This moves sample scan support to below tableam. It's not optional as
there is, in contrast to e.g. bitmap heap scans, no alternative way to
perform tablesample queries. If an AM can't deal with the block based
API, it will have to throw an ERROR.
The tableam callbacks for this are block based, but given the current
TsmRoutine interface, that seems to be required.
The new interface doesn't require TsmRoutines to perform visibility
checks anymore - that requires the TsmRoutine to know details about
the AM, which we want to avoid. To continue to allow taking the
returned number of tuples account SampleScanState now has a donetuples
field (which previously e.g. existed in SystemRowsSamplerData), which
is only incremented after the visibility check succeeds.
Author: Andres Freund
Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
Previously, the planner created RangeTblEntry and RelOptInfo structs
for every partition of a partitioned table, even though many of them
might later be deemed uninteresting thanks to partition pruning logic.
This incurred significant overhead when there are many partitions.
Arrange to postpone creation of these data structures until after
we've processed the query enough to identify restriction quals for
the partitioned table, and then apply partition pruning before not
after creation of each partition's data structures. In this way
we need not open the partition relations at all for partitions that
the planner has no real interest in.
For queries that can be proven at plan time to access only a small
number of partitions, this patch improves the practical maximum
number of partitions from under 100 to perhaps a few thousand.
Amit Langote, reviewed at various times by Dilip Kumar, Jesper Pedersen,
Yoshikazu Imai, and David Rowley
Discussion: https://postgr.es/m/9d7c5112-cb99-6a47-d3be-cf1ee6862a1d@lab.ntt.co.jp
This is an SQL-standard feature that allows creating columns that are
computed from expressions rather than assigned, similar to a view or
materialized view but on a column basis.
This implements one kind of generated column: stored (computed on
write). Another kind, virtual (computed on read), is planned for the
future, and some room is left for it.
Reviewed-by: Michael Paquier <michael@paquier.xyz>
Reviewed-by: Pavel Stehule <pavel.stehule@gmail.com>
Discussion: https://www.postgresql.org/message-id/flat/b151f851-4019-bdb1-699e-ebab07d2f40a@2ndquadrant.com
To support building indexes over tables of different AMs, the scans to
do so need to be routed through the table AM. While moving a fair
amount of code, nearly all the changes are just moving code to below a
callback.
Currently the range based interface wouldn't make much sense for non
block based table AMs. But that seems aceptable for now.
Author: Andres Freund
Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
We've been creating duplicate RTEs for partitioned tables just
because we do so for regular inheritance parent tables. But unlike
regular-inheritance parents which are themselves regular tables
and thus need to be scanned, partitioned tables don't need the
extra RTE.
This makes the conditions for building a child RTE the same as those
for building an AppendRelInfo, allowing minor simplification in
expand_single_inheritance_child. Since the planner's actual processing
is driven off the AppendRelInfo list, nothing much changes beyond that,
we just have one fewer useless RTE entry.
Amit Langote, reviewed and hacked a bit by me
Discussion: https://postgr.es/m/9d7c5112-cb99-6a47-d3be-cf1ee6862a1d@lab.ntt.co.jp
If there's only one child relation, the Append or MergeAppend isn't
doing anything useful, and can be elided. It does have a purpose
during planning though, which is to serve as a buffer between parent
and child Var numbering. Therefore we keep it all the way through
to setrefs.c, and get rid of it only after fixing references in the
plan level(s) above it. This works largely the same as setrefs.c's
ancient hack to get rid of no-op SubqueryScan nodes, and can even
share some code with that.
Note the change to make setrefs.c use apply_tlist_labeling rather than
ad-hoc code. This has the effect of propagating the child's resjunk
and ressortgroupref labels, which formerly weren't propagated when
removing a SubqueryScan. Doing that is demonstrably necessary for
the [Merge]Append cases, and seems harmless for SubqueryScan, if only
because trivial_subqueryscan is afraid to collapse cases where the
resjunk marking differs. (I suspect that restriction could now be
removed, though it's unclear that it'd make any new matches possible,
since the outer query can't have references to a child resjunk column.)
David Rowley, reviewed by Alvaro Herrera and Tomas Vondra
Discussion: https://postgr.es/m/CAKJS1f_7u8ATyJ1JGTMHFoKDvZdeF-iEBhs+sM_SXowOr9cArg@mail.gmail.com
The code would do "PQclear(res)" twice if lo_unlink failed, evidently
due to careless thinking about how far out a "break" would break.
