Alvaro Herrera [Thu, 25 May 2023 10:36:18 +0000 (12:36 +0200)]
Fix pgbench in prepared mode with an empty pipeline
It crashes because it references memory that's not allocated in that
particular case. Fix by allocating it.
Reported-by: Alexander Lakhin <[email protected]>
Discussion: https://postgr.es/m/
bcf802a6-afc1-95b9-7bf4-
c5dd868ec144@gmail.com
Michael Paquier [Thu, 25 May 2023 07:26:08 +0000 (16:26 +0900)]
Fix typo in TAP tests of vacuumdb
Noticed while scanning the area, introduced in
582edc3.
Bruce Momjian [Thu, 25 May 2023 02:03:40 +0000 (22:03 -0400)]
doc: PG 16 relnotes, add author from previous merge
Reported-by: John Naylor
Discussion: https://postgr.es/m/CAFBsxsEMoNdg6CCMuQ-6YJ8G=AgGNMVuiTb5Di5bsoPfBb=9CA@mail.gmail.com
Bruce Momjian [Wed, 24 May 2023 16:18:43 +0000 (12:18 -0400)]
doc: PG 16 relnotes, wording adjustments
Reported-by: Erik Rijkers
Discussion: https://postgr.es/m/
c4a88acf-ab1a-a912-d4a9-
5e19b323c5ed@xs4all.nl
Bruce Momjian [Wed, 24 May 2023 13:54:34 +0000 (09:54 -0400)]
doc: PG 16 relnotes, merge and move vector items
Reported-by: John Naylor
Discussion: https://postgr.es/m/CAFBsxsEPg8L2MmGqavc8JByC=WF_Mnkhn-KKnFPkcqh0hydung@mail.gmail.com
Bruce Momjian [Wed, 24 May 2023 04:09:13 +0000 (00:09 -0400)]
doc: PG 16 relnotes, update xid/subxid searches item
Reported-by: John Naylor
Discussion: https://postgr.es/m/CAFBsxsEPg8L2MmGqavc8JByC=WF_Mnkhn-KKnFPkcqh0hydung@mail.gmail.com
Nathan Bossart [Wed, 24 May 2023 02:34:09 +0000 (19:34 -0700)]
Document deprecated createuser option.
2dcd1578c4 left the --role option undocumented, which is
inconsistent with other deprecated options such as pg_dump's
--blobs and --no-blobs. This change adds --role back to
createuser's documentation and usage output and marks it as
deprecated.
Suggested-by: Peter Eisentraut
Reviewed-by: Michael Paquier
Discussion: https://postgr.es/m/
0e85c9e7-4804-1cdb-5a4a-
c72c328f9ad8%40enterprisedb.com
Tom Lane [Tue, 23 May 2023 22:51:43 +0000 (18:51 -0400)]
Doc: update src/tools/pginclude/README.
Commit
f62975b2a made headerscheck and cpluspluscheck a little
smarter than this documentation gave them credit for. Make
their docs match current reality. Also add a warning about the
layer of dust that has settled on the rest of these scripts.
Discussion: https://postgr.es/m/769950.
1684797558@sss.pgh.pa.us
Tom Lane [Tue, 23 May 2023 15:15:57 +0000 (11:15 -0400)]
Fix the install rule for snowball_create.sql.
This file could be in the current (build) directory if we just
built it. However, when installing from a VPATH build from a
tarball, it will exist in the source directory and gmake will
therefore not rebuild it. Use the $< macro to find out where
gmake found it.
Oversight in
b3a0d8324, which also exposes a buildfarm testing gap:
we test install from VPATH builds from bare source trees, but not
from tarballs.
Per report from Christoph Berg.
Discussion: https://postgr.es/m/
[email protected]
Peter Eisentraut [Tue, 23 May 2023 13:19:33 +0000 (15:19 +0200)]
Use lower case for icu_validation_level values
Similar to client_min_messages etc.
Peter Eisentraut [Tue, 23 May 2023 13:19:12 +0000 (15:19 +0200)]
Punctuation improvement in postgresql.conf.sample
Peter Eisentraut [Tue, 23 May 2023 13:18:06 +0000 (15:18 +0200)]
Add newline at end of file
Bruce Momjian [Tue, 23 May 2023 04:27:01 +0000 (00:27 -0400)]
doc: PG 16 relnotes, SIMD improvements
Reported-by: John Naylor
Discussion: https://postgr.es/m/CAFBsxsEuAx4_nq=200u=70S5r83C2hzO-a9+c6YXTCbOFeDAfw@mail.gmail.com
Tom Lane [Mon, 22 May 2023 21:08:08 +0000 (17:08 -0400)]
Stamp 16beta1.
