The strategy of GIN index scan is driven by opclass-specific extract_query
method. This method that needed search mode is GIN_SEARCH_MODE_ALL. This
mode means that matching tuple may contain none of extracted entries. Simple
example is '!term' tsquery, which doesn't need any term to exist in matching
tsvector.
In order to handle such scan key GIN calculates virtual entry, which contains
all TIDs of all entries of attribute. In fact this is full scan of index
attribute. And typically this is very slow, but allows to handle some queries
correctly in GIN. However, current algorithm calculate such virtual entry for
each GIN_SEARCH_MODE_ALL scan key even if they are multiple for the same
attribute. This is clearly not optimal.
This commit improves the situation by introduction of "exclude only" scan keys.
Such scan keys are not capable to return set of matching TIDs. Instead, they
are capable only to filter TIDs produced by normal scan keys. Therefore,
each attribute should contain at least one normal scan key, while rest of them
may be "exclude only" if search mode is GIN_SEARCH_MODE_ALL.
The same optimization might be applied to the whole scan, not per-attribute.
But that leads to NULL values elimination problem. There is trade-off between
multiple possible ways to do this. We probably want to do this later using
some cost-based decision algorithm.
Discussion: https://postgr.es/m/CAOBaU_YGP5-BEt5Cc0%3DzMve92vocPzD%2BXiZgiZs1kjY0cj%3DXBg%40mail.gmail.com
Author: Nikita Glukhov, Alexander Korotkov, Tom Lane, Julien Rouhaud
Reviewed-by: Julien Rouhaud, Tomas Vondra, Tom Lane
This provides for cheaper mapping of child columns back to parent
columns. The one existing use-case in examine_simple_variable()
would hardly justify this by itself; but an upcoming bug fix will
make use of this array in a mainstream code path, and it seems
likely that we'll find other uses for it as we continue to build
out the partitioning infrastructure.
Discussion: https://postgr.es/m/12424.1575168015@sss.pgh.pa.us
The fix for CVE-2017-7484 disallowed use of pg_statistic data for
planning purposes if the user would not be able to select the associated
column and a non-leakproof function is to be applied to the statistics
values. That turns out to disable use of pg_statistic data in some
common cases involving inheritance/partitioning, where the user does
have permission to select from the parent table that was actually named
in the query, but not from a child table whose stats are needed. Since,
in non-corner cases, the user *can* select the child table's data via
the parent, this restriction is not actually useful from a security
standpoint. Improve the logic so that we also check the permissions of
the originally-named table, and allow access if select permission exists
for that.
When checking access to stats for a simple child column, we can map
the child column number back to the parent, and perform this test
exactly (including not allowing access if the child column isn't
exposed by the parent). For expression indexes, the current logic
just insists on whole-table select access, and this patch allows
access if the user can select the whole parent table. In principle,
if the child table has extra columns, this might allow access to
stats on columns the user can't read. In practice, it's unlikely
that the planner is going to do any stats calculations involving
expressions that are not visible to the query, so we'll ignore that
fine point for now. Perhaps someday we'll improve that logic to
detect exactly which columns are used by an expression index ...
but today is not that day.
Back-patch to v11. The issue was created in 9.2 and up by the
CVE-2017-7484 fix, but this patch depends on the append_rel_array[]
planner data structure which only exists in v11 and up. In
practice the issue is most urgent with partitioned tables, so
fixing v11 and later should satisfy much of the practical need.
Dilip Kumar and Amit Langote, with some kibitzing by me
Discussion: https://postgr.es/m/3876.1531261875@sss.pgh.pa.us
Trying to use hypothetical indexes with BRIN currently fails when trying
to access a relation that does not exist when looking for the
statistics. With the current API, it is not possible to easily pass
a value for pages_per_range down to the hypothetical index, so this
makes use of the default value of BRIN_DEFAULT_PAGES_PER_RANGE, which
should be fine enough in most cases.
Being able to refine or enforce the hypothetical costs in more
optimistic ways would require more refactoring by filling in the
statistics when building IndexOptInfo in plancat.c. This would involve
ABI breakages around the costing routines, something not fit for stable
branches.
This is broken since 7e534ad, so backpatch down to v10.
Author: Julien Rouhaud, Heikki Linnakangas
Reviewed-by: Álvaro Herrera, Tom Lane, Michael Paquier
Discussion: https://postgr.es/m/CAOBaU_ZH0LKEA8VFCocr6Lpte1ab0b6FpvgS0y4way+RPSXfYg@mail.gmail.com
Backpatch-through: 10
When estimating number of distinct groups, we failed to ignore system
attributes when matching the group expressions to mvdistinct stats,
causing failures like
ERROR: negative bitmapset member not allowed
Fix that by simply skipping anything that is not a regular attribute.
Backpatch to PostgreSQL 10, where the extended stats were introduced.
Bug: #16111
Reported-by: Tuomas Leikola
Author: Tomas Vondra
Backpatch-through: 10
Discussion: https://postgr.es/m/16111-687799584c3a7e73@postgresql.org
As of d9dd406fe2, we require MSVC 2013,
which means _MSC_VER >= 1800. This means that conditionals about
older versions of _MSC_VER can be removed or simplified.
Previous code was also in some cases handling MinGW, where _MSC_VER is
not defined at all, incorrectly, such as in pg_ctl.c and win32_port.h,
leading to some compiler warnings. This should now be handled better.
Reviewed-by: Michael Paquier <michael@paquier.xyz>
In the wake of commit 1cff1b95a, the result of list_concat no longer
shares the ListCells of the second input. Therefore, we can replace
"list_concat(x, list_copy(y))" with just "list_concat(x, y)".
To improve call sites that were list_copy'ing the first argument,
or both arguments, invent "list_concat_copy()" which produces a new
list sharing no ListCells with either input. (This is a bit faster
than "list_concat(list_copy(x), y)" because it makes the result list
the right size to start with.)
In call sites that were not list_copy'ing the second argument, the new
semantics mean that we are usually leaking the second List's storage,
since typically there is no remaining pointer to it. We considered
inventing another list_copy variant that would list_free the second
input, but concluded that for most call sites it isn't worth worrying
about, given the relative compactness of the new List representation.
(Note that in cases where such leakage would happen, the old code
already leaked the second List's header; so we're only discussing
the size of the leak not whether there is one. I did adjust two or
three places that had been troubling to free that header so that
they manually free the whole second List.)
Patch by me; thanks to David Rowley for review.
Discussion: https://postgr.es/m/11587.1550975080@sss.pgh.pa.us
Formerly, lcons was about the same speed as lappend, but with the new
List implementation, that's not so; with a long List, data movement
imposes an O(N) cost on lcons and list_delete_first, but not lappend.
Hence, invent list_delete_last with semantics parallel to
list_delete_first (but O(1) cost), and change various places to use
lappend and list_delete_last where this can be done without much
violence to the code logic.
There are quite a few places that construct result lists using lcons not
lappend. Some have semantic rationales for that; I added comments about
it to a couple that didn't have them already. In many such places though,
I think the coding is that way only because back in the dark ages lcons
was faster than lappend. Hence, switch to lappend where this can be done
without causing semantic changes.
In ExecInitExprRec(), this results in aggregates and window functions that
are in the same plan node being executed in a different order than before.
Generally, the executions of such functions ought to be independent of
each other, so this shouldn't result in visibly different query results.
But if you push it, as one regression test case does, you can show that
the order is different. The new order seems saner; it's closer to
the order of the functions in the query text. And we never documented
or promised anything about this, anyway.
Also, in gistfinishsplit(), don't bother building a reverse-order list;
it's easy now to iterate backwards through the original list.
It'd be possible to go further towards removing uses of lcons and
list_delete_first, but it'd require more extensive logic changes,
and I'm not convinced it's worth it. Most of the remaining uses
deal with queues that probably never get long enough to be worth
sweating over. (Actually, I doubt that any of the changes in this
patch will have measurable performance effects either. But better
to have good examples than bad ones in the code base.)
Patch by me, thanks to David Rowley and Daniel Gustafsson for review.
Discussion: https://postgr.es/m/21272.1563318411@sss.pgh.pa.us
Originally, Postgres Lists were a more or less exact reimplementation of
Lisp lists, which consist of chains of separately-allocated cons cells,
each having a value and a next-cell link. We'd hacked that once before
(commit d0b4399d8) to add a separate List header, but the data was still
in cons cells. That makes some operations -- notably list_nth() -- O(N),
and it's bulky because of the next-cell pointers and per-cell palloc
overhead, and it's very cache-unfriendly if the cons cells end up
scattered around rather than being adjacent.
In this rewrite, we still have List headers, but the data is in a
resizable array of values, with no next-cell links. Now we need at
most two palloc's per List, and often only one, since we can allocate
some values in the same palloc call as the List header. (Of course,
extending an existing List may require repalloc's to enlarge the array.
But this involves just O(log N) allocations not O(N).)
Of course this is not without downsides. The key difficulty is that
addition or deletion of a list entry may now cause other entries to
move, which it did not before.
For example, that breaks foreach() and sister macros, which historically
used a pointer to the current cons-cell as loop state. We can repair
those macros transparently by making their actual loop state be an
integer list index; the exposed "ListCell *" pointer is no longer state
carried across loop iterations, but is just a derived value. (In
practice, modern compilers can optimize things back to having just one
loop state value, at least for simple cases with inline loop bodies.)
In principle, this is a semantics change for cases where the loop body
inserts or deletes list entries ahead of the current loop index; but
I found no such cases in the Postgres code.
The change is not at all transparent for code that doesn't use foreach()
but chases lists "by hand" using lnext(). The largest share of such
code in the backend is in loops that were maintaining "prev" and "next"
variables in addition to the current-cell pointer, in order to delete
list cells efficiently using list_delete_cell(). However, we no longer
need a previous-cell pointer to delete a list cell efficiently. Keeping
a next-cell pointer doesn't work, as explained above, but we can improve
matters by changing such code to use a regular foreach() loop and then
using the new macro foreach_delete_current() to delete the current cell.
(This macro knows how to update the associated foreach loop's state so
that no cells will be missed in the traversal.)
There remains a nontrivial risk of code assuming that a ListCell *
pointer will remain good over an operation that could now move the list
contents. To help catch such errors, list.c can be compiled with a new
define symbol DEBUG_LIST_MEMORY_USAGE that forcibly moves list contents
whenever that could possibly happen. This makes list operations
significantly more expensive so it's not normally turned on (though it
is on by default if USE_VALGRIND is on).
There are two notable API differences from the previous code:
* lnext() now requires the List's header pointer in addition to the
current cell's address.
* list_delete_cell() no longer requires a previous-cell argument.
These changes are somewhat unfortunate, but on the other hand code using
either function needs inspection to see if it is assuming anything
it shouldn't, so it's not all bad.
