gset_data (aka gd) in planner.c is always non-null if and only if
parse->groupingSets is non-null, but Coverity doesn't know that and
complains. Feed it an assertion to see if that keeps it happy.
copyObject() is declared to return void *, which allows easily assigning
the result independent of the input, but it loses all type checking.
If the compiler supports typeof or something similar, cast the result to
the input type. This creates a greater amount of type safety. In some
cases, where the result is assigned to a generic type such as Node * or
Expr *, new casts are now necessary, but in general casts are now
unnecessary in the normal case and indicate that something unusual is
happening.
Reviewed-by: Mark Dilger <hornschnorter@gmail.com>
This extends the Aggregate node with two new features: HashAggregate
can now run multiple hashtables concurrently, and a new strategy
MixedAggregate populates hashtables while doing sorted grouping.
The planner will now attempt to save as many sorts as possible when
planning grouping sets queries, while not exceeding work_mem for the
estimated combined sizes of all hashtables used. No SQL-level changes
are required. There should be no user-visible impact other than the
new EXPLAIN output and possible changes to result ordering when ORDER
BY was not used (which affected a few regression tests). The
enable_hashagg option is respected.
Author: Andrew Gierth
Reviewers: Mark Dilger, Andres Freund
Discussion: https://postgr.es/m/87vatszyhj.fsf@news-spur.riddles.org.uk
This replaces the old, recursive tree-walk based evaluation, with
non-recursive, opcode dispatch based, expression evaluation.
Projection is now implemented as part of expression evaluation.
This both leads to significant performance improvements, and makes
future just-in-time compilation of expressions easier.
The speed gains primarily come from:
- non-recursive implementation reduces stack usage / overhead
- simple sub-expressions are implemented with a single jump, without
function calls
- sharing some state between different sub-expressions
- reduced amount of indirect/hard to predict memory accesses by laying
out operation metadata sequentially; including the avoidance of
nearly all of the previously used linked lists
- more code has been moved to expression initialization, avoiding
constant re-checks at evaluation time
Future just-in-time compilation (JIT) has become easier, as
demonstrated by released patches intended to be merged in a later
release, for primarily two reasons: Firstly, due to a stricter split
between expression initialization and evaluation, less code has to be
handled by the JIT. Secondly, due to the non-recursive nature of the
generated "instructions", less performance-critical code-paths can
easily be shared between interpreted and compiled evaluation.
The new framework allows for significant future optimizations. E.g.:
- basic infrastructure for to later reduce the per executor-startup
overhead of expression evaluation, by caching state in prepared
statements. That'd be helpful in OLTPish scenarios where
initialization overhead is measurable.
- optimizing the generated "code". A number of proposals for potential
work has already been made.
- optimizing the interpreter. Similarly a number of proposals have
been made here too.
The move of logic into the expression initialization step leads to some
backward-incompatible changes:
- Function permission checks are now done during expression
initialization, whereas previously they were done during
execution. In edge cases this can lead to errors being raised that
previously wouldn't have been, e.g. a NULL array being coerced to a
different array type previously didn't perform checks.
- The set of domain constraints to be checked, is now evaluated once
during expression initialization, previously it was re-built
every time a domain check was evaluated. For normal queries this
doesn't change much, but e.g. for plpgsql functions, which caches
ExprStates, the old set could stick around longer. The behavior
around might still change.
Author: Andres Freund, with significant changes by Tom Lane,
changes by Heikki Linnakangas
Reviewed-By: Tom Lane, Heikki Linnakangas
Discussion: https://postgr.es/m/20161206034955.bh33paeralxbtluv@alap3.anarazel.de
Partitioned tables do not contain any data; only their unpartitioned
descendents need to be scanned. However, the partitioned tables still
need to be locked, even though they're not scanned. To make that
work, Append and MergeAppend relations now need to carry a list of
(unscanned) partitioned relations that must be locked, and InitPlan
must lock all partitioned result relations.
Aside from the obvious advantage of avoiding some work at execution
time, this has two other advantages. First, it may improve the
planner's decision-making in some cases since the empty relation
might throw things off. Second, it paves the way to getting rid of
the storage for partitioned tables altogether.
Amit Langote, reviewed by me.
Discussion: http://postgr.es/m/6837c359-45c4-8044-34d1-736756335a15@lab.ntt.co.jp
Neha Sharma reported these to Rushabh Lathia just after I commit
355d3993c53ed62c5b53d020648e4fbcfbf5f155 went in. The patch is
Rushabh's, with input from me.
