The patches committed as 6e71dd7ce9
and 3a1f8611f2 were developed in
parallel but dependent on each other in a way that I failed to
notice.
This patch to fix the problem was prepared by Amit Kapila.
Commit 4a4e6893aa, which introduced this
mechanism, failed to account for the fact that the RECORD pseudo-type
uses transient typmods that are only meaningful within a single
backend. Transferring such tuples without modification between two
cooperating backends does not work. This commit installs a system
for passing the tuple descriptors over the same shm_mq being used to
send the tuples themselves. The two sides might not assign the same
transient typmod to any given tuple descriptor, so we must also
substitute the appropriate receiver-side typmod for the one used by
the sender. That adds some CPU overhead, but still seems better than
being unable to pass records between cooperating parallel processes.
Along the way, move the logic for handling multiple tuple queues from
tqueue.c to nodeGather.c; tqueue.c now provides a TupleQueueReader,
which reads from a single queue, rather than a TupleQueueFunnel, which
potentially reads from multiple queues. This change was suggested
previously as a way to make sure that nodeGather.c rather than tqueue.c
had policy control over the order in which to read from queues, but
it wasn't clear to me until now how good an idea it was. typmod
mapping needs to be performed separately for each queue, and it is
much simpler if the tqueue.c code handles that and leaves multiplexing
multiple queues to higher layers of the stack.
Commit d1b7c1ffe7 introduced a mechanism
for serializing a ParamListInfo structure to be passed to a parallel
worker. However, this mechanism failed to handle external expanded
values, as pointed out by Noah Misch. Repair.
Moreover, plpgsql_param_fetch requires adjustment because the
serialization mechanism needs it to skip evaluating unused parameters
just as we would do when it is called from copyParamList, but params
== estate->paramLI in that case. To fix, make the bms_is_member test
in that function unconditional.
Finally, have setup_param_list set a new ParamListInfo field,
paramMask, to the parameters actually used in the expression, so that
we don't try to fetch those that are not needed when serializing a
parameter list. This isn't necessary for correctness, but it makes
the performance of the parallel executor code comparable to what we
do for cases involving cursors.
Design suggestions and extensive review by Noah Misch. Patch by me.
Commit b0b0d84b3d purported to make it
possible to relaunch workers using the same parallel context, but it had
an unpleasant race condition: we might reinitialize after the workers
have sent their last control message but before they have dettached the
DSM, leaving to crashes. Repair by introducing a new ParallelContext
operation, ReinitializeParallelDSM.
Adjust execParallel.c to use this new support, so that we can rescan a
Gather node by relaunching workers but without needing to recreate the
DSM.
Amit Kapila, with some adjustments by me. Extracted from latest parallel
sequential scan patch.
The original Gather code failed to mark a Gather node as not able to
do projection, but it couldn't, even though it did call initialize its
projection info via ExecAssignProjectionInfo. There doesn't seem to
be any good reason for this node not to have projection capability,
so clean things up so that it does. Without this, plans using Gather
nodes might need to carry extra Result nodes to do projection.
Commit 816e336f12 added the wrong error
check to async.c; sending restrictions is restricted to the leader,
not altogether unsafe.
Commit 3bd909b220 added ExecShutdownNode
to traverse the planstate tree and call shutdown functions, but made
a Gather node, the only node that actually has such a function, abort
the tree traversal, which is wrong.
In the previous coding, before returning from ExecutorRun, we'd shut
down all parallel workers. This was dead wrong if ExecutorRun was
called with a non-zero tuple count; it had the effect of truncating
the query output. To fix, give ExecutePlan control over whether to
enter parallel mode, and have it refuse to do so if the tuple count
is non-zero. Rewrite the Gather logic so that it can cope with being
called outside parallel mode.
Commit 7aea8e4f2d is largely to blame
for this problem, though this patch modifies some subsequently-committed
code which relied on the guarantees it purported to make.