Programmers should be aware of these significant performance changes:
* list_nth() and related functions are now O(1); so there's no
major access-speed difference between a list and an array.
* Inserting or deleting a list element now takes time proportional to
the distance to the end of the list, due to moving the array elements.
(However, it typically *doesn't* require palloc or pfree, so except in
long lists it's probably still faster than before.) Notably, lcons()
used to be about the same cost as lappend(), but that's no longer true
if the list is long. Code that uses lcons() and list_delete_first()
to maintain a stack might usefully be rewritten to push and pop at the
end of the list rather than the beginning.
* There are now list_insert_nth...() and list_delete_nth...() functions
that add or remove a list cell identified by index. These have the
data-movement penalty explained above, but there's no search penalty.
* list_concat() and variants now copy the second list's data into
storage belonging to the first list, so there is no longer any
sharing of cells between the input lists. The second argument is
now declared "const List *" to reflect that it isn't changed.
This patch just does the minimum needed to get the new implementation
in place and fix bugs exposed by the regression tests. As suggested
by the foregoing, there's a fair amount of followup work remaining to
do.
Also, the ENABLE_LIST_COMPAT macros are finally removed in this
commit. Code using those should have been gone a dozen years ago.
Patch by me; thanks to David Rowley, Jesper Pedersen, and others
for review.
Discussion: https://postgr.es/m/11587.1550975080@sss.pgh.pa.us
Commit 3ca930fc3 modified get_actual_variable_range() to use a new
"SnapshotNonVacuumable" snapshot type for selecting tuples that it
would consider valid. However, because that snapshot type can accept
recently-dead tuples, this caused a bug when using a recently-created
index: we might accept a recently-dead tuple that is an early member
of a broken HOT chain and does not actually match the index entry.
Then, the data extracted from the heap tuple would not necessarily be
an endpoint value of the column; it could even be NULL, leading to
get_actual_variable_range() itself reporting "found unexpected null
value in index". Even without an error, this could lead to poor
plan choices due to an erroneous notion of the endpoint value.
We can improve matters by changing the code to use the index-only
scan technique (which didn't exist when get_actual_variable_range was
originally written). If any of the tuples in a HOT chain are live
enough to satisfy SnapshotNonVacuumable, we take the data from the
index entry, ignoring what is in the heap. This fixes the problem
without changing the live-vs-dead-tuple behavior from what was
intended by commit 3ca930fc3.
A side benefit is that for static tables we might not have to touch
the heap at all (when the extremal value is in an all-visible page).
In addition, we can save some overhead by not having to create a
complete ExecutorState, and we don't need to run FormIndexDatum,
avoiding more cycles as well as the possibility of failure for
indexes on expressions. (I'm not sure that this code would ever
be used to determine the extreme value of an expression, in the
current state of the planner; but it's definitely possible that
lower-order columns of the selected index could be expressions.
So one could construct perhaps-artificial examples in which the
old code unexpectedly failed due to trying to compute an
expression's value for a now-dead row.)
Per report from Manuel Rigger. Back-patch to v11 where commit
3ca930fc3 came in.
Discussion: https://postgr.es/m/CA+u7OA7W4NWEhCvftdV6_8bbm2vgypi5nuxfnSEJQqVKFSUoMg@mail.gmail.com
In examine_variable() and examine_simple_variable(), when checking the
user's table and column privileges to determine whether to grant
access to the pg_statistic data, use checkAsUser for the privilege
checks, if it's set. This will be the case if we're accessing the
table via a view, to indicate that we should perform privilege checks
as the view owner rather than the current user.
This change makes this planner check consistent with the check in the
executor, so the planner will be able to make use of statistics if the
table is accessible via the view. This fixes a performance regression
introduced by commit e2d4ef8de8, which affects queries against
non-security barrier views in the case where the user doesn't have
privileges on the underlying table, but the view owner does.
Note that it continues to provide the same safeguards controlling
access to pg_statistic for direct table access (in which case
checkAsUser won't be set) and for security barrier views, because of
the nearby checks on rte->security_barrier and rte->securityQuals.
Back-patch to all supported branches because e2d4ef8de8 was.
Dean Rasheed, reviewed by Jonathan Katz and Stephen Frost.
In commit e2d4ef8de8, security checks were added to prevent
user-supplied operators from running over data from pg_statistic
unless the user has table or column privileges on the table, or the
operator is leakproof. For a table with RLS, however, checking for
table or column privileges is insufficient, since that does not
guarantee that the user has permission to view all of the column's
data.
Fix this by also checking for securityQuals on the RTE, and insisting
that the operator be leakproof if there are any. Thus the
leakproofness check will only be skipped if there are no securityQuals
and the user has table or column privileges on the table -- i.e., only
if we know that the user has access to all the data in the column.
Back-patch to 9.5 where RLS was added.
Dean Rasheed, reviewed by Jonathan Katz and Stephen Frost.
Security: CVE-2019-10130
The assertions added by commit b04aeb0a0 exposed that there are some
code paths wherein the executor will try to open an index without
holding any lock on it. We do have some lock on the index's table,
so it seems likely that there's no fatal problem with this (for
instance, the index couldn't get dropped from under us). Still,
it's bad practice and we should fix it.
To do so, remove the optimizations in ExecInitIndexScan and friends
that tried to avoid taking a lock on an index belonging to a target
relation, and just take the lock always. In non-bug cases, this
will result in no additional shared-memory access, since we'll find
in the local lock table that we already have a lock of the desired
type; hence, no significant performance degradation should occur.
Also, adjust the planner and executor so that the type of lock taken
on an index is always identical to the type of lock taken for its table,
by relying on the recently added RangeTblEntry.rellockmode field.
This avoids some corner cases where that might not have been true
before (possibly resulting in extra locking overhead), and prevents
future maintenance issues from having multiple bits of logic that
all needed to be in sync. In addition, this change removes all core
calls to ExecRelationIsTargetRelation, which avoids a possible O(N^2)
startup penalty for queries with large numbers of target relations.
(We'd probably remove that function altogether, were it not that we
advertise it as something that FDWs might want to use.)
Also adjust some places in selfuncs.c to not take any lock on indexes
they are transiently opening, since we can assume that plancat.c
did that already.
In passing, change gin_clean_pending_list() to take RowExclusiveLock
not AccessShareLock on its target index. Although it's not clear that
that's actually a bug, it seemed very strange for a function that's
explicitly going to modify the index to use only AccessShareLock.
David Rowley, reviewed by Julien Rouhaud and Amit Langote,
a bit of further tweaking by me
Discussion: https://postgr.es/m/19465.1541636036@sss.pgh.pa.us
We were getting just DEFAULT_INEQ_SEL for comparisons such as
"ctid >= constant", but it's possible to do a lot better if we don't
mind some assumptions about the table's tuple density being reasonably
uniform. There are already assumptions much like that elsewhere in
the planner, so that hardly seems like much of an objection.
Extracted from a patch set that also proposes to introduce a special
executor node type for such queries. Not sure if that's going to make
it into v12, but improving the selectivity estimate is useful
independently of that.
Edmund Horner, reviewed by David Rowley
Discussion: https://postgr.es/m/CAMyN-kB-nFTkF=VA_JPwFNo08S0d-Yk0F741S2B7LDmYAi8eyA@mail.gmail.com
Too allow table accesses to be not directly dependent on heap, several
new abstractions are needed. Specifically:
1) Heap scans need to be generalized into table scans. Do this by
introducing TableScanDesc, which will be the "base class" for
individual AMs. This contains the AM independent fields from
HeapScanDesc.
The previous heap_{beginscan,rescan,endscan} et al. have been
replaced with a table_ version.
There's no direct replacement for heap_getnext(), as that returned
a HeapTuple, which is undesirable for a other AMs. Instead there's
table_scan_getnextslot(). But note that heap_getnext() lives on,
it's still used widely to access catalog tables.
This is achieved by new scan_begin, scan_end, scan_rescan,
scan_getnextslot callbacks.
2) The portion of parallel scans that's shared between backends need
to be able to do so without the user doing per-AM work. To achieve
that new parallelscan_{estimate, initialize, reinitialize}
callbacks are introduced, which operate on a new
ParallelTableScanDesc, which again can be subclassed by AMs.
As it is likely that several AMs are going to be block oriented,
block oriented callbacks that can be shared between such AMs are
provided and used by heap. table_block_parallelscan_{estimate,
intiialize, reinitialize} as callbacks, and
table_block_parallelscan_{nextpage, init} for use in AMs. These
operate on a ParallelBlockTableScanDesc.
3) Index scans need to be able to access tables to return a tuple, and
there needs to be state across individual accesses to the heap to
store state like buffers. That's now handled by introducing a
sort-of-scan IndexFetchTable, which again is intended to be
subclassed by individual AMs (for heap IndexFetchHeap).
The relevant callbacks for an AM are index_fetch_{end, begin,
reset} to create the necessary state, and index_fetch_tuple to
retrieve an indexed tuple. Note that index_fetch_tuple
implementations need to be smarter than just blindly fetching the
tuples for AMs that have optimizations similar to heap's HOT - the
currently alive tuple in the update chain needs to be fetched if
appropriate.
Similar to table_scan_getnextslot(), it's undesirable to continue
to return HeapTuples. Thus index_fetch_heap (might want to rename
that later) now accepts a slot as an argument. Core code doesn't
have a lot of call sites performing index scans without going
through the systable_* API (in contrast to loads of heap_getnext
calls and working directly with HeapTuples).
Index scans now store the result of a search in
IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the
target is not generally a HeapTuple anymore that seems cleaner.
To be able to sensible adapt code to use the above, two further
callbacks have been introduced:
a) slot_callbacks returns a TupleTableSlotOps* suitable for creating
slots capable of holding a tuple of the AMs
type. table_slot_callbacks() and table_slot_create() are based
upon that, but have additional logic to deal with views, foreign
tables, etc.
While this change could have been done separately, nearly all the
call sites that needed to be adapted for the rest of this commit
also would have been needed to be adapted for
table_slot_callbacks(), making separation not worthwhile.
b) tuple_satisfies_snapshot checks whether the tuple in a slot is
currently visible according to a snapshot. That's required as a few
places now don't have a buffer + HeapTuple around, but a
slot (which in heap's case internally has that information).
Additionally a few infrastructure changes were needed:
I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now
internally uses a slot to keep track of tuples. While
systable_getnext() still returns HeapTuples, and will so for the
foreseeable future, the index API (see 1) above) now only deals with
slots.
The remainder, and largest part, of this commit is then adjusting all
scans in postgres to use the new APIs.
