plan node supports projection,
* we can just modify its output targetlist to do the required calculations
* directly, and not need a Result. In some places in the planner we can just
* jam the desired PathTarget into the input path node (and adjust its cost
* accordingly), so we don't need a ProjectionPath. But in other places
* it's necessary to not modify the input path node, so we need a separate
* ProjectionPath node, which is marked dummy to indicate that we intend to
* assign the work to the input plan node. The estimated cost for the
* ProjectionPath node will account for whether a Result will be used or not.
*/
typedef struct ProjectionPath
{
Path path;
Path *subpath; /* path representing input source */
bool dummypp; /* true if no separate Result is needed */
} ProjectionPath;
/*

3
src/include/optimizer/pathnode.h
View File

@ -143,8 +143,7 @@ extern ProjectionPath *create_projection_path(PlannerInfo *root,
extern Path *apply_projection_to_path(PlannerInfo *root,
RelOptInfo *rel,
Path *path,
PathTarget *target,
bool target_parallel);
PathTarget *target);
extern SortPath *create_sort_path(PlannerInfo *root,
RelOptInfo *rel,
Path *subpath,