Like Gather, we spawn multiple workers and run the same plan in each
one; however, Gather Merge is used when each worker produces the same
output ordering and we want to preserve that output ordering while
merging together the streams of tuples from various workers. (In a
way, Gather Merge is like a hybrid of Gather and MergeAppend.)
This works out to a win if it saves us from having to perform an
expensive Sort. In cases where only a small amount of data would need
to be sorted, it may actually be faster to use a regular Gather node
and then sort the results afterward, because Gather Merge sometimes
needs to wait synchronously for tuples whereas a pure Gather generally
doesn't. But if this avoids an expensive sort then it's a win.
Rushabh Lathia, reviewed and tested by Amit Kapila, Thomas Munro,
and Neha Sharma, and reviewed and revised by me.
Discussion: http://postgr.es/m/CAGPqQf09oPX-cQRpBKS0Gq49Z+m6KBxgxd_p9gX8CKk_d75HoQ@mail.gmail.com
XMLTABLE is defined by the SQL/XML standard as a feature that allows
turning XML-formatted data into relational form, so that it can be used
as a <table primary> in the FROM clause of a query.
This new construct provides significant simplicity and performance
benefit for XML data processing; what in a client-side custom
implementation was reported to take 20 minutes can be executed in 400ms
using XMLTABLE. (The same functionality was said to take 10 seconds
using nested PostgreSQL XPath function calls, and 5 seconds using
XMLReader under PL/Python).
The implemented syntax deviates slightly from what the standard
requires. First, the standard indicates that the PASSING clause is
optional and that multiple XML input documents may be given to it; we
make it mandatory and accept a single document only. Second, we don't
currently support a default namespace to be specified.
This implementation relies on a new executor node based on a hardcoded
method table. (Because the grammar is fixed, there is no extensibility
in the current approach; further constructs can be implemented on top of
this such as JSON_TABLE, but they require changes to core code.)
Author: Pavel Stehule, Álvaro Herrera
Extensively reviewed by: Craig Ringer
Discussion: https://postgr.es/m/CAFj8pRAgfzMD-LoSmnMGybD0WsEznLHWap8DO79+-GTRAPR4qA@mail.gmail.com
In combination with 569174f1be92be93f5366212cc46960d28a5c5cd, which
taught the btree AM how to perform parallel index scans, this allows
parallel index scan plans on btree indexes. This infrastructure
should be general enough to support parallel index scans for other
index AMs as well, if someone updates them to support parallel
scans.
Amit Kapila, reviewed and tested by Anastasia Lubennikova, Tushar
Ahuja, and Haribabu Kommi, and me.
I noticed while hacking on join UNION transforms that planner.c's
function get_base_rel_indexes() just duplicates the functionality of
get_relids_in_jointree(). It doesn't even have the excuse of being
older code :-(. Drop it and use the latter function instead.
If we forcibly place a Material node atop a finished subplan, we need
to move any initPlans attached to the subplan up to the Material node,
in order to keep SS_finalize_plan() happy. I'd figured this out in
commit 7b67a0a49 for the case of materializing a cursor plan, but out of
an abundance of caution, I put the initPlan movement hack at the call
site for that case, rather than inside materialize_finished_plan().
That was the wrong thing, because it turns out to also be necessary for
the only other caller of materialize_finished_plan(), ie subselect.c.
We lacked any test cases that exposed the mistake, but bug#14524 from
Wei Congrui shows that it's possible to get an initPlan reference into
the top tlist in that case too, and then SS_finalize_plan() complains.
Hence, move the hack into materialize_finished_plan().
In HEAD, also relocate some recently-added tests in subselect.sql, which
I'd unthinkingly dropped into the middle of a sequence of related tests.
Report: https://postgr.es/m/20170202060020.1400.89021@wrigleys.postgresql.org
Evaluation of set returning functions (SRFs_ in the targetlist (like SELECT
generate_series(1,5)) so far was done in the expression evaluation (i.e.
ExecEvalExpr()) and projection (i.e. ExecProject/ExecTargetList) code.
This meant that most executor nodes performing projection, and most
expression evaluation functions, had to deal with the possibility that an
evaluated expression could return a set of return values.
That's bad because it leads to repeated code in a lot of places. It also,
and that's my (Andres's) motivation, made it a lot harder to implement a
more efficient way of doing expression evaluation.