Author: Andres Freund, Haribabu Kommi, Alvaro Herrera
Discussion:
https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.dehttps://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
It seems to make more sense for this to be in selfuncs.c, since it's
largely a statistical-estimation thing, and it's related to other
functions like estimate_hash_bucket_stats that are there.
While at it, change the result type from Size to double. Perhaps at one
point it was impossible for the result to overflow an integer, but
I've got no confidence in that proposition anymore. Nothing's actually
done with the result except to compare it to a work_mem-based limit,
so as long as we don't get an overflow on the way to that comparison,
things should be fine even with very large dNumGroups.
Code movement proposed by Antonin Houska, type change by me
Discussion: https://postgr.es/m/25767.1549359615@localhost
Get rid of deconstruct_indexquals() in favor of just iterating over the
IndexClause list directly. The extra services that that function used to
provide, such as hiding clause commutation and associating the right index
column with each clause, are no longer useful given the new data structure.
I'd originally thought that it'd provide a useful amount of abstraction
by freeing callers from paying attention to the exact clause type of each
indexqual, but that hope proves to have been vain, because few callers can
ignore the semantic differences between different clause types. Indeed,
removing it results in a net code savings, and probably some cycles shaved
by not having to build an extra list-of-structs data structure.
Also, export a few formerly-static support functions, with the goal
of allowing extension AMs to write functionality equivalent to
genericcostestimate() without pointless code duplication.
Discussion: https://postgr.es/m/24586.1550106354@sss.pgh.pa.us
In commit 1a8d5afb0, I thought it'd be a good idea to define
IndexClause.indexquals as NIL in the most common case where the given
clause (IndexClause.rinfo) is usable exactly as-is. It'd be more
consistent to define the indexquals in that case as being a one-element
list containing IndexClause.rinfo, but I thought saving the palloc
overhead for making such a list would be worthwhile.
In hindsight, that was a great example of "premature optimization is the
root of all evil": it's complicated everyplace that needs to deal with
the indexquals, requiring duplicative code to handle both the simple
case and the not-simple case. I'd initially found that tolerable but
it's getting less so as I mop up some areas that I'd not touched in
1a8d5afb0. In any case, two more pallocs during a planner run are
surely at the noise level (a conclusion confirmed by a bit of
microbenchmarking). So let's change this decision before it becomes
set in stone, and insist that IndexClause.indexquals always be a valid
list of the actual index quals for the clause.
Discussion: https://postgr.es/m/24586.1550106354@sss.pgh.pa.us
While at it, refactor patternsel() a bit so that it can be used from
the LIKE/regex planner support functions as well. This makes the
planner able to deal equally well with either operator or function
syntax for these operations. I'm not excited about that as a feature
in itself, but it provides a nice model for extensions to follow if
they want such behavior for their operations.
This change localizes the use of pattern_fixed_prefix() and
make_greater_string() so that they no longer need be exported.
(We might get pushback from extensions about that, perhaps,
in which case I'd be inclined to re-export them in a new header
file like_support.h.)
This reduces the bulk of selfuncs.c a fair amount, removing ~1370
lines or about one-sixth of that file; it's still too big, but this
is progress.
Discussion: https://postgr.es/m/24537.1550093915@sss.pgh.pa.us
We're only going to consider key columns when creating indexquals,
so there is no point in having the outer loops in indxpath.c iterate
further than nkeycolumns.
Doing so in match_pathkeys_to_index() is actually wrong, and would have
caused crashes by now, except that we have no index AMs supporting both
amcanorderbyop and amcaninclude.
It's also wrong in relation_has_unique_index_for(). The effect there is
to fail to prove uniqueness even when the index does prove it, if there
are extra columns.
Also future-proof examine_variable() for the day when extra columns can
be expressions, and fix what's either a thinko or just an oversight in
btcostestimate(): we should consider the number of key columns, not the
total, when deciding whether to derate correlation.
None of these things seemed important enough to risk changing in a
just-before-wrap patch, but since we're past the release wrap window,
time to fix 'em.
Discussion: https://postgr.es/m/25526.1549847928@sss.pgh.pa.us
Add support function requests for estimating the selectivity, cost,
and number of result rows (if a SRF) of the target function.
The lack of a way to estimate selectivity of a boolean-returning
function in WHERE has been a recognized deficiency of the planner
since Berkeley days. This commit finally fixes it.
In addition, non-constant estimates of cost and number of output
rows are now possible. We still fall back to looking at procost
and prorows if the support function doesn't service the request,
of course.
To make concrete use of the possibility of estimating output rowcount
for SRFs, this commit adds support functions for array_unnest(anyarray)
and the integer variants of generate_series; the lack of plausible
rowcount estimates for those, even when it's obvious to a human,
has been a repeated subject of complaints. Obviously, much more
could now be done in this line, but I'm mostly just trying to get
the infrastructure in place.
Discussion: https://postgr.es/m/15193.1548028093@sss.pgh.pa.us
In place of three separate but interrelated lists (indexclauses,
indexquals, and indexqualcols), an IndexPath now has one list
"indexclauses" of IndexClause nodes. This holds basically the same
information as before, but in a more useful format: in particular, there
is now a clear connection between an indexclause (an original restriction
clause from WHERE or JOIN/ON) and the indexquals (directly usable index
conditions) derived from it.
We also change the ground rules a bit by mandating that clause commutation,
if needed, be done up-front so that what is stored in the indexquals list
is always directly usable as an index condition. This gets rid of repeated
re-determination of which side of the clause is the indexkey during costing
and plan generation, as well as repeated lookups of the commutator
operator. To minimize the added up-front cost, the typical case of
commuting a plain OpExpr is handled by a new special-purpose function
commute_restrictinfo(). For RowCompareExprs, generating the new clause
properly commuted to begin with is not really any more complex than before,
it's just different --- and we can save doing that work twice, as the
pretty-klugy original implementation did.
Tracking the connection between original and derived clauses lets us
also track explicitly whether the derived clauses are an exact or lossy
translation of the original. This provides a cheap solution to getting
rid of unnecessary rechecks of boolean index clauses, which previously
seemed like it'd be more expensive than it was worth.
Another pleasant (IMO) side-effect is that EXPLAIN now always shows
index clauses with the indexkey on the left; this seems less confusing.
This commit leaves expand_indexqual_conditions() and some related
functions in a slightly messy state. I didn't bother to change them
any more than minimally necessary to work with the new data structure,
because all that code is going to be refactored out of existence in
a follow-on patch.
Discussion: https://postgr.es/m/22182.1549124950@sss.pgh.pa.us
Create a new header optimizer/optimizer.h, which exposes just the
planner functions that can be used "at arm's length", without need
to access Paths or the other planner-internal data structures defined
in nodes/relation.h. This is intended to provide the whole planner
API seen by most of the rest of the system; although FDWs still need
to use additional stuff, and more thought is also needed about just
what selfuncs.c should rely on.
The main point of doing this now is to limit the amount of new
#include baggage that will be needed by "planner support functions",
which I expect to introduce later, and which will be in relevant
datatype modules rather than anywhere near the planner.
This commit just moves relevant declarations into optimizer.h from
other header files (a couple of which go away because everything
got moved), and adjusts #include lists to match. There's further
cleanup that could be done if we want to decide that some stuff
being exposed by optimizer.h doesn't belong in the planner at all,
but I'll leave that for another day.
Discussion: https://postgr.es/m/11460.1548706639@sss.pgh.pa.us
While it's perhaps unlikely that users would write an explicit test
like "ctid IS NULL", this function is also used in range estimation,
and an incorrect answer can throw off the results for tight ranges.
Anyway it's not much code so we might as well do it.
Edmund Horner
Discussion: https://postgr.es/m/CAMyN-kCa3BFUFrCTtQeprxTU1anCd3Pua7zXstGCKq4pXgjukw@mail.gmail.com
The code in tqual.c is largely heap specific. Due to the upcoming
pluggable storage work, it therefore makes sense to move it into
access/heap/ (as the file's header notes, the tqual name isn't very
good).
But the various statically allocated snapshot and snapshot
initialization functions are now (see previous commit) generic and do
not depend on functions declared in tqual.h anymore. Therefore move.
Also move XidInMVCCSnapshot as that's useful for future AMs, and
already used outside of tqual.c.
Author: Andres Freund
Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
heapam.h previously was included in a number of widely used
headers (e.g. execnodes.h, indirectly in executor.h, ...). That's
problematic on its own, as heapam.h contains a lot of low-level
details that don't need to be exposed that widely, but becomes more
problematic with the upcoming introduction of pluggable table storage
- it seems inappropriate for heapam.h to be included that widely
afterwards.
heapam.h was largely only included in other headers to get the
HeapScanDesc typedef (which was defined in heapam.h, even though
HeapScanDescData is defined in relscan.h). The better solution here
seems to be to just use the underlying struct (forward declared where
necessary). Similar for BulkInsertState.
Another problem was that LockTupleMode was used in executor.h - parts
of the file tried to cope without heapam.h, but due to the fact that
it indirectly included it, several subsequent violations of that goal
were not not noticed. We could just reuse the approach of declaring
parameters as int, but it seems nicer to move LockTupleMode to
lockoptions.h - that's not a perfect location, but also doesn't seem
bad.
As a number of files relied on implicitly included heapam.h, a
significant number of files grew an explicit include. It's quite
probably that a few external projects will need to do the same.
Author: Andres Freund
Reviewed-By: Alvaro Herrera
Discussion: https://postgr.es/m/20190114000701.y4ttcb74jpskkcfb@alap3.anarazel.de
Now that name comparison has effectively the same behavior as text
comparison, we might as well merge the name_ops opfamily into text_ops,
allowing cross-type comparisons to be processed without forcing a
datatype coercion first. We need do little more than add cross-type
operators to make the opfamily complete, and fix one or two places
in the planner that assumed text_ops was a single-datatype opfamily.
I chose to unify hash name_ops into hash text_ops as well, since the
types have compatible hashing semantics. This allows marking the
new cross-type equality operators as oprcanhash.
(Note: this doesn't remove the name_ops opclasses, so there's no
breakage of index definitions. Those opclasses are just reparented
into the text_ops opfamily.)
Discussion: https://postgr.es/m/15938.1544377821@sss.pgh.pa.us
The "name" comparison operators now all support collations, making them
functionally equivalent to "text" comparisons, except for the different
physical representation of the datatype. They do, in fact, mostly share
the varstr_cmp and varstr_sortsupport infrastructure, which has been
slightly enlarged to handle the case.
To avoid changes in the default behavior of the datatype, set name's
typcollation to C_COLLATION_OID not DEFAULT_COLLATION_OID, so that
by default comparisons to a name value will continue to use strcmp
semantics. (This would have been the case for system catalog columns
anyway, because of commit 6b0faf723, but doing this makes it true for
user-created name columns as well. In particular, this avoids
locale-dependent changes in our regression test results.)