This fixes a long-standing bug which was discovered while investigating
the interaction between the new join pushdown code and the EvalPlanQual
machinery: if a ForeignScan appears on the inner side of a paramaterized
nestloop, an EPQ recheck would re-return the original tuple even if
it no longer satisfied the pushed-down quals due to changed parameter
values.
This fix adds a new member to ForeignScan and ForeignScanState and a
new argument to make_foreignscan, and requires changes to FDWs which
push down quals to populate that new argument with a list of quals they
have chosen to push down. Therefore, I'm only back-patching to 9.5,
even though the bug is not new in 9.5.
Etsuro Fujita, reviewed by me and by Kyotaro Horiguchi.
On reflection, the submitted patch didn't really work to prevent the
request size from exceeding MaxAllocSize, because of the fact that we'd
happily round nbuckets up to the next power of 2 after we'd limited it to
max_pointers. The simplest way to enforce the limit correctly is to
round max_pointers down to a power of 2 when it isn't one already.
(Note that the constraint to INT_MAX / 2, if it were doing anything useful
at all, is properly applied after that.)
Limit the size of the hashtable pointer array to not more than
MaxAllocSize, per reports from Kouhei Kaigai and others of "invalid memory
alloc request size" failures. There was discussion of allowing the array
to get larger than that by using the "huge" palloc API, but so far no proof
that that is actually a good idea, and at this point in the 9.5 cycle major
changes from old behavior don't seem like the way to go.
Fix a rather serious secondary bug in the new code, which was that it
didn't ensure nbuckets remained a power of 2 when recomputing it for the
multiple-batch case.
Clean up sloppy division of labor between ExecHashIncreaseNumBuckets and
its sole call site.
A Gather executor node runs any number of copies of a plan in an equal
number of workers and merges all of the results into a single tuple
stream. It can also run the plan itself, if the workers are
unavailable or haven't started up yet. It is intended to work with
the Partial Seq Scan node which will be added in future commits.
It could also be used to implement parallel query of a different sort
by itself, without help from Partial Seq Scan, if the single_copy mode
is used. In that mode, a worker executes the plan, and the parallel
leader does not, merely collecting the worker's results. So, a Gather
node could be inserted into a plan to split the execution of that plan
across two processes. Nested Gather nodes aren't currently supported,
but we might want to add support for that in the future.
There's nothing in the planner to actually generate Gather nodes yet,
so it's not quite time to break out the champagne. But we're getting
close.
Amit Kapila. Some designs suggestions were provided by me, and I also
reviewed the patch. Single-copy mode, documentation, and other minor
changes also by me.
This code provides infrastructure for a parallel leader to start up
parallel workers to execute subtrees of the plan tree being executed
in the master. User-supplied parameters from ParamListInfo are passed
down, but PARAM_EXEC parameters are not. Various other constructs,
such as initplans, subplans, and CTEs, are also not currently shared.
Nevertheless, there's enough here to support a basic implementation of
parallel query, and we can lift some of the current restrictions as
needed.
Amit Kapila and Robert Haas
When taking the UPDATE path in an INSERT .. ON CONFLICT .. UPDATE tables
with oids were not supported. The tuple generated by the update target
list was projected without space for an oid - a simple oversight.
Reported-By: Peter Geoghegan
Author: Andres Freund
Backpatch: 9.5, where ON CONFLICT was introduced
The shm_mq mechanism was built to send error (and notice) messages and
tuples between backends. However, shm_mq itself only deals in raw
bytes. Since commit 2bd9e412f9, we have
had infrastructure for one message to redirect protocol messages to a
queue and for another backend to parse them and do useful things with
them. This commit introduces a somewhat analogous facility for tuples
by adding a new type of DestReceiver, DestTupleQueue, which writes
each tuple generated by a query into a shm_mq, and a new
TupleQueueFunnel facility which reads raw tuples out of the queue and
reconstructs the HeapTuple format expected by the executor.