To fix this, introduce a new executor node (ProjectSet) that can evaluate
targetlists containing one or more SRFs. To avoid the complexity of the old
way of handling nested expressions returning sets (e.g. having to pass up
ExprDoneCond, and dealing with arguments to functions returning sets etc.),
those SRFs can only be at the top level of the node's targetlist. The
planner makes sure (via split_pathtarget_at_srfs()) that SRF evaluation is
only necessary in ProjectSet nodes and that SRFs are only present at the
top level of the node's targetlist. If there are nested SRFs the planner
creates multiple stacked ProjectSet nodes. The ProjectSet nodes always get
input from an underlying node.
We also discussed and prototyped evaluating targetlist SRFs using ROWS
FROM(), but that turned out to be more complicated than we'd hoped.
While moving SRF evaluation to ProjectSet would allow to retain the old
"least common multiple" behavior when multiple SRFs are present in one
targetlist (i.e. continue returning rows until all SRFs are at the end of
their input at the same time), we decided to instead only return rows till
all SRFs are exhausted, returning NULL for already exhausted ones. We
deemed the previous behavior to be too confusing, unexpected and actually
not particularly useful.
As a side effect, the previously prohibited case of multiple set returning
arguments to a function, is now allowed. Not because it's particularly
desirable, but because it ends up working and there seems to be no argument
for adding code to prohibit it.
Currently the behavior for COALESCE and CASE containing SRFs has changed,
returning multiple rows from the expression, even when the SRF containing
"arm" of the expression is not evaluated. That's because the SRFs are
evaluated in a separate ProjectSet node. As that's quite confusing, we're
likely to instead prohibit SRFs in those places. But that's still being
discussed, and the code would reside in places not touched here, so that's
a task for later.
There's a lot of, now superfluous, code dealing with set return expressions
around. But as the changes to get rid of those are verbose largely boring,
it seems better for readability to keep the cleanup as a separate commit.
Author: Tom Lane and Andres Freund
Discussion: https://postgr.es/m/20160822214023.aaxz5l4igypowyri@alap3.anarazel.de
In an RLS query, we must ensure that security filter quals are evaluated
before ordinary query quals, in case the latter contain "leaky" functions
that could expose the contents of sensitive rows. The original
implementation of RLS planning ensured this by pushing the scan of a
secured table into a sub-query that it marked as a security-barrier view.
Unfortunately this results in very inefficient plans in many cases, because
the sub-query cannot be flattened and gets planned independently of the
rest of the query.
To fix, drop the use of sub-queries to enforce RLS qual order, and instead
mark each qual (RestrictInfo) with a security_level field establishing its
priority for evaluation. Quals must be evaluated in security_level order,
except that "leakproof" quals can be allowed to go ahead of quals of lower
security_level, if it's helpful to do so. This has to be enforced within
the ordering of any one list of quals to be evaluated at a table scan node,
and we also have to ensure that quals are not chosen for early evaluation
(i.e., use as an index qual or TID scan qual) if they're not allowed to go
ahead of other quals at the scan node.
This is sufficient to fix the problem for RLS quals, since we only support
RLS policies on simple tables and thus RLS quals will always exist at the
table scan level only. Eventually these qual ordering rules should be
enforced for join quals as well, which would permit improving planning for
explicit security-barrier views; but that's a task for another patch.
Note that FDWs would need to be aware of these rules --- and not, for
example, send an insecure qual for remote execution --- but since we do
not yet allow RLS policies on foreign tables, the case doesn't arise.
This will need to be addressed before we can allow such policies.
Patch by me, reviewed by Stephen Frost and Dean Rasheed.
Discussion: https://postgr.es/m/8185.1477432701@sss.pgh.pa.us
This patch makes several changes that improve the consistency of
representation of lists of statements. It's always been the case
that the output of parse analysis is a list of Query nodes, whatever
the types of the individual statements in the list. This patch brings
similar consistency to the outputs of raw parsing and planning steps:
* The output of raw parsing is now always a list of RawStmt nodes;
the statement-type-dependent nodes are one level down from that.
* The output of pg_plan_queries() is now always a list of PlannedStmt
nodes, even for utility statements. In the case of a utility statement,
"planning" just consists of wrapping a CMD_UTILITY PlannedStmt around
the utility node. This list representation is now used in Portal and
CachedPlan plan lists, replacing the former convention of intermixing
PlannedStmts with bare utility-statement nodes.
Now, every list of statements has a consistent head-node type depending
on how far along it is in processing. This allows changing many places
that formerly used generic "Node *" pointers to use a more specific
pointer type, thus reducing the number of IsA() tests and casts needed,
as well as improving code clarity.
Also, the post-parse-analysis representation of DECLARE CURSOR is changed
so that it looks more like EXPLAIN, PREPARE, etc. That is, the contained
SELECT remains a child of the DeclareCursorStmt rather than getting flipped
around to be the other way. It's now true for both Query and PlannedStmt
that utilityStmt is non-null if and only if commandType is CMD_UTILITY.