Bruce Momjian [Mon, 22 May 2023 17:58:24 +0000 (13:58 -0400)]
doc: PG 16 relnotes, add major features list
Reported-by: Jonathan Katz
Discussion: https://postgr.es/m/
2fd2cc0e-df39-3e77-8fcf-
35aad5796b0a@postgresql.org
Tom Lane [Mon, 22 May 2023 15:50:20 +0000 (11:50 -0400)]
Spell the values of libpq's gssdelegation parameter as "0" and "1".
That's how other boolean options are handled, so do likewise.
The previous coding with "enable" and "disable" was seemingly
modeled on gssencmode, but that's a three-way flag.
While at it, add PGGSSDELEGATION to the set of environment
variables cleared by pg_regress and Utils.pm.
Abhijit Menon-Sen, per gripe from Alvaro Herrera
Discussion: https://postgr.es/m/
20230522091609[email protected]
Bruce Momjian [Mon, 22 May 2023 14:20:53 +0000 (10:20 -0400)]
doc: PG 16 relnotes, improve desc. of pg_log_standby_snapshot()
Peter Eisentraut [Mon, 22 May 2023 10:44:31 +0000 (12:44 +0200)]
Translation updates
Source-Git-URL: https://git.postgresql.org/git/pgtranslation/messages.git
Source-Git-Hash:
642d41265b1ea68ae71a66ade5c5440ba366a890
Alvaro Herrera [Mon, 22 May 2023 10:26:40 +0000 (12:26 +0200)]
pg_dump doc: mention 'long' is valid for --compress
Apparently an oversight in
2820adf7755d.
Bruce Momjian [Mon, 22 May 2023 03:39:27 +0000 (23:39 -0400)]
doc: PG 16 relnotes, improve description of standby log. decode
Improve description of "Allow logical decoding on standbys".
Nathan Bossart [Mon, 22 May 2023 03:03:56 +0000 (20:03 -0700)]
Rename some createuser options.
This change renames --admin to --with-admin, --role to --member-of,
and --member to --with-member. Many people found the previous
names to be confusing. The --admin and --member options are new in
v16, but --role has been there for a while, so that one has been
kept (but left undocumented) for backward compatibility.
Suggested-by: Peter Eisentraut
Reviewed-by: Tom Lane, Michael Paquier
Discussion: https://postgr.es/m/ZFvVZvQDliIWmOwg%40momjian.us
Bruce Momjian [Mon, 22 May 2023 02:52:42 +0000 (22:52 -0400)]
doc: PG 16 relnotes, misc merged items and bootstrap detail
Reported-by: Andres Freund, jian he
Discussion: https://postgr.es/m/
20230521171341[email protected],
20230521171341[email protected]
Bruce Momjian [Sun, 21 May 2023 19:57:37 +0000 (15:57 -0400)]
doc: PG 16 relnotes, misc. updates
Reported-by: Tom Lane
Discussion: https://postgr.es/m/277016.
1684689065@sss.pgh.pa.us
Tom Lane [Sun, 21 May 2023 19:25:43 +0000 (15:25 -0400)]
In clause_is_computable_at(), test required_relids for clone clauses.
Use the clause's required_relids not clause_relids for testing
whether it is computable at the current join level, if it is a
clone clause generated by deconstruct_distribute_oj_quals().
Arguably, this is more correct and we should do it for all clauses;
that would at least remove the handwavy claim that we are doing
it to save cycles compared to inspecting Vars individually.
However, attempting to do that exposes that we are not being careful
to compute an accurate value for required_relids in all cases.
I'm unsure whether it's a good idea to attempt to do that for v16,
or leave it as future clean-up. In the meantime, this quick hack
demonstrably fixes some cases, so let's squeeze it in for beta1.
Patch by me, but great thanks to Richard Guo for investigation
and testing. The new test cases are all modeled on his examples.