In consequence, tweak a couple of places that made assumptions about
collatable base types always having typcollation DEFAULT_COLLATION_OID.
I have not, however, attempted to relax the restriction that user-
defined collatable types must have that. Hence, "name" doesn't
behave quite like a user-defined type; it acts more like a domain
with COLLATE "C". (Conceivably, if we ever get rid of the need for
catalog name columns to be fixed-length, "name" could actually become
such a domain over text. But that'd be a pretty massive undertaking,
and I'm not volunteering.)
Discussion: https://postgr.es/m/15938.1544377821@sss.pgh.pa.us
When we first put in collations support, we basically punted on teaching
pg_statistic, ANALYZE, and the planner selectivity functions about that.
They've just used DEFAULT_COLLATION_OID independently of the actual
collation of the data. It's time to improve that, so:
* Add columns to pg_statistic that record the specific collation associated
with each statistics slot.
* Teach ANALYZE to use the column's actual collation when comparing values
for statistical purposes, and record this in the appropriate slot. (Note
that type-specific typanalyze functions are now expected to fill
stats->stacoll with the appropriate collation, too.)
* Teach assorted selectivity functions to use the actual collation of
the stats they are looking at, instead of just assuming it's
DEFAULT_COLLATION_OID.
This should give noticeably better results in selectivity estimates for
columns with nondefault collations, at least for query clauses that use
that same collation (which would be the default behavior in most cases).
It's still true that comparisons with explicit COLLATE clauses different
from the stored data's collation won't be well-estimated, but that's no
worse than before. Also, this patch does make the first step towards
doing better with that, which is that it's now theoretically possible to
collect stats for a collation other than the column's own collation.
Patch by me; thanks to Peter Eisentraut for review.
Discussion: https://postgr.es/m/14706.1544630227@sss.pgh.pa.us
We should never estimate the output of a semijoin to be more rows than
we estimate for an inner join with the same input rels and join condition;
it's obviously impossible for that to happen. However, given the
relatively poor quality of our semijoin selectivity estimates ---
particularly, but not only, in cases where we punt and return a default
estimate --- we did often deliver such estimates. To improve matters,
calculate both estimates inside eqjoinsel() and take the smaller one.
The bulk of this patch is just mechanical refactoring to avoid repetitive
information lookup when we call both eqjoinsel_semi and eqjoinsel_inner.
The actual new behavior is just
selec = Min(selec, inner_rel->rows * selec_inner);
which looks a bit odd but is correct because of our different definitions
for inner and semi join selectivity.
There is one ensuing plan change in the regression tests, but it looks
reasonable enough (and checking the actual row counts shows that the
estimate moved closer to reality, not further away).
Per bug #15160 from Alexey Ermakov. Although this is arguably a bug fix,
I won't risk destabilizing plan choices in stable branches by
back-patching.
Tom Lane, reviewed by Melanie Plageman
Discussion: https://postgr.es/m/152395805004.19366.3107109716821067806@wrigleys.postgresql.org
Previously tables declared WITH OIDS, including a significant fraction
of the catalog tables, stored the oid column not as a normal column,
but as part of the tuple header.
This special column was not shown by default, which was somewhat odd,
as it's often (consider e.g. pg_class.oid) one of the more important
parts of a row. Neither pg_dump nor COPY included the contents of the
oid column by default.
The fact that the oid column was not an ordinary column necessitated a
significant amount of special case code to support oid columns. That
already was painful for the existing, but upcoming work aiming to make
table storage pluggable, would have required expanding and duplicating
that "specialness" significantly.
WITH OIDS has been deprecated since 2005 (commit ff02d0a05280e0).
Remove it.
Removing includes:
- CREATE TABLE and ALTER TABLE syntax for declaring the table to be
WITH OIDS has been removed (WITH (oids[ = true]) will error out)
- pg_dump does not support dumping tables declared WITH OIDS and will
issue a warning when dumping one (and ignore the oid column).
- restoring an pg_dump archive with pg_restore will warn when
restoring a table with oid contents (and ignore the oid column)
- COPY will refuse to load binary dump that includes oids.
- pg_upgrade will error out when encountering tables declared WITH
OIDS, they have to be altered to remove the oid column first.
- Functionality to access the oid of the last inserted row (like
plpgsql's RESULT_OID, spi's SPI_lastoid, ...) has been removed.
The syntax for declaring a table WITHOUT OIDS (or WITH (oids = false)
for CREATE TABLE) is still supported. While that requires a bit of
support code, it seems unnecessary to break applications / dumps that
do not use oids, and are explicit about not using them.
The biggest user of WITH OID columns was postgres' catalog. This
commit changes all 'magic' oid columns to be columns that are normally
declared and stored. To reduce unnecessary query breakage all the
newly added columns are still named 'oid', even if a table's column
naming scheme would indicate 'reloid' or such. This obviously
requires adapting a lot code, mostly replacing oid access via
HeapTupleGetOid() with access to the underlying Form_pg_*->oid column.
The bootstrap process now assigns oids for all oid columns in
genbki.pl that do not have an explicit value (starting at the largest
oid previously used), only oids assigned later by oids will be above
FirstBootstrapObjectId. As the oid column now is a normal column the
special bootstrap syntax for oids has been removed.
Oids are not automatically assigned during insertion anymore, all
backend code explicitly assigns oids with GetNewOidWithIndex(). For
the rare case that insertions into the catalog via SQL are called for
the new pg_nextoid() function can be used (which only works on catalog
tables).
The fact that oid columns on system tables are now normal columns
means that they will be included in the set of columns expanded
by * (i.e. SELECT * FROM pg_class will now include the table's oid,
previously it did not). It'd not technically be hard to hide oid
column by default, but that'd mean confusing behavior would either
have to be carried forward forever, or it'd cause breakage down the
line.
While it's not unlikely that further adjustments are needed, the
scope/invasiveness of the patch makes it worthwhile to get merge this
now. It's painful to maintain externally, too complicated to commit
after the code code freeze, and a dependency of a number of other
patches.
Catversion bump, for obvious reasons.
Author: Andres Freund, with contributions by John Naylor
Discussion: https://postgr.es/m/20180930034810.ywp2c7awz7opzcfr@alap3.anarazel.de
Upcoming work intends to allow pluggable ways to introduce new ways of
storing table data. Accessing those table access methods from the
executor requires TupleTableSlots to be carry tuples in the native
format of such storage methods; otherwise there'll be a significant
conversion overhead.
Different access methods will require different data to store tuples
efficiently (just like virtual, minimal, heap already require fields
in TupleTableSlot). To allow that without requiring additional pointer
indirections, we want to have different structs (embedding
TupleTableSlot) for different types of slots. Thus different types of
slots are needed, which requires adapting creators of slots.
The slot that most efficiently can represent a type of tuple in an
executor node will often depend on the type of slot a child node
uses. Therefore we need to track the type of slot is returned by
nodes, so parent slots can create slots based on that.
Relatedly, JIT compilation of tuple deforming needs to know which type
of slot a certain expression refers to, so it can create an
appropriate deforming function for the type of tuple in the slot.
But not all nodes will only return one type of slot, e.g. an append
node will potentially return different types of slots for each of its
subplans.
Therefore add function that allows to query the type of a node's
result slot, and whether it'll always be the same type (whether it's
fixed). This can be queried using ExecGetResultSlotOps().
The scan, result, inner, outer type of slots are automatically
inferred from ExecInitScanTupleSlot(), ExecInitResultSlot(),
left/right subtrees respectively. If that's not correct for a node,
that can be overwritten using new fields in PlanState.
This commit does not introduce the actually abstracted implementation
of different kind of TupleTableSlots, that will be left for a followup
commit. The different types of slots introduced will, for now, still
use the same backing implementation.
While this already partially invalidates the big comment in
tuptable.h, it seems to make more sense to update it later, when the
different TupleTableSlot implementations actually exist.
Author: Ashutosh Bapat and Andres Freund, with changes by Amit Khandekar
Discussion: https://postgr.es/m/20181105210039.hh4vvi4vwoq5ba2q@alap3.anarazel.de
Upcoming changes introduce further types of tuple table slots, in
preparation of making table storage pluggable. New storage methods
will have different representation of tuples, therefore the slot
accessor should refer explicitly to heap tuples.
Instead of just renaming the functions, split it into one function
that accepts heap tuples not residing in buffers, and one accepting
ones in buffers. Previously one function was used for both, but that
was a bit awkward already, and splitting will allow us to represent
slot types for tuples in buffers and normal memory separately.
This is split out from the patch introducing abstract slots, as this
largely consists out of mechanical changes.
Author: Ashutosh Bapat
Reviewed-By: Andres Freund
Discussion: https://postgr.es/m/20180220224318.gw4oe5jadhpmcdnm@alap3.anarazel.de
We include <float.h> in every place that needs isnan(), because MSVC
used to require it. However, since MSVC 2013 that's no longer necessary
(cf. commit cec8394b5c), so we can retire the inclusion to a
version-specific stanza in win32_port.h, where it doesn't need to
pollute random .c files. The header is of course still needed in a few
places for other reasons.
I (Álvaro) removed float.h from a few more files than in Emre's original
patch. This doesn't break the build in my system, but we'll see what
the buildfarm has to say about it all.
Author: Emre Hasegeli
Discussion: https://postgr.es/m/CAE2gYzyc0+5uG+Cd9-BSL7NKC8LSHLNg1Aq2=8ubjnUwut4_iw@mail.gmail.com
- Explicitly forbids opclass, collation and indoptions (like DESC/ASC etc) for
including columns. Throw an error if user points that.
- Truncated storage arrays for such attributes to store only key atrributes,
added assertion checks.
- Do not check opfamily and collation for including columns in
CompareIndexInfo()
Discussion: https://www.postgresql.org/message-id/5ee72852-3c4e-ee35-e2ed-c1d053d45c08@sigaev.ru
This patch introduces INCLUDE clause to index definition. This clause
specifies a list of columns which will be included as a non-key part in
the index. The INCLUDE columns exist solely to allow more queries to
benefit from index-only scans. Also, such columns don't need to have
appropriate operator classes. Expressions are not supported as INCLUDE
columns since they cannot be used in index-only scans.
Index access methods supporting INCLUDE are indicated by amcaninclude flag
in IndexAmRoutine. For now, only B-tree indexes support INCLUDE clause.
In B-tree indexes INCLUDE columns are truncated from pivot index tuples
(tuples located in non-leaf pages and high keys). Therefore, B-tree indexes
now might have variable number of attributes. This patch also provides
generic facility to support that: pivot tuples contain number of their
attributes in t_tid.ip_posid. Free 13th bit of t_info is used for indicating
that. This facility will simplify further support of index suffix truncation.