Remove the extra PQclear and adjust the loop logic so that we'll fall
out of both levels of loop after an error, as was clearly the intent.
Spotted by Coverity. I have no idea why it took this long to notice,
since the bug has been there since commit 67ccbb080. Accordingly,
back-patch to all supported branches.
This adds new, required, table AM callbacks for insert/delete/update
and lock_tuple. To be able to reasonably use those, the EvalPlanQual
mechanism had to be adapted, moving more logic into the AM.
Previously both delete/update/lock call-sites and the EPQ mechanism had
to have awareness of the specific tuple format to be able to fetch the
latest version of a tuple. Obviously that needs to be abstracted
away. To do so, move the logic that find the latest row version into
the AM. lock_tuple has a new flag argument,
TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last
version, rather than the current one. It'd have been possible to do
so via a separate callback as well, but finding the last version
usually also necessitates locking the newest version, making it
sensible to combine the two. This replaces the previous use of
EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously
signaled either a concurrent update or delete, is now split into two,
to avoid callers needing AM specific knowledge to differentiate.
The move of finding the latest row version into tuple_lock means that
encountering a row concurrently moved into another partition will now
raise an error about "tuple to be locked" rather than "tuple to be
updated/deleted" - which is accurate, as that always happens when
locking rows. While possible slightly less helpful for users, it seems
like an acceptable trade-off.
As part of this commit HTSU_Result has been renamed to TM_Result, and
its members been expanded to differentiated between updating and
deleting. HeapUpdateFailureData has been renamed to TM_FailureData.
The interface to speculative insertion is changed so nodeModifyTable.c
does not have to set the speculative token itself anymore. Instead
there's a version of tuple_insert, tuple_insert_speculative, that
performs the speculative insertion (without requiring a flag to signal
that fact), and the speculative insertion is either made permanent
with table_complete_speculative(succeeded = true) or aborted with
succeeded = false).
Note that multi_insert is not yet routed through tableam, nor is
COPY. Changing multi_insert requires changes to copy.c that are large
enough to better be done separately.
Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED
VIEW are also only going to be adjusted in a later commit.
Author: Andres Freund and Haribabu Kommi
Discussion:
https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.dehttps://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.dehttps://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
This adds a flag "deterministic" to collations. If that is false,
such a collation disables various optimizations that assume that
strings are equal only if they are byte-wise equal. That then allows
use cases such as case-insensitive or accent-insensitive comparisons
or handling of strings with different Unicode normal forms.
This functionality is only supported with the ICU provider. At least
glibc doesn't appear to have any locales that work in a
nondeterministic way, so it's not worth supporting this for the libc
provider.
The term "deterministic comparison" in this context is from Unicode
Technical Standard #10
(https://unicode.org/reports/tr10/#Deterministic_Comparison).
This patch makes changes in three areas:
- CREATE COLLATION DDL changes and system catalog changes to support
this new flag.
- Many executor nodes and auxiliary code are extended to track
collations. Previously, this code would just throw away collation
information, because the eventually-called user-defined functions
didn't use it since they only cared about equality, which didn't
need collation information.
- String data type functions that do equality comparisons and hashing
are changed to take the (non-)deterministic flag into account. For
comparison, this just means skipping various shortcuts and tie
breakers that use byte-wise comparison. For hashing, we first need
to convert the input string to a canonical "sort key" using the ICU
analogue of strxfrm().
Reviewed-by: Daniel Verite <daniel@manitou-mail.org>
Reviewed-by: Peter Geoghegan <pg@bowt.ie>
Discussion: https://www.postgresql.org/message-id/flat/1ccc668f-4cbc-0bef-af67-450b47cdfee7@2ndquadrant.com
Teach contrib/amcheck's bt_index_parent_check() function to take
advantage of the uniqueness property of heapkeyspace indexes in support
of a new verification option: non-pivot tuples (non-highkey tuples on
the leaf level) can optionally be re-found using a new search for each,
that starts from the root page. If a tuple cannot be re-found, report
that the index is corrupt.
The new "rootdescend" verification option is exhaustive, and can
therefore make a call to bt_index_parent_check() take a lot longer.
Re-finding tuples during verification is mostly intended as an option
for backend developers, since the corruption scenarios that it alone is
uniquely capable of detecting seem fairly far-fetched.