Discussion: https://postgr.es/m/CAMbWs4-_vwkBij4XOQ5ukxUvLgwTm0kS5_DO9CicUeKbEfKjUw@mail.gmail.com
Andres Freund [Sun, 21 May 2023 16:48:37 +0000 (09:48 -0700)]
Remove over-eager assertion in ExtendBufferedRelTo()
The assertion checked that the size of the relation is not "too large" - but
the code is explicitly dealing with the possibility of another backend
extending the relation concurrently. In that case the new relation size could
be bigger than what the current backend needs, wrongly triggering an assertion
failure.
Unfortunately it is hard to write a reliable and affordable regression tests
for this, as a lot of concurrency is needed to encounter the bug.
Introduced in
31966b151e6a.
Reported-by: Melanie Plageman <[email protected]>
Andres Freund [Sun, 21 May 2023 16:44:55 +0000 (09:44 -0700)]
Optimize walsender wake up logic using condition variables
WalSndWakeup() currently loops through all the walsenders slots, with a
spinlock acquisition and release for every iteration, to wake up waiting
walsenders.
This commonly was not a problem before
e101dfac3a53c. But, to allow logical
decoding on standbys, we need to wake up logical walsenders after every WAL
record is applied on the standby, rather just when flushing WAL or switching
timelines. This causes a performance regression for workloads replaying a lot
of WAL records.
To solve this, we use condition variable (CV) to efficiently wake up
walsenders in WalSndWakeup().
Every walsender prepares to sleep on a shared memory CV. Note that it just
prepares to sleep on the CV (i.e., adds itself to the CV's waitlist), but does
not actually wait on the CV (IOW, it never calls ConditionVariableSleep()). It
still uses WaitEventSetWait() for waiting, because CV infrastructure doesn't
handle FeBe socket events currently. The processes (startup process,
walreceiver etc.) wanting to wake up walsenders use
ConditionVariableBroadcast(), which in turn calls SetLatch(), helping
walsenders come out of WaitEventSetWait().
We use separate shared memory CVs for physical and logical walsenders for
selective wake ups, see WalSndWakeup() for more details.
This approach is simple and reasonably efficient. But not very elegant. But
for 16 it seems to be a better path than a larger redesign of the CV
mechanism. A desirable future improvement would be to add support for CVs
into WaitEventSetWait().
This still leaves us with a small regression in very extreme workloads (due to
the spinlock acquisition in ConditionVariableBroadcast() when there are no
waiters) - but that seems acceptable.
Reported-by: Andres Freund <[email protected]>
Suggested-by: Andres Freund <[email protected]>
Author: Bharath Rupireddy <
[email protected]>
Reviewed-by: "Drouvot, Bertrand" <[email protected]>
Reviewed-by: Zhijie Hou <[email protected]>
Discussion: https://www.postgresql.org/message-id/
20230509190247.3rrplhdgem6su6cg%40awork3.anarazel.de
Bruce Momjian [Sun, 21 May 2023 15:55:50 +0000 (11:55 -0400)]
doc: PG 16 relnotes, add commits
Reported-by: Ian Lawrence Barwick
Discussion: https://postgr.es/m/CAB8KJ=hLvQDb53WdigV7OBEBh5tzQ7eA=pG0ZpccoRVBUNCXYA@mail.gmail.com
Tom Lane [Sun, 21 May 2023 15:21:19 +0000 (11:21 -0400)]
Doc: fix some rendering problems in the PDF docs build.
Tweak column widths in a couple of tables to avoid "contents
... exceed the available area" warnings. Remove usage of
some non-Latin-1 characters.
Discussion: https://postgr.es/m/614600.
1684537037@sss.pgh.pa.us
Tom Lane [Sun, 21 May 2023 14:55:18 +0000 (10:55 -0400)]
Expand some more uses of "deleg" to "delegation" or "delegated".
Complete the task begun in
9c0a0e2ed: we don't want to use the
abbreviation "deleg" for GSS delegation in any user-visible places.
(For consistency, this also changes most internal uses too.)
Abhijit Menon-Sen and Tom Lane
Discussion: https://postgr.es/m/949048.
1684639317@sss.pgh.pa.us
Nathan Bossart [Sun, 21 May 2023 03:18:51 +0000 (20:18 -0700)]
Fix remaining references to gss_accept_deleg.
These were missed in
9c0a0e2ed9.