The TupleQueueFunnel abstraction supports reading from multiple tuple
streams at the same time, but only in round-robin fashion. Someone
could imaginably want other policies, but this should be good enough
to meet our short-term needs related to parallel query, and we can
always extend it later.
This also makes one minor addition to the shm_mq API that didn'
seem worth breaking out as a separate patch.
Extracted from Amit Kapila's parallel sequential scan patch. This
code was originally written by me, and then it was revised by Amit,
and then it was revised some more by me.
Commit 924bcf4f16 introduced a framework
for parallel computation in PostgreSQL that makes most but not all
built-in functions safe to execute in parallel mode. In order to have
parallel query, we'll need to be able to determine whether that query
contains functions (either built-in or user-defined) that cannot be
safely executed in parallel mode. This requires those functions to be
labeled, so this patch introduces an infrastructure for that. Some
functions currently labeled as safe may need to be revised depending on
how pending issues related to heavyweight locking under paralllelism
are resolved.
Parallel plans can't be used except for the case where the query will
run to completion. If portal execution were suspended, the parallel
mode restrictions would need to remain in effect during that time, but
that might make other queries fail. Therefore, this patch introduces
a framework that enables consideration of parallel plans only when it
is known that the plan will be run to completion. This probably needs
some refinement; for example, at bind time, we do not know whether a
query run via the extended protocol will be execution to completion or
run with a limited fetch count. Having the client indicate its
intentions at bind time would constitute a wire protocol break. Some
contexts in which parallel mode would be safe are not adjusted by this
patch; the default is not to try parallel plans except from call sites
that have been updated to say that such plans are OK.
This commit doesn't introduce any parallel paths or plans; it just
provides a way to determine whether they could potentially be used.
I'm committing it on the theory that the remaining parallel sequential
scan patches will also get committed to this release, hopefully in the
not-too-distant future.
Robert Haas and Amit Kapila. Reviewed (in earlier versions) by Noah
Misch.
This refactors rewrite/rowsecurity.c to simplify the handling of the
default deny case (reducing the number of places where we check for and
add the default deny policy from three to one) by splitting up the
retrival of the policies from the application of them.
This also allowed us to do away with the policy_id field. A policy_name
field was added for WithCheckOption policies and is used in error
reporting, when available.
Patch by Dean Rasheed, with various mostly cosmetic changes by me.
Back-patch to 9.5 where RLS was introduced to avoid unnecessary
differences, since we're still in alpha, per discussion with Robert.
Per discussion, nowadays it is possible to have tablespaces that have
wildly different I/O characteristics from others. Setting different
effective_io_concurrency parameters for those has been measured to
improve performance.
Author: Julien Rouhaud
Reviewed by: Andres Freund
For no obvious reason, spi_printtup() was coded to enlarge the tuple
pointer table by just 256 slots at a time, rather than doubling the size at
each reallocation, as is our usual habit. For very large SPI results, this
makes for O(N^2) time spent in repalloc(), which of course soon comes to
dominate the runtime. Use the standard doubling approach instead.
This is a longstanding performance bug, so back-patch to all active
branches.
Neil Conway
Commit 8cce08f168 used a left-shift
on a literal of 1 that could (in large allocations) be shifted by
31 or more bits. This was assigned to a local variable that was
already declared to be a long to protect against overruns of int,
but the literal in this shift needs to be declared long to allow it
to work correctly in some compilers.
Backpatch to 9.5, where the bug was introduced.
Report and patch by KaiGai Kohei, slighly modified based on
discussion.
It was a bool, even though it should be CEOUC_WAIT_MODE. That's unlikely
to have a negative effect with the current definition of bool (char),
but it's definitely wrong.
Discussion: 20150812084351.GD8470@awork2.anarazel.de
Backpatch: 9.5, where ON CONFLICT was merged
If there are two different aggregates in the query with same inputs, and
the aggregates have the same initial condition and transition function,
only calculate the state value once, and only call the final functions
separately. For example, AVG(x) and SUM(x) aggregates have the same
transition function, which accumulates the sum and number of input tuples.