The changes of above are backward-compatible, pg_upgrade doesn't need special
handling of B-tree indexes for that.
Bump catalog version
Author: Anastasia Lubennikova with contribition by Alexander Korotkov and me
Reviewed by: Peter Geoghegan, Tomas Vondra, Antonin Houska, Jeff Janes,
David Rowley, Alexander Korotkov
Discussion: https://www.postgresql.org/message-id/flat/56168952.4010101@postgrespro.ru
The prefix operator along with SP-GiST indexes can be used as an alternative
for LIKE 'word%' commands and it doesn't have a limitation of string/prefix
length as B-Tree has.
Bump catalog version
Author: Ildus Kurbangaliev with some editorization by me
Review by: Arthur Zakirov, Alexander Korotkov, and me
Discussion: https://www.postgresql.org/message-id/flat/20180202180327.222b04b3@wp.localdomain
If convert_to_scalar is passed a pair of datatypes it can't cope with,
its former behavior was just to elog(ERROR). While this is OK so far as
the core code is concerned, there's extension code that would like to use
scalarltsel/scalargtsel/etc as selectivity estimators for operators that
work on non-core datatypes, and this behavior is a show-stopper for that
use-case. If we simply allow convert_to_scalar to return FALSE instead of
outright failing, then the main logic of scalarltsel/scalargtsel will work
fine for any operator that behaves like a scalar inequality comparison.
The lack of conversion capability will mean that we can't estimate to
better than histogram-bin-width precision, since the code will effectively
assume that the comparison constant falls at the middle of its bin. But
that's still a lot better than nothing. (Someday we should provide a way
for extension code to supply a custom version of convert_to_scalar, but
today is not that day.)
While poking at this issue, we noted that the existing code for handling
type bytea in convert_to_scalar is several bricks shy of a load.
It assumes without checking that if the comparison value is type bytea,
the bounds values are too; in the worst case this could lead to a crash.
It also fails to detoast the input values, so that the comparison result is
complete garbage if any input is toasted out-of-line, compressed, or even
just short-header. I'm not sure how often such cases actually occur ---
the bounds values, at least, are probably safe since they are elements of
an array and hence can't be toasted. But that doesn't make this code OK.
Back-patch to all supported branches, partly because author requested that,
but mostly because of the bytea bugs. The change in API for the exposed
routine convert_network_to_scalar() is theoretically a back-patch hazard,
but it seems pretty unlikely that any third-party code is calling that
function directly.
Tomas Vondra, with some adjustments by me
Discussion: https://postgr.es/m/b68441b6-d18f-13ab-b43b-9a72188a4e02@2ndquadrant.com
Previously, this function estimated the selectivity as 1 minus eqjoinsel()
for the negator equality operator, regardless of join type (I think there
was an expectation that eqjoinsel would handle the join type). But
actually this is completely wrong for semijoin cases: the fraction of the
LHS that has a non-matching row is not one minus the fraction of the LHS
that has a matching row. In reality a semijoin with <> will nearly always
succeed: it can only fail when the RHS is empty, or it contains a single
distinct value that is equal to the particular LHS value, or the LHS value
is null. The only one of those things we should have much confidence in
estimating is the fraction of LHS values that are null, so let's just take
the selectivity as 1 minus outer nullfrac.
Per coding convention, antijoin should be estimated the same as semijoin.
Arguably this is a bug fix, but in view of the lack of field complaints
and the risk of destabilizing plans, no back-patch.
Thomas Munro, reviewed by Ashutosh Bapat
Discussion: https://postgr.es/m/CAEepm=270ze2hVxWkJw-5eKzc3AB4C9KpH3L2kih75R5pdSogg@mail.gmail.com
Improve query_is_distinct_for() to accept SRFs in the targetlist when
we can prove distinctness from a DISTINCT clause. In that case the
de-duplication will surely happen after SRF expansion, so the proof
still works. Continue to punt in the case where we'd try to prove
distinctness from GROUP BY (or, in the future, source relations).
To do that, we'd have to determine whether the SRFs were in the
grouping columns or elsewhere in the tlist, and it still doesn't
seem worth the trouble. But this trivial change allows us to
recognize that "SELECT DISTINCT unnest(foo) FROM ..." produces
unique-ified output, which seems worth having.
Also, fix estimate_num_groups() to consider the possibility of SRFs in
the grouping columns. Its failure to do so was masked before v10 because
grouping_planner() scaled up plan rowcount estimates by the estimated SRF
multiplier after performing grouping. That doesn't happen anymore, which
is more correct, but it means we need an adjustment in the estimate for
the number of groups. Failure to do this leads to an underestimate for
the number of output rows of subqueries like "SELECT DISTINCT unnest(foo)"
compared to what 9.6 and earlier estimated, thus breaking plan choices
in some cases.
Per report from Dmitry Shalashov. Back-patch to v10 to avoid degraded
plan choices compared to previous releases.
Discussion: https://postgr.es/m/CAKPeCUGAeHgoh5O=SvcQxREVkoX7UdeJUMj1F5=aBNvoTa+O8w@mail.gmail.com
The lower case spellings are C and C++ standard and are used in most
parts of the PostgreSQL sources. The upper case spellings are only used
in some files/modules. So standardize on the standard spellings.
The APIs for ICU, Perl, and Windows define their own TRUE and FALSE, so
those are left as is when using those APIs.
In code comments, we use the lower-case spelling for the C concepts and
keep the upper-case spelling for the SQL concepts.
Reviewed-by: Michael Paquier <michael.paquier@gmail.com>
Allowing arrays with a domain type as their element type was left un-done
in the original domain patch, but not for any very good reason. This
omission leads to such surprising results as array_agg() not working on
a domain column, because the parser can't identify a suitable output type
for the polymorphic aggregate.
In order to fix this, first clean up the APIs of coerce_to_domain() and
some internal functions in parse_coerce.c so that we consistently pass
around a CoercionContext along with CoercionForm. Previously, we sometimes
passed an "isExplicit" boolean flag instead, which is strictly less
information; and coerce_to_domain() didn't even get that, but instead had
to reverse-engineer isExplicit from CoercionForm. That's contrary to the
documentation in primnodes.h that says that CoercionForm only affects
display and not semantics. I don't think this change fixes any live bugs,
but it makes things more consistent. The main reason for doing it though
is that now build_coercion_expression() receives ccontext, which it needs
in order to be able to recursively invoke coerce_to_target_type().
Next, reimplement ArrayCoerceExpr so that the node does not directly know
any details of what has to be done to the individual array elements while
performing the array coercion. Instead, the per-element processing is
represented by a sub-expression whose input is a source array element and
whose output is a target array element. This simplifies life in
parse_coerce.c, because it can build that sub-expression by a recursive
invocation of coerce_to_target_type(). The executor now handles the
per-element processing as a compiled expression instead of hard-wired code.
The main advantage of this is that we can use a single ArrayCoerceExpr to
handle as many as three successive steps per element: base type conversion,
typmod coercion, and domain constraint checking. The old code used two
stacked ArrayCoerceExprs to handle type + typmod coercion, which was pretty
inefficient, and adding yet another array deconstruction to do domain
constraint checking seemed very unappetizing.
In the case where we just need a single, very simple coercion function,
doing this straightforwardly leads to a noticeable increase in the
per-array-element runtime cost. Hence, add an additional shortcut evalfunc
in execExprInterp.c that skips unnecessary overhead for that specific form
of expression. The runtime speed of simple cases is within 1% or so of
where it was before, while cases that previously required two levels of
array processing are significantly faster.
Finally, create an implicit array type for every domain type, as we do for
base types, enums, etc. Everything except the array-coercion case seems
to just work without further effort.
Tom Lane, reviewed by Andrew Dunstan
Discussion: https://postgr.es/m/9852.1499791473@sss.pgh.pa.us
Historically, the selectivity functions have simply not distinguished
< from <=, or > from >=, arguing that the fraction of the population that
satisfies the "=" aspect can be considered to be vanishingly small, if the
comparison value isn't any of the most-common-values for the variable.
(If it is, the code path that executes the operator against each MCV will
take care of things properly.) But that isn't really true unless we're
dealing with a continuum of variable values, and in practice we seldom are.
If "x = const" would estimate a nonzero number of rows for a given const
value, then it follows that we ought to estimate different numbers of rows
for "x < const" and "x <= const", even if the const is not one of the MCVs.
Handling this more honestly makes a significant difference in edge cases,
such as the estimate for a tight range (x BETWEEN y AND z where y and z
are close together).
Hence, split scalarltsel into scalarltsel/scalarlesel, and similarly
split scalargtsel into scalargtsel/scalargesel. Adjust <= and >=
operator definitions to reference the new selectivity functions.
Improve the core ineq_histogram_selectivity() function to make a
correction for equality. (Along the way, I learned quite a bit about
exactly why that function gives good answers, which I tried to memorialize
in improved comments.)
The corresponding join selectivity functions were, and remain, just stubs.
But I chose to split them similarly, to avoid confusion and to prevent the
need for doing this exercise again if someone ever makes them less stubby.
In passing, change ineq_histogram_selectivity's clamp for extreme
probability estimates so that it varies depending on the histogram
size, instead of being hardwired at 0.0001. With the default histogram
size of 100 entries, you still get the old clamp value, but bigger
histograms should allow us to put more faith in edge values.
Tom Lane, reviewed by Aleksander Alekseev and Kuntal Ghosh
Discussion: https://postgr.es/m/12232.1499140410@sss.pgh.pa.us
In commit fccebe421, we hacked get_actual_variable_range() to scan the
index with SnapshotDirty, so that if there are many uncommitted tuples
at the end of the index range, it wouldn't laboriously scan through all
of them looking for a live value to return. However, that didn't fix it
for the case of many recently-dead tuples at the end of the index;
SnapshotDirty recognizes those as committed dead and so we're back to
the same problem.
To improve the situation, invent a "SnapshotNonVacuumable" snapshot type
and use that instead. The reason this helps is that, if the snapshot
rejects a given index entry, we know that the indexscan will mark that
index entry as killed. This means the next get_actual_variable_range()
scan will proceed past that entry without visiting the heap, making the
scan a lot faster. We may end up accepting a recently-dead tuple as
being the estimated extremal value, but that doesn't seem much worse than
the compromise we made before to accept not-yet-committed extremal values.
The cost of the scan is still proportional to the number of dead index
entries at the end of the range, so in the interval after a mass delete
but before VACUUM's cleaned up the mess, it's still possible for
get_actual_variable_range() to take a noticeable amount of time, if you've
got enough such dead entries. But the constant factor is much much better
than before, since all we need to do with each index entry is test its
"killed" bit.