That allows simplifying a lot of places that were testing both fields.
(I think some of those were just defensive programming, but in many places,
it was actually necessary to avoid confusing DECLARE CURSOR with SELECT.)
Because PlannedStmt carries a canSetTag field, we're also able to get rid
of some ad-hoc rules about how to reconstruct canSetTag for a bare utility
statement; specifically, the assumption that a utility is canSetTag if and
only if it's the only one in its list. While I see no near-term need for
relaxing that restriction, it's nice to get rid of the ad-hocery.
The API of ProcessUtility() is changed so that what it's passed is the
wrapper PlannedStmt not just the bare utility statement. This will affect
all users of ProcessUtility_hook, but the changes are pretty trivial; see
the affected contrib modules for examples of the minimum change needed.
(Most compilers should give pointer-type-mismatch warnings for uncorrected
code.)
There's also a change in the API of ExplainOneQuery_hook, to pass through
cursorOptions instead of expecting hook functions to know what to pick.
This is needed because of the DECLARE CURSOR changes, but really should
have been done in 9.6; it's unlikely that any extant hook functions
know about using CURSOR_OPT_PARALLEL_OK.
Finally, teach gram.y to save statement boundary locations in RawStmt
nodes, and pass those through to Query and PlannedStmt nodes. This allows
more intelligent handling of cases where a source query string contains
multiple statements. This patch doesn't actually do anything with the
information, but a follow-on patch will. (Passing this information through
cleanly is the true motivation for these changes; while I think this is all
good cleanup, it's unlikely we'd have bothered without this end goal.)
catversion bump because addition of location fields to struct Query
affects stored rules.
This patch is by me, but it owes a good deal to Fabien Coelho who did
a lot of preliminary work on the problem, and also reviewed the patch.
Discussion: https://postgr.es/m/alpine.DEB.2.20.1612200926310.29821@lancre
Andreas Seltenreich found another case where we were being too optimistic
about allowing a plan to be considered parallelizable despite it containing
initPlans. It seems like the real issue here is that if we know we are
going to tack initPlans onto the topmost Plan node for a subquery, we
had better mark that subquery's result Paths as not-parallel-safe. That
fixes this problem and allows reversion of a kluge (added in commit
7b67a0a49 and extended in f24cf960d) to not trust the parallel_safe flag
at top level.
Discussion: <874m2w4k5d.fsf@ex.ansel.ydns.eu>
We mustn't force parallel mode if the query has any subplans, since
ExecSerializePlan doesn't transmit them to workers. Testing
top_plan->initPlan is inadequate because (1) there might be initPlans
attached to lower plan nodes, and (2) non-initPlan subplans don't
work either. There's certainly room for improvement in those
restrictions, but for the moment that's what we've got.
Amit Kapila, per report from Andreas Seltenreich
Discussion: <8737im6pmh.fsf@credativ.de>
The plan generated for sorted partial aggregation with "GROUP BY constant"
included a Sort node with no sort keys, which the executor does not like.
Per report from Steve Randall. I'd add a regression test case if I could
think of a compact one, but it doesn't seem worth expending lots of cycles
on.
Report: <CABVd52UAdGXpg_rCk46egpNKYdXOzCjuJ1zG26E2xBe_8bj+Fg@mail.gmail.com>
The more efficient hashtable speeds up hash-aggregations with more than
a few hundred groups significantly. Improvements of over 120% have been
measured.
Due to the the different hash table queries that not fully
determined (e.g. GROUP BY without ORDER BY) may change their result
order.
The conversion is largely straight-forward, except that, due to the
static element types of simplehash.h type hashes, the additional data
some users store in elements (e.g. the per-group working data for hash
aggregaters) is now stored in TupleHashEntryData->additional. The
meaning of BuildTupleHashTable's entrysize (renamed to additionalsize)
has been changed to only be about the additionally stored size. That
size is only used for the initial sizing of the hash-table.
Reviewed-By: Tomas Vondra
Discussion: <20160727004333.r3e2k2y6fvk2ntup@alap3.anarazel.de>
Teach the parser to reject misplaced set-returning functions during parse
analysis using p_expr_kind, in much the same way as we do for aggregates
and window functions (cf commit eaccfded9). While this isn't complete
(it misses nesting-based restrictions), it's much better than the previous
error reporting for such cases, and it allows elimination of assorted
ad-hoc expression_returns_set() error checks. We could add nesting checks
later if it seems important to catch all cases at parse time.