For a query like "SELECT AVG(x), SUM(x) FROM x", we can therefore
accumulate the state function only once, which gives a nice speedup.
David Rowley, reviewed and edited by me.
It's against project policy to use elog() for user-facing errors, or to
omit an errcode() selection for errors that aren't supposed to be "can't
happen" cases. Fix all the violations of this policy that result in
ERRCODE_INTERNAL_ERROR log entries during the standard regression tests,
as errors that can reliably be triggered from SQL surely should be
considered user-facing.
I also looked through all the files touched by this commit and fixed
other nearby problems of the same ilk. I do not claim to have fixed
all violations of the policy, just the ones in these files.
In a few places I also changed existing ERRCODE choices that didn't
seem particularly appropriate; mainly replacing ERRCODE_SYNTAX_ERROR
by something more specific.
Back-patch to 9.5, but no further; changing ERRCODE assignments in
stable branches doesn't seem like a good idea.
Although I think on all modern machines floating division by zero
results in Infinity not SIGFPE, we still don't want infinities
running around in the planner's costing estimates; too much risk
of that leading to insane behavior.
grouping_planner() failed to consider the possibility that final_rel
might be known dummy and hence have zero rowcount. (I wonder if it
would be better to set a rows estimate of 1 for dummy relations?
But at least in the back branches, changing this convention seems
like a bad idea, so I'll leave that for another day.)
Make certain that get_variable_numdistinct() produces a nonzero result.
The case that can be shown to be broken is with stadistinct < 0.0 and
small ntuples; we did not prevent the result from rounding to zero.
For good luck I applied clamp_row_est() to all the nonconstant return
values.
In ExecChooseHashTableSize(), Assert that we compute positive nbuckets
and nbatch. I know of no reason to think this isn't the case, but it
seems like a good safety check.
Per reports from Piotr Stefaniak. Back-patch to all active branches.
The pg_stats view is supposed to be restricted to only show rows
about tables the user can read. However, it sometimes can leak
information which could not otherwise be seen when row level security
is enabled. Fix that by not showing pg_stats rows to users that would
be subject to RLS on the table the row is related to. This is done
by creating/using the newly introduced SQL visible function,
row_security_active().
Along the way, clean up three call sites of check_enable_rls(). The second
argument of that function should only be specified as other than
InvalidOid when we are checking as a different user than the current one,
as in when querying through a view. These sites were passing GetUserId()
instead of InvalidOid, which can cause the function to return incorrect
results if the current user has the BYPASSRLS privilege and row_security
has been set to OFF.
Additionally fix a bug causing RI Trigger error messages to unintentionally
leak information when RLS is enabled, and other minor cleanup and
improvements. Also add WITH (security_barrier) to the definition of pg_stats.
Bumped CATVERSION due to new SQL functions and pg_stats view definition.
Back-patch to 9.5 where RLS was introduced. Reported by Yaroslav.
Patch by Joe Conway and Dean Rasheed with review and input by
Michael Paquier and Stephen Frost.
There is no full discussion of speculative insertions in the executor
README. There is a high-level explanation in execIndexing.c, but it doesn't
seem necessary to refer it from here.
Peter Geoghegan
The original implementation of TABLESAMPLE modeled the tablesample method
API on index access methods, which wasn't a good choice because, without
specialized DDL commands, there's no way to build an extension that can
implement a TSM. (Raw inserts into system catalogs are not an acceptable
thing to do, because we can't undo them during DROP EXTENSION, nor will
pg_upgrade behave sanely.) Instead adopt an API more like procedural
language handlers or foreign data wrappers, wherein the only SQL-level
support object needed is a single handler function identified by having
a special return type. This lets us get rid of the supporting catalog
altogether, so that no custom DDL support is needed for the feature.