For example, "rootdescend" verification is much more likely to detect
corruption of the least significant byte of a key from a pivot tuple in
the root page of a B-Tree that already has at least three levels.
Typically, only a few tuples on a cousin leaf page are at risk of
"getting overlooked" by index scans in this scenario. The corrupt key
in the root page is only slightly corrupt: corrupt enough to give wrong
answers to some queries, and yet not corrupt enough to allow the problem
to be detected without verifying agreement between the leaf page and the
root page, skipping at least one internal page level. The existing
bt_index_parent_check() checks never cross more than a single level.
Author: Peter Geoghegan
Reviewed-By: Heikki Linnakangas
Discussion: https://postgr.es/m/CAH2-Wz=yTWnVu+HeHGKb2AGiADL9eprn-cKYAto4MkKOuiGtRQ@mail.gmail.com
Make nbtree treat all index tuples as having a heap TID attribute.
Index searches can distinguish duplicates by heap TID, since heap TID is
always guaranteed to be unique. This general approach has numerous
benefits for performance, and is prerequisite to teaching VACUUM to
perform "retail index tuple deletion".
Naively adding a new attribute to every pivot tuple has unacceptable
overhead (it bloats internal pages), so suffix truncation of pivot
tuples is added. This will usually truncate away the "extra" heap TID
attribute from pivot tuples during a leaf page split, and may also
truncate away additional user attributes. This can increase fan-out,
especially in a multi-column index. Truncation can only occur at the
attribute granularity, which isn't particularly effective, but works
well enough for now. A future patch may add support for truncating
"within" text attributes by generating truncated key values using new
opclass infrastructure.
Only new indexes (BTREE_VERSION 4 indexes) will have insertions that
treat heap TID as a tiebreaker attribute, or will have pivot tuples
undergo suffix truncation during a leaf page split (on-disk
compatibility with versions 2 and 3 is preserved). Upgrades to version
4 cannot be performed on-the-fly, unlike upgrades from version 2 to
version 3. contrib/amcheck continues to work with version 2 and 3
indexes, while also enforcing stricter invariants when verifying version
4 indexes. These stricter invariants are the same invariants described
by "3.1.12 Sequencing" from the Lehman and Yao paper.
A later patch will enhance the logic used by nbtree to pick a split
point. This patch is likely to negatively impact performance without
smarter choices around the precise point to split leaf pages at. Making
these two mostly-distinct sets of enhancements into distinct commits
seems like it might clarify their design, even though neither commit is
particularly useful on its own.
The maximum allowed size of new tuples is reduced by an amount equal to
the space required to store an extra MAXALIGN()'d TID in a new high key
during leaf page splits. The user-facing definition of the "1/3 of a
page" restriction is already imprecise, and so does not need to be
revised. However, there should be a compatibility note in the v12
release notes.
Author: Peter Geoghegan
Reviewed-By: Heikki Linnakangas, Alexander Korotkov
Discussion: https://postgr.es/m/CAH2-WzkVb0Kom=R+88fDFb=JSxZMFvbHVC6Mn9LJ2n=X=kS-Uw@mail.gmail.com
Use dedicated struct to represent nbtree insertion scan keys. Having a
dedicated struct makes the difference between search type scankeys and
insertion scankeys a lot clearer, and simplifies the signature of
several related functions. This is based on a suggestion by Andrey
Lepikhov.
Streamline how unique index insertions cache binary search progress.
Cache the state of in-progress binary searches within _bt_check_unique()
for later instead of having callers avoid repeating the binary search in
an ad-hoc manner. This makes it easy to add a new optimization:
_bt_check_unique() now falls out of its loop immediately in the common
case where it's already clear that there couldn't possibly be a
duplicate.
The new _bt_check_unique() scheme makes it a lot easier to manage cached
binary search effort afterwards, from within _bt_findinsertloc(). This
is needed for the upcoming patch to make nbtree tuples unique by
treating heap TID as a final tiebreaker column. Unique key binary
searches need to restore lower and upper bounds. They cannot simply
continue to use the >= lower bound as the offset to insert at, because
the heap TID tiebreaker column must be used in comparisons for the
restored binary search (unlike the original _bt_check_unique() binary
search, where scankey's heap TID column must be omitted).