There is one case the parser will now throw error for although previous
versions allowed it, which is SRFs in the tlist of an UPDATE. That never
behaved sensibly (since it's ill-defined which generated row should be
used to perform the update) and it's hard to see why it should not be
treated as an error. It's a release-note-worthy change though.
Also, add a new Query field hasTargetSRFs reporting whether there are
any SRFs in the targetlist (including GROUP BY/ORDER BY expressions).
The parser can now set that basically for free during parse analysis,
and we can use it in a number of places to avoid expression_returns_set
searches. (There will be more such checks soon.) In some places, this
allows decontorting the logic since it's no longer expensive to check for
SRFs in the tlist --- so I made the checks parallel to the handling of
hasAggs/hasWindowFuncs wherever it seemed appropriate.
catversion bump because adding a Query field changes stored rules.
Andres Freund and Tom Lane
Discussion: <24639.1473782855@sss.pgh.pa.us>
We need to scan the whole parse tree for parallel-unsafe functions.
If there are none, we'll later need to determine whether particular
subtrees contain any parallel-restricted functions. The previous coding
retained no knowledge from the first scan, even though this is very
wasteful in the common case where the query contains only parallel-safe
functions. We can bypass all of the later scans by remembering that fact.
This provides a small but measurable speed improvement when the case
applies, and shouldn't cost anything when it doesn't.
Patch by me, reviewed by Robert Haas
Discussion: <3740.1471538387@sss.pgh.pa.us>
We must not push down a foreign join when the foreign tables involved
should be accessed under different user mappings. Previously we tried
to enforce that rule literally during planning, but that meant that the
resulting plans were dependent on the current contents of the
pg_user_mapping catalog, and we had to blow away all cached plans
containing any remote join when anything at all changed in pg_user_mapping.
This could have been improved somewhat, but the fact that a syscache inval
callback has very limited info about what changed made it hard to do better
within that design. Instead, let's change the planner to not consider user
mappings per se, but to allow a foreign join if both RTEs have the same
checkAsUser value. If they do, then they necessarily will use the same
user mapping at runtime, and we don't need to know specifically which one
that is. Post-plan-time changes in pg_user_mapping no longer require any
plan invalidation.
This rule does give up some optimization ability, to wit where two foreign
table references come from views with different owners or one's from a view
and one's directly in the query, but nonetheless the same user mapping
would have applied. We'll sacrifice the first case, but to not regress
more than we have to in the second case, allow a foreign join involving
both zero and nonzero checkAsUser values if the nonzero one is the same as
the prevailing effective userID. In that case, mark the plan as only
runnable by that userID.
The plancache code already had a notion of plans being userID-specific,
in order to support RLS. It was a little confused though, in particular
lacking clarity of thought as to whether it was the rewritten query or just
the finished plan that's dependent on the userID. Rearrange that code so
that it's clearer what depends on which, and so that the same logic applies
to both RLS-injected role dependency and foreign-join-injected role
dependency.
Note that this patch doesn't remove the other issue mentioned in the
original complaint, which is that while we'll reliably stop using a foreign
join if it's disallowed in a new context, we might fail to start using a
foreign join if it's now allowed, but we previously created a generic
cached plan that didn't use one. It was agreed that the chance of winning
that way was not high enough to justify the much larger number of plan
invalidations that would have to occur if we tried to cause it to happen.
In passing, clean up randomly-varying spelling of EXPLAIN commands in
postgres_fdw.sql, and fix a COSTS ON example that had been allowed to
leak into the committed tests.
This reverts most of commits fbe5a3fb7 and 5d4171d1c, which were the
previous attempt at ensuring we wouldn't push down foreign joins that
span permissions contexts.
Etsuro Fujita and Tom Lane
Discussion: <d49c1e5b-f059-20f4-c132-e9752ee0113e@lab.ntt.co.jp>
Test the external-sort code path in CLUSTER for two different scenarios:
multiple-pass external sorting, and the best case for replacement
selection, where only one run is produced, so that no merge is required.
This test would have caught the bug fixed in commit 1b0fc8507, at
least when run with valgrind enabled.
In passing, add a short-circuit test in plan_cluster_use_sort() to make
dead certain that it selects sorting when enable_indexscan is off. As
things stand, that would happen anyway, but it seems like good future
proofing for this test.
Peter Geoghegan
Discussion: <CAM3SWZSgxehDkDMq1FdiW2A0Dxc79wH0hz1x-TnGy=1BXEL+nw@mail.gmail.com>
We were just leaving the cost fields zeroes, which produces obviously bogus
output with force_parallel_mode = on. With force_parallel_mode = regress,
the zeroes are hidden, but I wonder if they wouldn't still confuse add-on
code such as auto_explain.