We chose to back-patch commit fccebe421 at the time, but I'm hesitant to
do so here, because this form of the problem seems to affect many fewer
people. Also, even when it happens, it's less bad than the case fixed
by commit fccebe421 because we don't get the contention effects from
expensive TransactionIdIsInProgress tests.
Dmitriy Sarafannikov, reviewed by Andrey Borodin
Discussion: https://postgr.es/m/05C72CF7-B5F6-4DB9-8A09-5AC897653113@yandex.ru
Previously, if we had to estimate the number of distinct values in a
VALUES column, we fell back on the default behavior used whenever we lack
statistics, which effectively is that there are Min(# of entries, 200)
distinct values. This can be very badly off with a large VALUES list,
as noted by Jeff Janes.
We could consider actually running an ANALYZE-like scan on the VALUES,
but that seems unduly expensive, and anyway it could not deliver reliable
info if the entries are not all constants. What seems like a better choice
is to assume that the values are all distinct. This will sometimes be just
as wrong as the old code, but it seems more likely to be more nearly right
in many common cases. Also, it is more consistent with what happens in
some related cases, for example WHERE x = ANY(ARRAY[1,2,3,...,n]) and
WHERE x = ANY(VALUES (1),(2),(3),...,(n)) now are estimated similarly.
This was discussed some time ago, but consensus was it'd be better
to slip it in at the start of a development cycle not near the end.
(It should've gone into v10, really, but I forgot about it.)
Discussion: https://postgr.es/m/CAMkU=1xHkyPa8VQgGcCNg3RMFFvVxUdOpus1gKcFuvVi0w6Acg@mail.gmail.com
The executor is capable of splitting buckets during a hash join if
too much memory is being used by a small number of buckets. However,
this only helps if a bucket's population is actually divisible; if
all the hash keys are alike, the tuples still end up in the same
new bucket. This can result in an OOM failure if there are enough
inner keys with identical hash values. The planner's cost estimates
will bias it against choosing a hash join in such situations, but not
by so much that it will never do so. To mitigate the OOM hazard,
explicitly estimate the hash bucket space needed by just the inner
side's most common value, and if that would exceed work_mem then
add disable_cost to the hash cost estimate.
This approach doesn't account for the possibility that two or more
common values would share the same hash value. On the other hand,
work_mem is normally a fairly conservative bound, so that eating
two or more times that much space is probably not going to kill us.
If we have no stats about the inner side, ignore this consideration.
There was some discussion of making a conservative assumption, but that
would effectively result in disabling hash join whenever we lack stats,
which seems like an overreaction given how seldom the problem manifests
in the field.
Per a complaint from David Hinkle. Although this could be viewed
as a bug fix, the lack of similar complaints weighs against back-
patching; indeed we waited for v11 because it seemed already rather
late in the v10 cycle to be making plan choice changes like this one.
Discussion: https://postgr.es/m/32013.1487271761@sss.pgh.pa.us
Don't move parenthesized lines to the left, even if that means they
flow past the right margin.
By default, BSD indent lines up statement continuation lines that are
within parentheses so that they start just to the right of the preceding
left parenthesis. However, traditionally, if that resulted in the
continuation line extending to the right of the desired right margin,
then indent would push it left just far enough to not overrun the margin,
if it could do so without making the continuation line start to the left of
the current statement indent. That makes for a weird mix of indentations
unless one has been completely rigid about never violating the 80-column
limit.
This behavior has been pretty universally panned by Postgres developers.
Hence, disable it with indent's new -lpl switch, so that parenthesized
lines are always lined up with the preceding left paren.
This patch is much less interesting than the first round of indent
changes, but also bulkier, so I thought it best to separate the effects.
Discussion: https://postgr.es/m/E1dAmxK-0006EE-1r@gemulon.postgresql.org
Discussion: https://postgr.es/m/30527.1495162840@sss.pgh.pa.us
Change pg_bsd_indent to follow upstream rules for placement of comments
to the right of code, and remove pgindent hack that caused comments
following #endif to not obey the general rule.
Commit e3860ffa4d wasn't actually using
the published version of pg_bsd_indent, but a hacked-up version that
tried to minimize the amount of movement of comments to the right of
code. The situation of interest is where such a comment has to be
moved to the right of its default placement at column 33 because there's
code there. BSD indent has always moved right in units of tab stops
in such cases --- but in the previous incarnation, indent was working
in 8-space tab stops, while now it knows we use 4-space tabs. So the
net result is that in about half the cases, such comments are placed
one tab stop left of before. This is better all around: it leaves
more room on the line for comment text, and it means that in such
cases the comment uniformly starts at the next 4-space tab stop after
the code, rather than sometimes one and sometimes two tabs after.
Also, ensure that comments following #endif are indented the same
as comments following other preprocessor commands such as #else.
That inconsistency turns out to have been self-inflicted damage
from a poorly-thought-through post-indent "fixup" in pgindent.
This patch is much less interesting than the first round of indent
changes, but also bulkier, so I thought it best to separate the effects.
Discussion: https://postgr.es/m/E1dAmxK-0006EE-1r@gemulon.postgresql.org
Discussion: https://postgr.es/m/30527.1495162840@sss.pgh.pa.us
The new indent version includes numerous fixes thanks to Piotr Stefaniak.
The main changes visible in this commit are:
* Nicer formatting of function-pointer declarations.
* No longer unexpectedly removes spaces in expressions using casts,
sizeof, or offsetof.
* No longer wants to add a space in "struct structname *varname", as
well as some similar cases for const- or volatile-qualified pointers.
* Declarations using PG_USED_FOR_ASSERTS_ONLY are formatted more nicely.
* Fixes bug where comments following declarations were sometimes placed
with no space separating them from the code.
* Fixes some odd decisions for comments following case labels.
* Fixes some cases where comments following code were indented to less
than the expected column 33.
On the less good side, it now tends to put more whitespace around typedef
names that are not listed in typedefs.list. This might encourage us to
put more effort into typedef name collection; it's not really a bug in
indent itself.
There are more changes coming after this round, having to do with comment
indentation and alignment of lines appearing within parentheses. I wanted
to limit the size of the diffs to something that could be reviewed without
one's eyes completely glazing over, so it seemed better to split up the
changes as much as practical.
Discussion: https://postgr.es/m/E1dAmxK-0006EE-1r@gemulon.postgresql.org
Discussion: https://postgr.es/m/30527.1495162840@sss.pgh.pa.us
In a CHECK clause, a null result means true, whereas in a WHERE clause
it means false. predtest.c provided different functions depending on
which set of semantics applied to the predicate being proved, but had
no option to control what a null meant in the clauses provided as
axioms. Add one.
Use that in the partitioning code when figuring out whether the
validation scan on a new partition can be skipped. Rip out the
old logic that attempted (not very successfully) to compensate
for the absence of the necessary support in predtest.c.
Ashutosh Bapat and Robert Haas, reviewed by Amit Langote and
incorporating feedback from Tom Lane.
Discussion: http://postgr.es/m/CAFjFpReT_kq_uwU_B8aWDxR7jNGE=P0iELycdq5oupi=xSQTOw@mail.gmail.com
These estimators returned 1 minus the corresponding equality/match
estimate, which is incorrect: we need to subtract off the fraction
of nulls in the column, since those are neither equal nor not equal
to the comparison value. The error only becomes obvious if the
nullfrac is large, but it could be very bad in a mostly-nulls
column, as reported in bug #14676 from Marko Tiikkaja.
To fix the <> case, refactor eqsel() and neqsel() to call a common
support routine, which can be made to account for nullfrac correctly.
The pattern-match cases were already factored that way, and it was
simply an oversight that patternsel() wasn't subtracting off nullfrac.
neqjoinsel() has a similar problem, but since we're elsewhere discussing
changing its behavior entirely, I left it alone for now.
This is a very longstanding bug, but I'm hesitant to back-patch a fix for
it. Given the lack of prior complaints, such cases must not come up often,
so it's probably not worth the risk of destabilizing plans in stable
branches.
Discussion: https://postgr.es/m/20170529153847.4275.95416@wrigleys.postgresql.org
Consistently refer to such an entry as a "statistics object", not just
"statistics" or "extended statistics". Previously we had a mismash of
terms, accompanied by utter confusion as to whether the term was
singular or plural. That's not only grating (at least to the ear of
a native English speaker) but could be outright misleading, eg in error
messages that seemed to be referring to multiple objects where only one
could be meant.
This commit fixes the code and a lot of comments (though I may have
missed a few). I also renamed two new SQL functions,
pg_get_statisticsextdef -> pg_get_statisticsobjdef
pg_statistic_ext_is_visible -> pg_statistics_obj_is_visible
to conform better with this terminology.
I have not touched the SGML docs other than fixing those function
names; the docs certainly need work but it seems like a separable task.
Discussion: https://postgr.es/m/22676.1494557205@sss.pgh.pa.us
The mess cleaned up in commit da0759600 is clear evidence that it's a
bug hazard to expect the caller of get_attstatsslot()/free_attstatsslot()
to provide the correct type OID for the array elements in the slot.
Moreover, we weren't even getting any performance benefit from that,
since get_attstatsslot() was extracting the real type OID from the array
anyway. So we ought to get rid of that requirement; indeed, it would
make more sense for get_attstatsslot() to pass back the type OID it found,
in case the caller isn't sure what to expect, which is likely in binary-
compatible-operator cases.
Another problem with the current implementation is that if the stats array
element type is pass-by-reference, we incur a palloc/memcpy/pfree cycle
for each element. That seemed acceptable when the code was written because
we were targeting O(10) array sizes --- but these days, stats arrays are
almost always bigger than that, sometimes much bigger. We can save a
significant number of cycles by doing one palloc/memcpy/pfree of the whole
array. Indeed, in the now-probably-common case where the array is toasted,
that happens anyway so this method is basically free. (Note: although the
catcache code will inline any out-of-line toasted values, it doesn't
decompress them. At the other end of the size range, it doesn't expand
short-header datums either. In either case, DatumGetArrayTypeP would have
to make a copy. We do end up using an extra array copy step if the element
type is pass-by-value and the array length is neither small enough for a
short header nor large enough to have suffered compression. But that
seems like a very acceptable price for winning in pass-by-ref cases.)
Hence, redesign to take these insights into account. While at it,
convert to an API in which we fill a struct rather than passing a bunch
of pointers to individual output arguments. That will make it less
painful if we ever want further expansion of what get_attstatsslot can
pass back.
It's certainly arguable that this is new development and not something to
push post-feature-freeze. However, I view it as primarily bug-proofing
and therefore something that's better to have sooner not later. Since
we aren't quite at beta phase yet, let's put it in.