Author: Peter Geoghegan, Heikki Linnakangas
Reviewed-By: Heikki Linnakangas, Andrey Lepikhov
Discussion: https://postgr.es/m/CAH2-WzmE6AhUdk9NdWBf4K3HjWXZBX3+umC7mH7+WDrKcRtsOw@mail.gmail.com
We should try to prewarm each database only once. Otherwise, if
prewarming fails for some reason, it will just keep retrying in an
infnite loop. This can happen if, for example, the database has been
dropped. The existing code was intended to implement the try-once
behavior, but failed to do so because it neglected to set
worker.bgw_restart_time to BGW_NEVER_RESTART.
Mithun Cy, per a report from Hans Buschmann
Discussion: http://postgr.es/m/CA+hUKGKpQJCWcgyy3QTC9vdn6uKAR_8r__A-MMm2GYfj45caag@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
... as well as its implementation from backend/access/hash/hashfunc.c to
backend/utils/hash/hashfn.c.
access/hash is the place for the hash index AM, not really appropriate
for generic facilities, which is what hash_any is; having things the old
way meant that anything using hash_any had to include the AM's include
file, pointlessly polluting its namespace with unrelated, unnecessary
cruft.
Also move the HTEqual strategy number to access/stratnum.h from
access/hash.h.
To avoid breaking third-party extension code, add an #include
"utils/hashutils.h" to access/hash.h. (An easily removed line by
committers who enjoy their asbestos suits to protect them from angry
extension authors.)
Discussion: https://postgr.es/m/201901251935.ser5e4h6djt2@alvherre.pgsql
This fixes two sets of issues related to the use of transient files in
the backend:
1) OpenTransientFile() has been used in some code paths with read-write
flags while read-only is sufficient, so switch those calls to be
read-only where necessary. These have been reported by Joe Conway.
2) When opening transient files, it is up to the caller to close the
file descriptors opened. In error code paths, CloseTransientFile() gets
called to clean up things before issuing an error. However in normal
exit paths, a lot of callers of CloseTransientFile() never actually
reported errors, which could leave a file descriptor open without
knowing about it. This is an issue I complained about a couple of
times, but never had the courage to write and submit a patch, so here we
go.
Note that one frontend code path is impacted by this commit so as an
error is issued when fetching control file data, making backend and
frontend to be treated consistently.
Reported-by: Joe Conway, Michael Paquier
Author: Michael Paquier
Reviewed-by: Álvaro Herrera, Georgios Kokolatos, Joe Conway
Discussion: https://postgr.es/m/20190301023338.GD1348@paquier.xyz
Discussion: https://postgr.es/m/c49b69ec-e2f7-ff33-4f17-0eaa4f2cef27@joeconway.com
In preparation for abstracting table storage, convert trigger.c to
track tuples in slots. Which also happens to make code calling
triggers simpler.
As the calling interface for triggers themselves is not changed in
this patch, HeapTuples still are extracted from the slot at that
time. But that's handled solely inside trigger.c, not visible to
callers. It's quite likely that we'll want to revise the external
trigger interface, but that's a separate large project.
As part of this work the slots used for old/new/return tuples are
moved from EState into ResultRelInfo, as different updated tables
might need different slots. The slots are now also now created
on-demand, which is good both from an efficiency POV, but also makes
the modifying code simpler.
Author: Andres Freund, Amit Khandekar and Ashutosh Bapat
Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
In commit b0eaa4c51b, we have avoided the creation of FSM for small tables.
So the functions that use FSM to compute the free space can return zero for
such tables. This was previously also possible for the cases where the
vacuum has not been triggered to update FSM.
This commit updates the comments in code and documentation to reflect this
behavior.
Author: John Naylor
Reviewed-by: Amit Kapila
Discussion: https://postgr.es/m/CACPNZCtba-3m1q3A8gxA_vxg=T7gQf7gMbpR4Ciy5LntY-j+0Q@mail.gmail.com
As things stood, buildfarm member dory failed. MSVC versions lacking
stdint.h are unusable for building PostgreSQL, but pg_config.h.win32
doesn't know that. Even so, we support other systems lacking stdint.h,
including buildfarm member gaur. Per a suggestion from Tom Lane.
Discussion: https://postgr.es/m/9598.1550353336@sss.pgh.pa.us
Upstream fixed bugs over the years, but none are confirmed to have
affected pgcrypto. We're better off naively tracking upstream than
reactively maintaining a twelve-year-old snapshot of upstream. Add a
header comment describing the synchronization procedure. Discard use of
INVERT_COMPARE_RESULT(); the domain of the comparisons in question is
{-1,0,1}, controlled entirely by code in imath.c.