In commit 915b703e1 I gave get_agg_clause_costs() the responsibility of
marking Aggref nodes with the appropriate aggtranstype. I failed to notice
that where it was being called from, it might see only a subset of the
Aggref nodes that were in the original targetlist. Specifically, if there
are duplicate aggregate calls in the tlist, either make_sort_input_target
or make_window_input_target might put just a single instance into the
grouping_target, and then only that instance would get marked. Fix by
moving the call back into grouping_planner(), before we start building
assorted PathTargets from the query tlist. Per report from Stefan Huehner.
Report: <20160702131056.GD3165@huehner.biz>
In commit 68fa28f77 I tried to teach SS_finalize_plan() to cope with
initPlans attached anywhere in the plan tree, by dint of moving its
handling of those into the recursion in finalize_plan(). It turns out that
that doesn't really work: if a lower-level plan node emits an initPlan
output parameter in its targetlist, it's legitimate for upper levels to
reference those Params --- and at the point where this code runs, those
references look just like the Param itself, so finalize_plan() quite
properly rejects them as being in the wrong place. We could lobotomize
the checks enough to allow that, probably, but then it's not clear that
we'd have any meaningful check for misplaced Params at all. What seems
better, at least in the near term, is to tweak standard_planner() a bit
so that initPlans are never placed anywhere but the topmost plan node
for a query level, restoring the behavior that occurred pre-9.6. Possibly
we can do better if this code is ever merged into setrefs.c: then it would
be possible to check a Param's placement only when we'd failed to replace
it with a Var referencing a child plan node's targetlist.
BTW, I'm now suspicious that finalize_plan is doing the wrong thing by
returning the node's allParam rather than extParam to be incorporated
in the parent node's set of used parameters. However, it makes no
difference given that initPlans only appear at top level, so I'll leave
that alone for now.
Another thing that emerged from this is that standard_planner() needs
to check for initPlans before deciding that it's safe to stick a Gather
node on top in force_parallel_mode mode. We previously guarded against
that by deciding the plan wasn't wholePlanParallelSafe if any subplans
had been found, but after commit 5ce5e4a12 it's necessary to have this
substitute test, because path parallel_safe markings don't account for
initPlans. (Normally, we'd have decided the paths weren't safe anyway
due to appearances of SubPlan nodes, Params, or CTE scans somewhere in
the tree --- but it's possible for those all to be optimized away while
initPlans still remain.)
Per fuzz testing by Andreas Seltenreich.
Report: <874m89rw7x.fsf@credativ.de>
In the previous design, the GetForeignUpperPaths FDW callback hook was
called before we got around to labeling upper relations with the proper
consider_parallel flag; this meant that any upper paths created by an FDW
would be marked not-parallel-safe. While that's probably just as well
right now, we aren't going to want it to be true forever. Hence, abandon
the idea that FDWs should be allowed to inject upper paths before the core
code has gotten around to creating the relevant upper relation. (Well,
actually they still can, but it's on their own heads how well it works.)
Instead, adopt the same API already designed for create_upper_paths_hook:
we call GetForeignUpperPaths after each upperrel has been created and
populated with the paths the core planner knows how to make.
Commit 3fc6e2d7f5b652b417fa6937c34de2438d60fa9f introduced new "upper"
RelOptInfo structures but didn't set consider_parallel for them
correctly, a point I completely missed when reviewing it. Later,
commit e06a38965b3bcdaa881e7e06892d4d8ab6c2c980 made the situation
worse by doing it incorrectly for the grouping relation. Try to
straighten all of that out. Along the way, get rid of the annoying
wholePlanParallelSafe flag, which was only necessarily because of
the fact that upper planning stages didn't use paths at the time
that code was written.
The most important immediate impact of these changes is that
force_parallel_mode will provide useful test coverage in quite a few
more scenarios than it did previously, but it's also necessary
preparation for fixing some problems related to subqueries.
Patch by me, reviewed by Tom Lane.
It's rather silly to make a separate pass over the tlist + HAVING qual,
and a separate set of visits to the syscache, when get_agg_clause_costs
already has all the required information in hand. This nets out as less
code as well as fewer cycles.
The original coding had three separate booleans representing partial
aggregation behavior, which was confusing, unreadable, and error-prone,
not least because the booleans weren't always listed in the same order.