Discussion: https://postgr.es/m/16364.1494520862@sss.pgh.pa.us
Some selectivity estimation functions run user-supplied operators over
data obtained from pg_statistic without security checks, which allows
those operators to leak pg_statistic data without having privileges on
the underlying tables. Fix by checking that one of the following is
satisfied: (1) the user has table or column privileges on the table
underlying the pg_statistic data, or (2) the function implementing the
user-supplied operator is leak-proof. If neither is satisfied, planning
will proceed as if there are no statistics available.
At least one of these is satisfied in most cases in practice. The only
situations that are negatively impacted are user-defined or
not-leak-proof operators on a security-barrier view.
Reported-by: Robert Haas <robertmhaas@gmail.com>
Author: Peter Eisentraut <peter_e@gmx.net>
Author: Tom Lane <tgl@sss.pgh.pa.us>
Security: CVE-2017-7484
This extends the castNode() notation introduced by commit 5bcab1114 to
provide, in one step, extraction of a list cell's pointer and coercion to
a concrete node type. For example, "lfirst_node(Foo, lc)" is the same
as "castNode(Foo, lfirst(lc))". Almost half of the uses of castNode
that have appeared so far include a list extraction call, so this is
pretty widely useful, and it saves a few more keystrokes compared to the
old way.
As with the previous patch, back-patch the addition of these macros to
pg_list.h, so that the notation will be available when back-patching.
Patch by me, after an idea of Andrew Gierth's.
Discussion: https://postgr.es/m/14197.1491841216@sss.pgh.pa.us
The original code was overly optimistic about the cost of scanning a
BRIN index, leading to BRIN indexes being selected when they'd be a
worse choice than some other index. This complete rewrite should be
more accurate.
Author: David Rowley, based on an earlier patch by Emre Hasegeli
Reviewed-by: Emre Hasegeli
Discussion: https://postgr.es/m/CAKJS1f9n-Wapop5Xz1dtGdpdqmzeGqQK4sV2MK-zZugfC14Xtw@mail.gmail.com
Follow on patch in the multi-variate statistics patch series.
CREATE STATISTICS s1 WITH (dependencies) ON (a, b) FROM t;
ANALYZE;
will collect dependency stats on (a, b) and then use the measured
dependency in subsequent query planning.
Commit 7b504eb282 added
CREATE STATISTICS with n-distinct coefficients. These are now
specified using the mutually exclusive option WITH (ndistinct).
Author: Tomas Vondra, David Rowley
Reviewed-by: Kyotaro HORIGUCHI, Álvaro Herrera, Dean Rasheed, Robert Haas
and many other comments and contributions
Discussion: https://postgr.es/m/56f40b20-c464-fad2-ff39-06b668fac47c@2ndquadrant.com
Currently, the only type of child relation is an "other member rel",
which is the child of a baserel, but in the future joins and even
upper relations may have child rels. To facilitate that, introduce
macros that test to test for particular RelOptKind values, and use
them in various places where they help to clarify the sense of a test.
(For example, a test may allow RELOPT_OTHER_MEMBER_REL either because
it intends to allow child rels, or because it intends to allow simple
rels.)
Also, remove find_childrel_top_parent, which will not work for a
child rel that is not a baserel. Instead, add a new RelOptInfo
member top_parent_relids to track the same kind of information in a
more generic manner.
Ashutosh Bapat, slightly tweaked by me. Review and testing of the
patch set from which this was taken by Rajkumar Raghuwanshi and Rafia
Sabih.
Discussion: http://postgr.es/m/CA+TgmoagTnF2yqR3PT2rv=om=wJiZ4-A+ATwdnriTGku1CLYxA@mail.gmail.com
The code for the reworked n-distinct estimation on commit 7b504eb282 was
written differently in a previous version of the patch, prior to commit;
on rewriting it, we missed updating an initializer. This caused the
code to (mistakenly) apply a fudge factor even in the case where a
single value is applied, leading to incorrect results.
This means that the 'relvarcount' variable name is now wrong. Add a
comment to try and make the situation clearer, and remove an incorrect
comment I added.
Problem noticed, and code patch, by Tomas Vondra. Additional commentary
by Álvaro.
Add support for explicitly declared statistic objects (CREATE
STATISTICS), allowing collection of statistics on more complex
combinations that individual table columns. Companion commands DROP
STATISTICS and ALTER STATISTICS ... OWNER TO / SET SCHEMA / RENAME are
added too. All this DDL has been designed so that more statistic types
can be added later on, such as multivariate most-common-values and
multivariate histograms between columns of a single table, leaving room
for permitting columns on multiple tables, too, as well as expressions.
This commit only adds support for collection of n-distinct coefficient
on user-specified sets of columns in a single table. This is useful to
estimate number of distinct groups in GROUP BY and DISTINCT clauses;
estimation errors there can cause over-allocation of memory in hashed
aggregates, for instance, so it's a worthwhile problem to solve. A new
special pseudo-type pg_ndistinct is used.
(num-distinct estimation was deemed sufficiently useful by itself that
this is worthwhile even if no further statistic types are added
immediately; so much so that another version of essentially the same
functionality was submitted by Kyotaro Horiguchi:
https://postgr.es/m/20150828.173334.114731693.horiguchi.kyotaro@lab.ntt.co.jp
though this commit does not use that code.)
Author: Tomas Vondra. Some code rework by Álvaro.
Reviewed-by: Dean Rasheed, David Rowley, Kyotaro Horiguchi, Jeff Janes,
Ideriha Takeshi
Discussion: https://postgr.es/m/543AFA15.4080608@fuzzy.czhttps://postgr.es/m/20170320190220.ixlaueanxegqd5gr@alvherre.pgsql
Add a column collprovider to pg_collation that determines which library
provides the collation data. The existing choices are default and libc,
and this adds an icu choice, which uses the ICU4C library.
The pg_locale_t type is changed to a union that contains the
provider-specific locale handles. Users of locale information are
changed to look into that struct for the appropriate handle to use.
Also add a collversion column that records the version of the collation
when it is created, and check at run time whether it is still the same.
This detects potentially incompatible library upgrades that can corrupt
indexes and other structures. This is currently only supported by
ICU-provided collations.
initdb initializes the default collation set as before from the `locale
-a` output but also adds all available ICU locales with a "-x-icu"
appended.
Currently, ICU-provided collations can only be explicitly named
collations. The global database locales are still always libc-provided.
ICU support is enabled by configure --with-icu.
Reviewed-by: Thomas Munro <thomas.munro@enterprisedb.com>
Reviewed-by: Andreas Karlsson <andreas@proxel.se>
This adds in support for EUI-64 MAC addresses by adding a new data type
called 'macaddr8' (using our usual convention of indicating the number
of bytes stored).
This was largely a copy-and-paste from the macaddr data type, with
appropriate adjustments for having 8 bytes instead of 6 and adding
support for converting a provided EUI-48 (6 byte format) to the EUI-64
format. Conversion from EUI-48 to EUI-64 inserts FFFE as the 4th and
5th bytes but does not perform the IPv6 modified EUI-64 action of
flipping the 7th bit, but we add a function to perform that specific
action for the user as it may be commonly done by users who wish to
calculate their IPv6 address based on their network prefix and 48-bit
MAC address.
Author: Haribabu Kommi, with a good bit of rework of macaddr8_in by me.
Reviewed by: Vitaly Burovoy, Kuntal Ghosh
Discussion: https://postgr.es/m/CAJrrPGcUi8ZH+KkK+=TctNQ+EfkeCEHtMU_yo1mvX8hsk_ghNQ@mail.gmail.com
This makes almost all core code follow the policy introduced in the
previous commit. Specific decisions:
- Text search support functions with char* and length arguments, such as
prsstart and lexize, may receive unaligned strings. I doubt
maintainers of non-core text search code will notice.
- Use plain VARDATA() on values detoasted or synthesized earlier in the
same function. Use VARDATA_ANY() on varlenas sourced outside the
function, even if they happen to always have four-byte headers. As an
exception, retain the universal practice of using VARDATA() on return
values of SendFunctionCall().
- Retain PG_GETARG_BYTEA_P() in pageinspect. (Page images are too large
for a one-byte header, so this misses no optimization.) Sites that do
not call get_page_from_raw() typically need the four-byte alignment.
- For now, do not change btree_gist. Its use of four-byte headers in
memory is partly entangled with storage of 4-byte headers inside
GBT_VARKEY, on disk.
- For now, do not change gtrgm_consistent() or gtrgm_distance(). They
incorporate the varlena header into a cache, and there are multiple
credible implementation strategies to consider.
Further fallout from commit c29aff959: there are some files that need
<float.h>, and were getting it from datatype/timestamp.h, but it was not
apparent in my (tgl's) testing because the requirement for <float.h>
exists only on certain Windows toolchains.
Report and patch by David Rowley.
Discussion: https://postgr.es/m/CAKJS1f-BHceaFzZScFapDV48gUVM2CAOBfhkgffdqXzFb+kwew@mail.gmail.com
In combination with 569174f1be, 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.
If we don't have any MCV statistics for the inner relation, and we don't
trust its numdistinct estimate either, eqjoinsel_semi falls back to a very
conservative estimate (that 50% of the outer rows have matches). This is
particularly problematic if the inner relation is completely empty, since
then even an explicit ANALYZE won't produce any pg_statistic entries,
so there's no way to budge the planner off the bad estimate.
We'd produce a better estimate in such cases if we used the nd2/nd1
selectivity heuristic, so an easy fix is to treat the nd2 estimate as
non-default if we derive it from clamping to the inner rel's rowcount
estimate. This won't fix every related case (mainly because the rowcount
estimate might be larger than DEFAULT_NUM_DISTINCT), but it seems like a
sane extension of the existing logic, so let's apply the change in HEAD
and see if anyone complains. Per bug #14438 from Nikolay Nikitin.
Report: https://postgr.es/m/20161128182113.6527.58926@wrigleys.postgresql.org
Discussion: https://postgr.es/m/31089.1480384713@sss.pgh.pa.us
If ANALYZE found no repeated non-null entries in its sample, it set the
column's stadistinct value to -1.0, intending to indicate that the entries
are all distinct. But what this value actually means is that the number
of distinct values is 100% of the table's rowcount, and thus it was
overestimating the number of distinct values by however many nulls there
are. This could lead to very poor selectivity estimates, as for example
in a recent report from Andreas Joseph Krogh. We should discount the
stadistinct value by whatever we've estimated the nulls fraction to be.
(That is what will happen if we choose to use a negative stadistinct for
a column that does have repeated entries, so this code path was just
inconsistent.)
In addition to fixing the stadistinct entries stored by several different
ANALYZE code paths, adjust the logic where get_variable_numdistinct()
forces an "all distinct" estimate on the basis of finding a relevant unique
index. Unique indexes don't reject nulls, so there's no reason to assume
that the null fraction doesn't apply.