Discussion: https://postgr.es/m/
20230521031757.GA3835667%40nathanxps13
Bruce Momjian [Sun, 21 May 2023 01:32:54 +0000 (21:32 -0400)]
rename "gss_accept_deleg" to "gss_accept_delegation".
This is more consistent with existing GUC spelling.
Discussion: https://postgr.es/m/
[email protected]
Bruce Momjian [Sun, 21 May 2023 01:03:13 +0000 (21:03 -0400)]
doc: PG 16 relnotes, misc updates
* document to_reg* accepting OIDs
* document pg_log_standby_snapshot()
* document pg_input_is_valid() and pg_input_error_info()
* handle rename of function to pg_split_walfile_name()
* fix character encoding problem for Przemyslaw Sztoch
* remove partition section
Reported-by: jian he, Tom Lane, Bertrand Drouvot
Tom Lane [Fri, 19 May 2023 21:28:04 +0000 (17:28 -0400)]
Add
0245f8db3 to .git-blame-ignore-revs.
Tom Lane [Fri, 19 May 2023 21:24:48 +0000 (17:24 -0400)]
Pre-beta mechanical code beautification.
Run pgindent, pgperltidy, and reformat-dat-files.
This set of diffs is a bit larger than typical. We've updated to
pg_bsd_indent 2.1.2, which properly indents variable declarations that
have multi-line initialization expressions (the continuation lines are
now indented one tab stop). We've also updated to perltidy version
20230309 and changed some of its settings, which reduces its desire to
add whitespace to lines to make assignments etc. line up. Going
forward, that should make for fewer random-seeming changes to existing
code.
Discussion: https://postgr.es/m/
20230428092545[email protected]
Tom Lane [Fri, 19 May 2023 20:43:57 +0000 (16:43 -0400)]
Make agreed-on updates in perltidy options.
Our standard version of perltidy is now
20230309.
Add a --valign-exclusion-list setting to reduce crosstalk
between nearby lines of Perl code.
Also, update instructions for running pgindent
(missed in
b16259b3c).
Discussion: https://postgr.es/m/
20230428092545[email protected]
Tom Lane [Fri, 19 May 2023 20:36:38 +0000 (16:36 -0400)]
Do pre-release housekeeping on catalog data.
Run renumber_oids.pl to move high-numbered OIDs down, as per pre-beta
tasks specified by RELEASE_CHANGES. For reference, the command was
./renumber_oids.pl --first-mapped-oid 8000 --target-oid 6200
Tom Lane [Fri, 19 May 2023 19:24:07 +0000 (15:24 -0400)]
Fix thinko in join removal.
In commit
9df8f903e I (tgl) switched join_is_removable() from
using the min relid sets of the join under consideration to
using its full syntactic relid sets. This was a mistake,
as it allowed join removal in cases where a reference to the
join output would survive in some syntactically-lower join
condition. Revert to the former coding.
Richard Guo
Discussion: https://postgr.es/m/CAMbWs4-EU9uBGSP7G-iTwLBhRQ=rnZKvFDhD+n+xhajokyPCKg@mail.gmail.com
Tom Lane [Fri, 19 May 2023 18:26:34 +0000 (14:26 -0400)]
Fix misbehavior of EvalPlanQual checks with multiple result relations.
The idea of EvalPlanQual is that we replace the query's scan of the
result relation with a single injected tuple, and see if we get a
tuple out, thereby implying that the injected tuple still passes the
query quals. (In join cases, other relations in the query are still
scanned normally.) This logic was not updated when commit
86dc90056
made it possible for a single DML query plan to have multiple result
relations, when the query target relation has inheritance or partition
children. We replaced the output for the current result relation
successfully, but other result relations were still scanned normally;
thus, if any other result relation contained a tuple satisfying the
quals, we'd think the EPQ check passed, even if it did not pass for
the injected tuple itself. This would lead to update or delete
actions getting performed when they should have been skipped due to
a conflicting concurrent update in READ COMMITTED isolation mode.
Fix by blocking all sibling result relations from emitting tuples
during an EvalPlanQual recheck. In the back branches, the fix is
complicated a bit by the need to not change the size of struct
EPQState (else we'd have ABI-breaking changes in offsets in
struct ModifyTableState). Like the back-patches of
3f7836ff6
and
4b3e37993, add a separately palloc'd struct to avoid that.
The logic is the same as in HEAD otherwise.