Andrew Gierth reviewed the Makefile change. Tom Lane reviewed the
synchronization procedure description.
Discussion: https://postgr.es/m/20190203035704.GA6226@rfd.leadboat.com
Historically we've always materialized the full output of a CTE query,
treating WITH as an optimization fence (so that, for example, restrictions
from the outer query cannot be pushed into it). This is appropriate when
the CTE query is INSERT/UPDATE/DELETE, or is recursive; but when the CTE
query is non-recursive and side-effect-free, there's no hazard of changing
the query results by pushing restrictions down.
Another argument for materialization is that it can avoid duplicate
computation of an expensive WITH query --- but that only applies if
the WITH query is called more than once in the outer query. Even then
it could still be a net loss, if each call has restrictions that
would allow just a small part of the WITH query to be computed.
Hence, let's change the behavior for WITH queries that are non-recursive
and side-effect-free. By default, we will inline them into the outer
query (removing the optimization fence) if they are called just once.
If they are called more than once, we will keep the old behavior by
default, but the user can override this and force inlining by specifying
NOT MATERIALIZED. Lastly, the user can force the old behavior by
specifying MATERIALIZED; this would mainly be useful when the query had
deliberately been employing WITH as an optimization fence to prevent a
poor choice of plan.
Andreas Karlsson, Andrew Gierth, David Fetter
Discussion: https://postgr.es/m/87sh48ffhb.fsf@news-spur.riddles.org.uk
Test for the compiler builtins __builtin_clz, __builtin_ctz, and
__builtin_popcount, and make use of these in preference to
handwritten C code if they're available. Create src/port
infrastructure for "leftmost one", "rightmost one", and "popcount"
so as to centralize these decisions.
On x86_64, __builtin_popcount generally won't make use of the POPCNT
opcode because that's not universally supported yet. Provide code
that checks CPUID and then calls POPCNT via asm() if available.
This requires indirecting through a function pointer, which is
an annoying amount of overhead for a one-instruction operation,
but it's probably not worth working harder than this for our
current use-cases.
I'm not sure we've found all the existing places that could profit
from this new infrastructure; but we at least touched all the
ones that used copied-and-pasted versions of the bitmapset.c code,
and got rid of multiple copies of the associated constant arrays.
While at it, replace c-compiler.m4's one-per-builtin-function
macros with a single one that can handle all the cases we need
to worry about so far. Also, because I'm paranoid, make those
checks into AC_LINK checks rather than just AC_COMPILE; the
former coding failed to verify that libgcc has support for the
builtin, in cases where it's not inline code.
David Rowley, Thomas Munro, Alvaro Herrera, Tom Lane
Discussion: https://postgr.es/m/CAKJS1f9WTAGG1tPeJnD18hiQW5gAk59fQ6WK-vfdAKEHyRg2RA@mail.gmail.com
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
Previously, floating-point output was done by rounding to a specific
decimal precision; by default, to 6 or 15 decimal digits (losing
information) or as requested using extra_float_digits. Drivers that
wanted exact float values, and applications like pg_dump that must
preserve values exactly, set extra_float_digits=3 (or sometimes 2 for
historical reasons, though this isn't enough for float4).
Unfortunately, decimal rounded output is slow enough to become a
noticable bottleneck when dealing with large result sets or COPY of
large tables when many floating-point values are involved.
Floating-point output can be done much faster when the output is not
rounded to a specific decimal length, but rather is chosen as the
shortest decimal representation that is closer to the original float
value than to any other value representable in the same precision. The
recently published Ryu algorithm by Ulf Adams is both relatively
simple and remarkably fast.
Accordingly, change float4out/float8out to output shortest decimal
representations if extra_float_digits is greater than 0, and make that
the new default. Applications that need rounded output can set
extra_float_digits back to 0 or below, and take the resulting
performance hit.
We make one concession to portability for systems with buggy
floating-point input: we do not output decimal values that fall
exactly halfway between adjacent representable binary values (which
would rely on the reader doing round-to-nearest-even correctly). This
is known to be a problem at least for VS2013 on Windows.
Our version of the Ryu code originates from
https://github.com/ulfjack/ryu/ at commit c9c3fb1979, but with the
following (significant) modifications:
- Output format is changed to use fixed-point notation for small
exponents, as printf would, and also to use lowercase 'e', a
minimum of 2 exponent digits, and a mandatory sign on the exponent,
to keep the formatting as close as possible to previous output.