Back-patch to all supported branches. Back-patching is a bit of a judgment
call, but this problem seems to affect only a few users (else we'd have
identified it long ago), and it's bad enough when it does happen that
destabilizing plan choices in a worse direction seems unlikely.
Patch by me, with documentation wording suggested by Dean Rasheed
Report: <VisenaEmail.26.df42f82acae38a58.156463942b8@tc7-visena>
Discussion: <16143.1470350371@sss.pgh.pa.us>
Now indexes (but only B-tree for now) can contain "extra" column(s) which
doesn't participate in index structure, they are just stored in leaf
tuples. It allows to use index only scan by using single index instead
of two or more indexes.
Author: Anastasia Lubennikova with minor editorializing by me
Reviewers: David Rowley, Peter Geoghegan, Jeff Janes
When adjusting the estimate for the number of distinct values from a
rel in a grouped query to take into account the selectivity of the
rel's restrictions, use a formula that is less likely to produce
under-estimates.
The old formula simply multiplied the number of distinct values in the
rel by the restriction selectivity, which would be correct if the
restrictions were fully correlated with the grouping expressions, but
can produce significant under-estimates in cases where they are not
well correlated.
The new formula is based on the random selection probability, and so
assumes that the restrictions are not correlated with the grouping
expressions. This is guaranteed to produce larger estimates, and of
course risks over-estimating in cases where the restrictions are
correlated, but that has less severe consequences than
under-estimating, which might lead to a HashAgg that consumes an
excessive amount of memory.
This could possibly be improved upon in the future by identifying
correlated restrictions and using a hybrid of the old and new
formulae.
Author: Tomas Vondra, with some hacking be me
Reviewed-by: Mark Dilger, Alexander Korotkov, Dean Rasheed and Tom Lane
Discussion: http://www.postgresql.org/message-id/flat/56CD0381.5060502@2ndquadrant.com
If the referenced rel was proven empty, we'd compute 0/0 here, which
results in the function returning NaN. That's a bit more serious
than the other zero-divide case. Still, it only seems to be possible
in HEAD, so no back-patch.
Per report from Piotr Stefaniak. I looked through the rest of selfuncs.c
and found no other likely trouble spots.
This avoids a possible divide-by-zero in the following calculation,
and rounding the number to an integer seems like saner behavior anyway.
Assuming IEEE math, the division would yield +Infinity which would get
replaced by 1.0 at the bottom of the function, so nothing really
interesting would ensue; but avoiding divide-by-zero seems like a
good idea on general principles.
Per report from Piotr Stefaniak. No back-patch since this seems
mostly cosmetic.
This enables external code to create access methods. This is useful so
that extensions can add their own access methods which can be formally
tracked for dependencies, so that DROP operates correctly. Also, having
explicit support makes pg_dump work correctly.
Currently only index AMs are supported, but we expect different types to
be added in the future.
Authors: Alexander Korotkov, Petr Jelínek
Reviewed-By: Teodor Sigaev, Petr Jelínek, Jim Nasby
Commitfest-URL: https://commitfest.postgresql.org/9/353/
Discussion: https://www.postgresql.org/message-id/CAPpHfdsXwZmojm6Dx+TJnpYk27kT4o7Ri6X_4OSWcByu1Rm+VA@mail.gmail.com
All along, this function should have treated WindowFuncs in a manner
similar to Aggrefs, ie with an option whether or not to recurse into them.
By not considering the case, it was always recursing, which is OK for most
callers (although I suspect that the case in prepare_sort_from_pathkeys
might represent a bug). But now we need return-without-recursing behavior
as well. There are also more than a few callers that should never see a
WindowFunc, and now we'll get some error checking on that.
In commit 1d97c19a0f and later c1d9579dd8, we extended
pull_var_clause's API by adding enum-type arguments. That's sort of a pain
to maintain, though, because it means every time we add a new behavior we
must touch every last one of the call sites, even if there's a reasonable
default behavior that most of them could use. Let's switch over to using a
bitmask of flags, instead; that seems more maintainable and might save a
nanosecond or two as well. This commit changes no behavior in itself,
though I'm going to follow it up with one that does add a new behavior.
In passing, remove flatten_tlist(), which has not been used since 9.1
and would otherwise need the same API changes.
Removing these enums means that optimizer/tlist.h no longer needs to
depend on optimizer/var.h. Changing that caused a number of C files to
need addition of #include "optimizer/var.h" (probably we can thank old
runs of pgrminclude for that); but on balance it seems like a good change
anyway.
This patch reduces pg_am to just two columns, a name and a handler
function. All the data formerly obtained from pg_am is now provided
in a C struct returned by the handler function. This is similar to
the designs we've adopted for FDWs and tablesample methods. There
are multiple advantages. For one, the index AM's support functions
are now simple C functions, making them faster to call and much less
error-prone, since the C compiler can now check function signatures.
For another, this will make it far more practical to define index access
methods in installable extensions.
A disadvantage is that SQL-level code can no longer see attributes
of index AMs; in particular, some of the crosschecks in the opr_sanity
regression test are no longer possible from SQL. We've addressed that
by adding a facility for the index AM to perform such checks instead.
(Much more could be done in that line, but for now we're content if the
amvalidate functions more or less replace what opr_sanity used to do.)
We might also want to expose some sort of reporting functionality, but
this patch doesn't do that.
Alexander Korotkov, reviewed by Petr Jelínek, and rather heavily
editorialized on by me.
A report from Andy Colson showed that gincostestimate() was not being
nearly paranoid enough about whether to believe the statistics it finds in
the index metapage. The problem is that the metapage stats (other than the
pending-pages count) are only updated by VACUUM, and in the worst case
could still reflect the index's original empty state even when it has grown
to many entries. We attempted to deal with that by scaling up the stats to
match the current index size, but if nEntries is zero then scaling it up
still gives zero. Moreover, the proportion of pages that are entry pages
vs. data pages vs. pending pages is unlikely to be estimated very well by
scaling if the index is now orders of magnitude larger than before.
We can improve matters by expanding the use of the rule-of-thumb estimates
I introduced in commit 7fb008c5ee59b040: if the index has grown by more
than a cutoff amount (here set at 4X growth) since VACUUM, then use the
rule-of-thumb numbers instead of scaling. This might not be exactly right
but it seems much less likely to produce insane estimates.
I also improved both the scaling estimate and the rule-of-thumb estimate
to account for numPendingPages, since it's reasonable to expect that that
is accurate in any case, and certainly pages that are in the pending list
are not either entry or data pages.
As a somewhat separate issue, adjust the estimation equations that are
concerned with extra fetches for partial-match searches. These equations
suppose that a fraction partialEntries / numEntries of the entry and data
pages will be visited as a consequence of a partial-match search. Now,
it's physically impossible for that fraction to exceed one, but our
estimate of partialEntries is mostly bunk, and our estimate of numEntries
isn't exactly gospel either, so we could arrive at a silly value. In the
example presented by Andy we were coming out with a value of 100, leading
to insane cost estimates. Clamp the fraction to one to avoid that.
Like the previous patch, back-patch to all supported branches; this
problem can be demonstrated in one form or another in all of them.
We tried to fetch statistics data from the index metapage, which does not
work if the index isn't actually present. If the index is hypothetical,
instead extrapolate some plausible internal statistics based on the index
page count provided by the index-advisor plugin.
There was already some code in gincostestimate() to invent internal stats
in this way, but since it was only meant as a stopgap for pre-9.1 GIN
indexes that hadn't been vacuumed since upgrading, it was pretty crude.
If we want it to support index advisors, we should try a little harder.
A small amount of testing says that it's better to estimate the entry pages
as 90% of the index, not 100%. Also, estimating the number of entries
(keys) as equal to the heap tuple count could be wildly wrong in either
direction. Instead, let's estimate 100 entries per entry page.
Perhaps someday somebody will want the index advisor to be able to provide
these numbers more directly, but for the moment this should serve.
Problem report and initial patch by Julien Rouhaud; modified by me to
invent less-bogus internal statistics. Back-patch to all supported
branches, since we've supported index advisors since 9.0.
Previously, a function call appearing at the top level of WHERE had a
hard-wired selectivity estimate of 0.3333333, a kludge conveniently dated
in the source code itself to July 1992. The expectation at the time was
that somebody would soon implement estimator support functions analogous
to those for operators; but no such code has appeared, nor does it seem
likely to in the near future. We do have an alternative solution though,
at least for immutable functions on single relations: creating an
expression index on the function call will allow ANALYZE to gather stats
about the function's selectivity. But the code in clause_selectivity()
failed to make use of such data even if it exists.
Refactor so that that will happen. I chose to make it try this technique
for any clause type for which clause_selectivity() doesn't have a special
case, not just functions. To avoid adding unnecessary overhead in the
common case where we don't learn anything new, make selfuncs.c provide an
API that hooks directly to examine_variable() and then var_eq_const(),
rather than the previous coding which laboriously constructed an OpExpr
only so that it could be expensively deconstructed again.
I preserved the behavior that the default estimate for a function call
is 0.3333333. (For any other expression node type, it's 0.5, as before.)
I had originally thought to make the default be 0.5 across the board, but
changing a default estimate that's survived for twenty-three years seems
like something not to do without a lot more testing than I care to put
into it right now.
Per a complaint from Jehan-Guillaume de Rorthais. Back-patch into 9.5,
but not further, at least for the moment.
Previously, convert_one_string_to_scalar() would examine up to 20 bytes of
the input string, producing a scalar conversion with theoretical precision
far greater than is of any possible use considering the other limitations
on the accuracy of the resulting selectivity estimate. (I think this
choice might pre-date the caller-level logic that strips any common prefix
of the strings; before that, there could have been value in scanning the
strings far enough to use all the precision available in a double.)
Aside from wasting cycles to little purpose, this choice meant that the
"denom" variable could grow to as much as 256^21 = 3.74e50, which could
overflow in some non-IEEE float arithmetics. While we don't really support
any machines with non-IEEE arithmetic anymore, this still seems like quite
an unnecessary platform dependency. Limit the scan to 12 bytes instead,
thus limiting "denom" to 256^13 = 2.03e31, a value more likely to be
computable everywhere.
Per testing by Greg Stark, which showed overflow failures in our standard
regression tests on VAX.
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.
This formalizes a decision implicit in commit
4ea51cdfe8 and adds clean detection of
affected systems. Vendor updates are available for each such known bug.
Back-patch to 9.5, where the aforementioned commit first appeared.
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
Alexander Korotkov
Bruce Momjian
Tom Lane
Michael Paquier
Tomas Vondra
Peter Eisentraut
Dean Rasheed
Andres Freund
Alvaro Herrera
Teodor Sigaev
Peter Eisentraut
Robert Haas
Simon Riggs
Stephen Frost
Noah Misch