It was also inadequate for the allegedly-desirable future extension to
support intermediate partial aggregation, because we'd need separate
markers for serialization and deserialization in such a case.
Merge these bools into an enum "AggSplit" to provide symbolic names for
the supported operating modes (and document what those are). By assigning
the values of the enum constants carefully, we can treat AggSplit values
as options bitmasks so that tests of what to do aren't noticeably more
expensive than before.
While at it, get rid of Aggref.aggoutputtype. That's not needed since
commit 59a3795c2 got rid of setrefs.c's special-purpose Aggref comparison
code, and it likewise seemed more confusing than helpful.
Assorted comment cleanup as well (there's still more that I want to do
in that line).
catversion bump for change in Aggref node contents. Should be the last
one for partial-aggregation changes.
Discussion: <29309.1466699160@sss.pgh.pa.us>
Commit e06a38965's original coding for constructing the execution-time
expression tree for a combining aggregate was rather messy, involving
duplicating quite a lot of code in setrefs.c so that it could inject
a nonstandard matching rule for Aggrefs. Get rid of that in favor of
explicitly constructing a combining Aggref with a partial Aggref as input,
then allowing setref's normal matching logic to match the partial Aggref
to the output of the lower plan node and hence replace it with a Var.
In passing, rename and redocument make_partialgroup_input_target to have
some connection to what it actually does.
The original upper-planner-pathification design (commit 3fc6e2d7f5b652b4)
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.
The previous code neglected the fact that the scanjoin_target might
carry sortgroupref labelings that we need to absorb. Instead, do
create_projection_path() unconditionally, and tweak the path's cost
estimate after the fact. (I'm now convinced that we ought to refactor
the way we account for sometimes not needing a separate projection step,
but right now is not the time for that sort of cleanup.)
Problem identified by Amit Kapila, patch by me.
Commit 04ae11f62e643e07c411c4935ea6af46cb112aa9 removed some broken
code to apply the scan/join target to partial paths, but its theory
that this processing step is totally unnecessary turns out to be wrong.
Put similar code back again, but this time, check for parallel-safety
and avoid in-place modifications to paths that may already have been
used as part of some other path.
(This is not an entirely elegant solution to this problem; it might
be better, for example, to postpone generate_gather_paths for the
topmost scan/join rel until after the scan/join target has been
applied. But this is not the time for such redesign work.)
Amit Kapila and Robert Haas
The struct definition for PathTarget specifies that a NULL sortgrouprefs
pointer means no sortgroupref labels. While it's likely that there
should always be at least one labeled column in the places that were
unconditionally fetching through the pointer, it seems wiser to adhere to
the data structure specification and test first. Add a macro to make this
convenient. Per experimentation with running the regression tests with a
very small parallelization threshold --- the crash I observed may well
represent a bug elsewhere, but still this coding was not very robust.
Report: <20756.1465834072@sss.pgh.pa.us>
This terminology provoked widespread complaints. So, instead, rename
the GUC max_parallel_degree to max_parallel_workers_per_gather
(leaving room for a possible future GUC max_parallel_workers that acts
as a system-wide limit), and rename the parallel_degree reloption to
parallel_workers. Rename structure members to match.
These changes create a dump/restore hazard for users of PostgreSQL
9.6beta1 who have set the reloption (or applied the GUC using ALTER
USER or ALTER DATABASE).
The partial paths that get modified may already have been used as
part of a GatherPath which appears in the path list, so modifying
them is not a good idea at this stage - especially because this
code has no check that the PathTarget is in fact parallel-safe.
When partial aggregation is being performed, this is actually
harmless because we'll end up replacing the pathtargets here with
the correct ones within create_grouping_paths(). But if we've got
a query tree containing only scan/join operations then this can
result in incorrectly pushing down parallel-restricted target
list entries. If those are, for example, references to subqueries,
that can crash the server; but it's wrong in any event.
Amit Kapila
subquery_planner() failed to apply expression preprocessing to the
arbiterElems and arbiterWhere fields of an OnConflictExpr. No doubt the
theory was that this wasn't necessary because we don't actually try to
execute those expressions; but that's wrong, because it results in failure
to match to index expressions or index predicates that are changed at all
by preprocessing. Per bug #14132 from Reynold Smith.
Also add pullup_replace_vars processing for onConflictWhere. Perhaps
it's impossible to have a subquery reference there, but I'm not exactly
convinced; and even if true today it's a failure waiting to happen.
Also add some comments to other places where one or another field of
OnConflictExpr is intentionally ignored, with explanation as to why it's
okay to do so.