Adjust the API so that it can support non-constant tablesample arguments
(the original coding assumed we could evaluate the argument expressions at
ExecInitSampleScan time, which is undesirable even if it weren't outright
unsafe), and discourage sampling methods from looking at invisible tuples.
Make sure that the BERNOULLI and SYSTEM methods are genuinely repeatable
within and across queries, as required by the SQL standard, and deal more
honestly with methods that can't support that requirement.
Make a full code-review pass over the tablesample additions, and fix
assorted bugs, omissions, infelicities, and cosmetic issues (such as
failure to put the added code stanzas in a consistent ordering).
Improve EXPLAIN's output of tablesample plans, too.
Back-patch to 9.5 so that we don't have to support the original API
in production.
Fix some places where pgindent did silly stuff, often because project
style wasn't followed to begin with. (I've not touched the atomics
headers, though.)
Fix confusion in documentation, substantial memory leakage if float8 or
float4 are pass-by-reference, and assorted comments that were obsoleted
by commit 98edd617f3.
Paul Ramsey reported that commit 35fcb1b3d0
induced a core dump on commuted ORDER BY expressions, because it was
assuming that the indexorderby expression could be found verbatim in the
relevant equivalence class, but it wasn't there. We really don't need
anything that complicated anyway; for the data types likely to be used for
index ORDER BY operators in the foreseeable future, the exprType() of the
ORDER BY expression will serve fine. (The case where we'd have to work
harder is where the ORDER BY expression's result is only binary-compatible
with the declared input type of the ordering operator; long before worrying
about that, one would need to get rid of GiST's hard-wired assumption that
said datatype is float8.)
Aside from fixing that crash and adding a regression test for the case,
I did some desultory code review:
nodeIndexscan.c was likewise overthinking how hard it ought to work to
identify the datatype of the ORDER BY expressions.
Add comments explaining how come nodeIndexscan.c can get away with
simplifying assumptions about NULLS LAST ordering and no backward scan.
Revert no-longer-needed changes of find_ec_member_for_tle(); while the
new definition was no worse than the old, it wasn't better either, and
it might cause back-patching pain.
Revert entirely bogus additions to genam.h.
Use "a" and "an" correctly, mostly in comments. Two error messages were
also fixed (they were just elogs, so no translation work required). Two
function comments in pg_proc.h were also fixed. Etsuro Fujita reported one
of these, but I found a lot more with grep.
Also fix a few other typos spotted while grepping for the a/an typos.
For example, "consists out of ..." -> "consists of ...". Plus a "though"/
"through" mixup reported by Euler Taveira.
Many of these typos were in old code, which would be nice to backpatch to
make future backpatching easier. But much of the code was new, and I didn't
feel like crafting separate patches for each branch. So no backpatching.
The previous coding was a leftover from attempting to hang all the on
conflict logic onto modify table's child nodes. It appears to not have
actually caused problems except for explain.
Add test exercising the broken and some other code paths.
Author: Peter Geoghegan and Andres Freund
This oversight results in a crash at executor startup if the plan has
been copied. outfuncs.c was missed as well.
While we could probably have taught both those files to cope with the
originally chosen representation of an Oid array, it would have been
painful, not least because there'd be no easy way to verify the array
length. An Oid List is far easier to work with. And AFAICS, there is
no particular notational benefit to using an array rather than a list
in the existing parts of the patch either. So just change it to a list.
Error in commit 35fcb1b3d0, which is new,
so no need for back-patch.
This SQL standard functionality allows to aggregate data by different
GROUP BY clauses at once. Each grouping set returns rows with columns
grouped by in other sets set to NULL.
This could previously be achieved by doing each grouping as a separate
query, conjoined by UNION ALLs. Besides being considerably more concise,
grouping sets will in many cases be faster, requiring only one scan over
the underlying data.
The current implementation of grouping sets only supports using sorting
for input. Individual sets that share a sort order are computed in one
pass. If there are sets that don't share a sort order, additional sort &
aggregation steps are performed. These additional passes are sourced by
the previous sort step; thus avoiding repeated scans of the source data.