This is only a live bug back to v14 where
86dc90056 came in.
However, I chose to back-patch the test cases further, on the
grounds that this whole area is none too well tested. I skipped
doing so in v11 though because none of the test applied cleanly,
and it didn't quite seem worth extra work for a branch with only
six months to live.
Per report from Ante Krešić (via Aleksander Alekseev)
Discussion: https://postgr.es/m/CAJ7c6TMBTN3rcz4=AjYhLPD_w3FFT0Wq_C15jxCDn8U4tZnH1g@mail.gmail.com
Alvaro Herrera [Fri, 19 May 2023 18:19:28 +0000 (20:19 +0200)]
psql: Tweak xheader_width and pager_min_lines input parsing
Don't throw away the previous value when an invalid value is proposed.
Discussion: https://postgr.es/m/
20230519110205[email protected]
Peter Eisentraut [Fri, 19 May 2023 16:45:29 +0000 (18:45 +0200)]
Message style improvements
Bruce Momjian [Fri, 19 May 2023 16:29:50 +0000 (12:29 -0400)]
doc: PG 16 relnotes, delete reverted grant, adjust version num
Reported-by: Nathan Bossart, Sehrope Sarkuni
Tomas Vondra [Fri, 19 May 2023 14:31:11 +0000 (16:31 +0200)]
Allocate hash join files in a separate memory context
Should a hash join exceed memory limit, the hashtable is split up into
multiple batches. The number of batches is doubled each time a given
batch is determined not to fit in memory. Each batch file is allocated
with a block-sized buffer for buffering tuples and parallel hash join
has additional sharedtuplestore accessor buffers.
In some pathological cases requiring a lot of batches, often with skewed
data, bad stats, or very large datasets, users can run out-of-memory
solely from the memory overhead of all the batch files' buffers.
Batch files were allocated in the ExecutorState memory context, making
it very hard to identify when this batch explosion was the source of an
OOM. This commit allocates the batch files in a dedicated memory
context, making it easier to identify the cause of an OOM and work to
avoid it.
Based on initial draft by Tomas Vondra, with significant reworks and
improvements by Jehan-Guillaume de Rorthais.
Author: Jehan-Guillaume de Rorthais <
[email protected]>
Author: Tomas Vondra <
[email protected]>
Reviewed-by: Melanie Plageman <[email protected]>
Discussion: https://postgr.es/m/
20190421114618.z3mpgmimc3rmubi4@development
Discussion: https://postgr.es/m/
20230504193006.
1b5b9622%40karst#
273020ff4061fc7a2fbb1ba96b281f17
Tomas Vondra [Fri, 19 May 2023 14:19:54 +0000 (16:19 +0200)]
Describe hash join implementation
Add a high level description of our implementation of the hybrid hash
join algorithm to the block comment in nodeHashjoin.c.
Author: Melanie Plageman <
[email protected]>
Reviewed-by: Tomas Vondra <[email protected]>
Reviewed-by: Jehan-Guillaume de Rorthais <[email protected]>
Discussion: https://postgr.es/m/
20230516160051.
4267a800%40karst
Tom Lane [Fri, 19 May 2023 14:57:46 +0000 (10:57 -0400)]
Avoid naming conflict between transactions.sql and namespace.sql.
Commits
681d9e462 et al added a test case in namespace.sql that
implicitly relied on there not being a table "public.abc".
However, the concurrently-run transactions.sql test creates precisely
such a table, so with the right timing you'd get a failure.
Creating a table named as generically as "abc" in a common schema
seems like bad practice, so fix this by changing the name of
transactions.sql's table. (Compare
2cf8c7aa4.)
Marina Polyakova
Discussion: https://postgr.es/m/
80d0201636665d82185942e7112257b4@postgrespro.ru
Peter Eisentraut [Fri, 19 May 2023 13:57:38 +0000 (15:57 +0200)]
Remove stray mid-sentence tabs in comments
Bruce Momjian [Fri, 19 May 2023 13:30:00 +0000 (09:30 -0400)]
doc: PG 16 relnotes, merge REINDEX and reindexdb items
Reported-by: Vibhor Kumar
Bruce Momjian [Fri, 19 May 2023 12:31:11 +0000 (08:31 -0400)]
doc: PG 16 relnotes, add missing parentheses
Reported-by: Hans Buschmann
Peter Eisentraut [Fri, 19 May 2023 11:42:06 +0000 (13:42 +0200)]
Move mdwriteback() to better place
The previous order in the file didn't make sense and matched neither
the header file nor the smgr API.