- The output of exact midpoint values is disabled as noted above.
- The integer fast-path code is changed somewhat (since we have
fixed-point output and the upstream did not).
- Our project style has been largely applied to the code with the
exception of C99 declaration-after-statement, which has been
retained as an exception to our present policy.
- Most of upstream's debugging and conditionals are removed, and we
use our own configure tests to determine things like uint128
availability.
Changing the float output format obviously affects a number of
regression tests. This patch uses an explicit setting of
extra_float_digits=0 for test output that is not expected to be
exactly reproducible (e.g. due to numerical instability or differing
algorithms for transcendental functions).
Conversions from floats to numeric are unchanged by this patch. These
may appear in index expressions and it is not yet clear whether any
change should be made, so that can be left for another day.
This patch assumes that the only supported floating point format is
now IEEE format, and the documentation is updated to reflect that.
Code by me, adapting the work of Ulf Adams and other contributors.
References:
https://dl.acm.org/citation.cfm?id=3192369
Reviewed-by: Tom Lane, Andres Freund, Donald Dong
Discussion: https://postgr.es/m/87r2el1bx6.fsf@news-spur.riddles.org.uk
Ever since its birth, ReorderBufferBuildTupleCidHash() has contained an
assertion that a catalog tuple cannot change Cmax after acquiring one. But
that's wrong: if a subtransaction executes DDL that affects that catalog
tuple, and later aborts and another DDL affects the same tuple, it will
change Cmax. Relax the assertion to merely verify that the Cmax remains
valid and monotonically increasing, instead.
Add a test that tickles the relevant code.
Diagnosed by, and initial patch submitted by: Arseny Sher
Co-authored-by: Arseny Sher
Discussion: https://postgr.es/m/874l9p8hyw.fsf@ars-thinkpad
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
Up to now postgres_fdw has been using create_foreignscan_path() to
generate not only base-relation paths, but also paths for foreign joins
and foreign upperrels. This is wrong, because create_foreignscan_path()
calls get_baserel_parampathinfo() which will only do the right thing for
baserels. It accidentally fails to fail for unparameterized paths, which
are the only ones postgres_fdw (thought it) was handling, but we really
need different APIs for the baserel and join cases.
In HEAD, the best thing to do seems to be to split up the baserel,
joinrel, and upperrel cases into three functions so that they can
have different APIs. I haven't actually given create_foreign_join_path
a different API in this commit: we should spend a bit of time thinking
about just what we want to do there, since perhaps FDWs would want to
do something different from the build-up-a-join-pairwise approach that
get_joinrel_parampathinfo expects. In the meantime, since postgres_fdw
isn't prepared to generate parameterized joins anyway, just give it a
defense against trying to plan joins with lateral refs.
In addition (and this is what triggered this whole mess) fix bug #15613
from Srinivasan S A, by teaching file_fdw and postgres_fdw that plain
baserel foreign paths still have outer refs if the relation has
lateral_relids. Add some assertions in relnode.c to catch future
occurrences of the same error --- in particular, to catch other FDWs
doing that, but also as backstop against core-code mistakes like the
one fixed by commit bdd9a99aa.
Bug #15613 also needs to be fixed in the back branches, but the
appropriate fix will look quite a bit different there, since we don't
want to assume that existing FDWs get the word right away.
Discussion: https://postgr.es/m/15613-092be1be9576c728@postgresql.org
The previous tacit assumption that index_form_tuple() hides differences
in the TOAST state of its input datums was wrong. Normalize input
varlena datums by decompressing compressed values, and forming a new
index tuple for fingerprinting using uncompressed inputs. The final
normalized representation may actually be compressed once again within
index_form_tuple(), though that shouldn't matter. When the original
tuple is found to have no datums that are compressed inline, fingerprint
the original tuple directly.
Normalization avoids false positive reports of corruption in certain
cases. For example, the executor can apply toasting with some inline
compression to an entire heap tuple because its input has a single
external TOAST pointer. Varlena datums for other attributes that are
not particularly good candidates for inline compression can be
compressed in the heap tuple in passing, without the representation of
the same values in index tuples ever receiving concomitant inline
compression.
Add a test case to recreate the issue in a simpler though less realistic
way: by exploiting differences in pg_attribute.attstorage between heap
and index relations.