The code is structured in a way that adding support for purely using
hash aggregation or a mix of hashing and sorting is possible. Sorting
was chosen to be supported first, as it is the most generic method of
implementation.
Instead of, as in an earlier versions of the patch, representing the
chain of sort and aggregation steps as full blown planner and executor
nodes, all but the first sort are performed inside the aggregation node
itself. This avoids the need to do some unusual gymnastics to handle
having to return aggregated and non-aggregated tuples from underlying
nodes, as well as having to shut down underlying nodes early to limit
memory usage. The optimizer still builds Sort/Agg node to describe each
phase, but they're not part of the plan tree, but instead additional
data for the aggregation node. They're a convenient and preexisting way
to describe aggregation and sorting. The first (and possibly only) sort
step is still performed as a separate execution step. That retains
similarity with existing group by plans, makes rescans fairly simple,
avoids very deep plans (leading to slow explains) and easily allows to
avoid the sorting step if the underlying data is sorted by other means.
A somewhat ugly side of this patch is having to deal with a grammar
ambiguity between the new CUBE keyword and the cube extension/functions
named cube (and rollup). To avoid breaking existing deployments of the
cube extension it has not been renamed, neither has cube been made a
reserved keyword. Instead precedence hacking is used to make GROUP BY
cube(..) refer to the CUBE grouping sets feature, and not the function
cube(). To actually group by a function cube(), unlikely as that might
be, the function name has to be quoted.
Needs a catversion bump because stored rules may change.
Author: Andrew Gierth and Atri Sharma, with contributions from Andres Freund
Reviewed-By: Andres Freund, Noah Misch, Tom Lane, Svenne Krap, Tomas
Vondra, Erik Rijkers, Marti Raudsepp, Pavel Stehule
Discussion: CAOeZVidmVRe2jU6aMk_5qkxnB7dfmPROzM7Ur8JPW5j8Y5X-Lw@mail.gmail.com
Add a TABLESAMPLE clause to SELECT statements that allows
user to specify random BERNOULLI sampling or block level
SYSTEM sampling. Implementation allows for extensible
sampling functions to be written, using a standard API.
Basic version follows SQLStandard exactly. Usable
concrete use cases for the sampling API follow in later
commits.
Petr Jelinek
Reviewed by Michael Paquier and Simon Riggs
We can only support a lossy distance function when the distance function's
datatype is comparable with the original ordering operator's datatype.
The distance function always returns a float8, so we are limited to float8,
and float4 (by a hard-coded cast of the float8 to float4).
In light of this limitation, it seems like a good idea to have a separate
'recheck' flag for the ORDER BY expressions, so that if you have a non-lossy
distance function, it still works with lossy quals. There are cases like
that with the build-in or contrib opclasses, but it's plausible.
There was a hidden assumption that the ORDER BY values returned by GiST
match the original ordering operator's return type, but there are plenty
of examples where that's not true, e.g. in btree_gist and pg_trgm. As long
as the distance function is not lossy, we can tolerate that and just not
return the distance to the executor (or rather, always return NULL). The
executor doesn't need the distances if there are no lossy results.
There was another little bug: the recheck variable was not initialized
before calling the distance function. That revealed the bigger issue,
as the executor tried to reorder tuples that didn't need reordering, and
that failed because of the datatype mismatch.
The distance function can now set *recheck = false, like index quals. The
executor will then re-check the ORDER BY expressions, and use a queue to
reorder the results on the fly.
This makes it possible to do kNN-searches on polygons and circles, which
don't store the exact value in the index, but just a bounding box.
Alexander Korotkov and me
Robert Haas
Peter Eisentraut
Tom Lane
Andres Freund
Stephen Frost
Alvaro Herrera
Kevin Grittner
Heikki Linnakangas
Joe Conway
Bruce Momjian
Simon Riggs