Also, catalog/dependency.c failed to record any dependency on the named
constraint in ON CONFLICT ON CONSTRAINT, allowing such a constraint to
be dropped while rules exist that depend on it, and allowing pg_dump to
dump such a rule before the constraint it refers to. The normal execution
path managed to error out reasonably for a dangling constraint reference,
but ruleutils.c dumped core; so in addition to fixing the omission, add
a protective check in ruleutils.c, since we can't retroactively add a
dependency in existing databases.
Back-patch to 9.5 where this code was introduced.
Report: <20160510190350.2608.48667@wrigleys.postgresql.org>
The original patch kind of ignored the fact that we were doing something
different from a costing point of view, but nobody noticed. This patch
fixes that oversight.
David Rowley
Per discussion, this gives potential users of the hook more flexibility,
because they can build custom Paths that implement only one stage of
upper processing atop core-provided Paths for earlier stages.
This is necessary infrastructure for supporting parallel aggregation
for aggregates whose transition type is "internal". Such values
can't be passed between cooperating processes, because they are
just pointers.
David Rowley, reviewed by Tomas Vondra and by me.
cost_subplan() supposed that the given subplan must have plan_rows > 0,
which as far as I can tell was true until recent refactoring of the
code in createplan.c; but now that code allows the Result for a provably
empty subquery to have plan_rows = 0. Rather than undo that change,
put in a clamp to prevent zero divide.
get_cheapest_fractional_path() likewise supposed that best_path->rows > 0.
This assumption has been wrong for longer. It's actually harmless given
IEEE float math, because a positive value divided by zero gives +Infinity
and compare_fractional_path_costs() will do the right thing with that.
Still, best not to assume that.
final_cost_nestloop() also seems to have some risks in this area, so
borrow the clamping logic already present in the mergejoin cost functions.
Lastly, remove unnecessary clamp_row_est() in planner.c's calls to
get_number_of_groups(). The only thing that function does with path_rows
is pass it to estimate_num_groups() which already has an internal clamp,
so we don't need the extra call; and if we did, the callers are arguably
the wrong place for it anyway.
First two items reported by Piotr Stefaniak, the others are products
of my nosing around for similar problems. No back-patch since there's
no evidence that problems arise in the back branches.
In commit 9118d03a8cca3d97 we taught the planner to postpone evaluation of
set-returning functions in a SELECT's targetlist until after any sort done
to satisfy ORDER BY. However, if we postpone some SRFs this way while
others do not get postponed (because they're sort or group key columns)
we will break the traditional behavior by which all SRFs in the tlist run
in-step during ExecTargetList(), so that you get the least common multiple
of their periods not the product. Fix make_sort_input_target() so it will
not split up SRF evaluation in such cases.
There is still a hazard of similar odd behavior if there's a SRF in a
grouping column and another one that isn't, but that was true before
and we're just trying to preserve bug-compatibility with the traditional
behavior. This whole area is overdue to be rethought and reimplemented,
but we'll try to avoid changing behavior until then.
Per report from Regina Obe.
Parallel workers can now partially aggregate the data and pass the
transition values back to the leader, which can combine the partial
results to produce the final answer.
David Rowley, based on earlier work by Haribabu Kommi. Reviewed by
Álvaro Herrera, Tomas Vondra, Amit Kapila, James Sewell, and me.
This is basically like the just-added create_upper_paths_hook, but
control is funneled only to the FDW responsible for all the baserels
of the current query; so providing such a callback is much less likely
to add useless overhead than using the hook function is.
The documentation is a bit sketchy. We'll likely want to improve it,
and/or adjust the call conventions, when we get some experience with
actually using this callback. Hopefully somebody will find time to
experiment with it before 9.6 feature freeze.
In the initial revision of the upper-planner pathification work, the only
available way for an FDW or custom-scan provider to inject Paths
representing post-scan-join processing was to insert them during scan-level
GetForeignPaths or similar processing. While that's not impossible, it'd
require quite a lot of duplicative processing to look forward and see if
the extension would be capable of implementing the whole query. To improve
matters for custom-scan providers, provide a hook function at the point
where the core code is about to start filling in upperrel Paths. At this
point Paths are available for the whole scan/join tree, which should reduce
the amount of redundant effort considerably.
(An alternative design that was suggested was to provide a separate hook
for each post-scan-join processing step, but that seems messy and not
clearly more useful.)
Following our time-honored tradition, there's no documentation for this
hook outside the source code.
As-is, this hook is only meant for custom scan providers, which we can't
assume very much about. A followon patch will implement an FDW callback
to let FDWs do the same thing in a somewhat more structured fashion.