This bug was discovered by me during testing of an upcoming set of nbtree
enhancements. It was also independently reported by Andreas Kunert, as
bug #15597. His test case was rather more realistic than the one I
ended up using.
Bug: #15597
Discussion: https://postgr.es/m/CAH2-WznrVd9ie+TTJ45nDT+v2nUt6YJwQrT9SebCdQKtAvfPZw@mail.gmail.com
Discussion: https://postgr.es/m/15597-294e5d3e7f01c407@postgresql.org
Backpatch: 11-, where heapallindexed verification was introduced.
In b0eaa4c51b, we allow FSM to be created only after 4 pages. One of the
tests check the FSM contents and to do that it populates many tuples in
the relation. The FSM contents depend on the availability of freespace in
the page and it could vary because of the alignment of tuples.
This commit removes the dependency on FSM contents.
Author: Amit Kapila
Discussion: https://postgr.es/m/CAA4eK1KADF6K1bagr0--mGv3dMcZ%3DH_Z-Qtvdfbp5PjaC6PJJA%40mail.gmail.com
Previously, all heaps had FSMs. For very small tables, this means that the
FSM took up more space than the heap did. This is wasteful, so now we
refrain from creating the FSM for heaps with 4 pages or fewer. If the last
known target block has insufficient space, we still try to insert into some
other page before giving up and extending the relation, since doing
otherwise leads to table bloat. Testing showed that trying every page
penalized performance slightly, so we compromise and try every other page.
This way, we visit at most two pages. Any pages with wasted free space
become visible at next relation extension, so we still control table bloat.
As a bonus, directly attempting one or two pages can even be faster than
consulting the FSM would have been.
Once the FSM is created for a heap we don't remove it even if somebody
deletes all the rows from the corresponding relation. We don't think it is
a useful optimization as it is quite likely that relation will again grow
to the same size.
Author: John Naylor, Amit Kapila
Reviewed-by: Amit Kapila
Tested-by: Mithun C Y
Discussion: https://www.postgresql.org/message-id/CAJVSVGWvB13PzpbLEecFuGFc5V2fsO736BsdTakPiPAcdMM5tQ@mail.gmail.com
The old name of this file was never a very good indication of what it
was for. Now that there's also access/relation.h, we have a potential
confusion hazard as well, so let's rename it to something more apropos.
Per discussion, "pathnodes.h" is reasonable, since a good fraction of
the file is Path node definitions.
While at it, tweak a couple of other headers that were gratuitously
importing relation.h into modules that don't need it.
Discussion: https://postgr.es/m/7719.1548688728@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
estimate_path_cost_size() failed to re-use cached costs when the cached
startup/total cost was 0, so it calculated the costs redundantly.
This is an oversight in commit aa09cd242f; but apply the patch to HEAD
only because there are no reports of actual trouble from that.
Author: Etsuro Fujita
Discussion: https://postgr.es/m/5C4AF3F3.4060409%40lab.ntt.co.jp
The fact that "SELECT expression" has no base relations has long been a
thorn in the side of the planner. It makes it hard to flatten a sub-query
that looks like that, or is a trivial VALUES() item, because the planner
generally uses relid sets to identify sub-relations, and such a sub-query
would have an empty relid set if we flattened it. prepjointree.c contains
some baroque logic that works around this in certain special cases --- but
there is a much better answer. We can replace an empty FROM clause with a
dummy RTE that acts like a table of one row and no columns, and then there
are no such corner cases to worry about. Instead we need some logic to
get rid of useless dummy RTEs, but that's simpler and covers more cases
than what was there before.
For really trivial cases, where the query is just "SELECT expression" and
nothing else, there's a hazard that adding the extra RTE makes for a
noticeable slowdown; even though it's not much processing, there's not
that much for the planner to do overall. However testing says that the
penalty is very small, close to the noise level. In more complex queries,
this is able to find optimizations that we could not find before.
The new RTE type is called RTE_RESULT, since the "scan" plan type it
gives rise to is a Result node (the same plan we produced for a "SELECT
expression" query before). To avoid confusion, rename the old ResultPath
path type to GroupResultPath, reflecting that it's only used in degenerate
grouping cases where we know the query produces just one grouped row.
(It wouldn't work to unify the two cases, because there are different
rules about where the associated quals live during query_planner.)
Note: although this touches readfuncs.c, I don't think a catversion
bump is required, because the added case can't occur in stored rules,
only plans.
Patch by me, reviewed by David Rowley and Mark Dilger
Discussion: https://postgr.es/m/15944.1521127664@sss.pgh.pa.us
Previously, all heaps had FSMs. For very small tables, this means that the
FSM took up more space than the heap did. This is wasteful, so now we
refrain from creating the FSM for heaps with 4 pages or fewer. If the last
known target block has insufficient space, we still try to insert into some
other page before giving up and extending the relation, since doing
otherwise leads to table bloat. Testing showed that trying every page
penalized performance slightly, so we compromise and try every other page.
This way, we visit at most two pages. Any pages with wasted free space
become visible at next relation extension, so we still control table bloat.
As a bonus, directly attempting one or two pages can even be faster than
consulting the FSM would have been.
Once the FSM is created for a heap we don't remove it even if somebody
deletes all the rows from the corresponding relation. We don't think it is
a useful optimization as it is quite likely that relation will again grow
to the same size.
Author: John Naylor with design inputs and some code contribution by Amit Kapila
Reviewed-by: Amit Kapila
Tested-by: Mithun C Y
Discussion: https://www.postgresql.org/message-id/CAJVSVGWvB13PzpbLEecFuGFc5V2fsO736BsdTakPiPAcdMM5tQ@mail.gmail.com
Before this change FunctionCallInfoData, the struct arguments etc for
V1 function calls are stored in, always had space for
FUNC_MAX_ARGS/100 arguments, storing datums and their nullness in two
arrays. For nearly every function call 100 arguments is far more than
needed, therefore wasting memory. Arg and argnull being two separate
arrays also guarantees that to access a single argument, two
cachelines have to be touched.
Change the layout so there's a single variable-length array with pairs
of value / isnull. That drastically reduces memory consumption for
most function calls (on x86-64 a two argument function now uses
64bytes, previously 936 bytes), and makes it very likely that argument
value and its nullness are on the same cacheline.
Arguments are stored in a new NullableDatum struct, which, due to
padding, needs more memory per argument than before. But as usually
far fewer arguments are stored, and individual arguments are cheaper
to access, that's still a clear win. It's likely that there's other
places where conversion to NullableDatum arrays would make sense,
e.g. TupleTableSlots, but that's for another commit.
Because the function call information is now variable-length
allocations have to take the number of arguments into account. For
heap allocations that can be done with SizeForFunctionCallInfoData(),
for on-stack allocations there's a new LOCAL_FCINFO(name, nargs) macro
that helps to allocate an appropriately sized and aligned variable.
Some places with stack allocation function call information don't know
the number of arguments at compile time, and currently variably sized
stack allocations aren't allowed in postgres. Therefore allow for
FUNC_MAX_ARGS space in these cases. They're not that common, so for
now that seems acceptable.
Because of the need to allocate FunctionCallInfo of the appropriate
size, older extensions may need to update their code. To avoid subtle
breakages, the FunctionCallInfoData struct has been renamed to
FunctionCallInfoBaseData. Most code only references FunctionCallInfo,
so that shouldn't cause much collateral damage.
This change is also a prerequisite for more efficient expression JIT
compilation (by allocating the function call information on the stack,
allowing LLVM to optimize it away); previously the size of the call
information caused problems inside LLVM's optimizer.
Author: Andres Freund
Reviewed-By: Tom Lane
Discussion: https://postgr.es/m/20180605172952.x34m5uz6ju6enaem@alap3.anarazel.de
Previously, estimate_path_cost_size() didn't account for tlist eval
costs, except when costing a foreign-grouping path using local
statistics, but such costs should be accounted for when costing that path
using remote estimates, because some of the tlist expressions might be
evaluated locally. Also, such costs should be accounted for in the case
of a foreign-scan or foreign-join path, because the tlist might contain
PlaceHolderVars, which postgres_fdw currently evaluates locally.
This also fixes an oversight in my commit f8f6e44676.
Like that commit, apply this to HEAD only to avoid destabilizing existing
plan choices.
Author: Etsuro Fujita
Discussion: https://postgr.es/m/5BFD3EAD.2060301%40lab.ntt.co.jp
Andres Freund
Tom Lane
Peter Eisentraut
Peter Geoghegan
Michael Paquier
Robert Haas
Alvaro Herrera
Amit Kapila
Noah Misch
Andrew Gierth
Thomas Munro
Etsuro Fujita