Discussion: https://www.postgresql.org/message-id/flat/
22fed8ba-01c3-2008-a256-
4ea912d68fab%40enterprisedb.com
Peter Eisentraut [Fri, 19 May 2023 08:52:04 +0000 (10:52 +0200)]
Reindent some comments
Most (older) comments in md.c and smgr.c are indented with a leading
tab on all lines, which isn't the current style and makes updating the
comments a bit annoying. This reindents all these lines with a single
space, as is the normal style. This issue exists in various shapes
throughout the code but it's pretty consistent here, and since there
is a patch pending to refresh some of the comments in these files, it
seems sensible to clean this up here separately.
Discussion: https://www.postgresql.org/message-id/flat/
22fed8ba-01c3-2008-a256-
4ea912d68fab%40enterprisedb.com
Michael Paquier [Fri, 19 May 2023 03:37:58 +0000 (12:37 +0900)]
pageinspect: Fix gist_page_items() with included columns
Non-leaf pages of GiST indexes contain key attributes, leaf pages
contain both key and non-key attributes, and gist_page_items() ignored
the handling of non-key attributes. This caused a few problems when
using gist_page_items() on a GiST index with INCLUDE:
- On a non-leaf page, the function would crash.
- On a leaf page, the function would work, but miss to display all the
values for included attributes.
This commit fixes gist_page_items() to handle such cases in a more
appropriate way, and now displays the values of key and non-key
attributes for each item separately in a style consistent with what
ruleutils.c would generate for the attribute list, depending on the page
type dealt with. In a way similar to how a record is displayed, values
would be double-quoted for key or non-key attributes if required.
ruleutils.c did not provide a routine able to control if non-key
attributes should be displayed, so an extended() routine for index
definitions is added to work around the leaf and non-leaf page
differences.
While on it, this commit fixes a third problem related to the amount of
data reported for key attributes. The code originally relied on
BuildIndexValueDescription() (used for error reports on constraints)
that would not print all the data stored in the index but the index
opclass's input type, so this limited the amount of information
available. This switch makes gist_page_items() much cheaper as there is
no need to run ACL checks for each item printed, which is not an issue
anyway as superuser rights are required to execute the functions of
pageinspect. Opclasses whose data cannot be displayed can rely on
gist_page_items_bytea().
The documentation of this function was slightly incorrect for the
output results generated on HEAD and v15, so adjust it on these
branches.
Author: Alexander Lakhin, Michael Paquier
Discussion: https://postgr.es/m/17884-
cb8c326522977acb@postgresql.org
Backpatch-through: 14
Bruce Momjian [Fri, 19 May 2023 03:17:03 +0000 (23:17 -0400)]
doc: improve pg_walinspect and role membership items
Reported-by: Peter Geoghegan
Bruce Momjian [Fri, 19 May 2023 02:22:17 +0000 (22:22 -0400)]
doc: improve description of adding roles as members
Discussion: https://postgr.es/m/
[email protected]
Bruce Momjian [Fri, 19 May 2023 01:44:22 +0000 (21:44 -0400)]
doc: update PG 16 relnotes for pg_walinspect changes
Reported-by: Peter Geoghegan
Tomas Vondra [Thu, 18 May 2023 23:53:30 +0000 (01:53 +0200)]
Show empty BRIN ranges in brin_page_items
Commit
3581cbdcd6 added a flag to identify empty BRIN ranges. This adds
the new flag to brin_page_items() output.
This is kept as a separate commit as it should not be backpatched.
Reviewed-by: Justin Pryzby, Matthias van de Meent, Alvaro Herrera
Discussion: https://postgr.es/m/
402430e4-7d9d-6cf1-09ef-
464d80afff3b@enterprisedb.com
Tomas Vondra [Thu, 18 May 2023 22:00:22 +0000 (00:00 +0200)]
Fix handling of empty ranges and NULLs in BRIN
BRIN indexes did not properly distinguish between summaries for empty
(no rows) and all-NULL ranges, treating them as essentially the same
thing. Summaries were initialized with allnulls=true, and opclasses
simply reset allnulls to false when processing the first non-NULL value.
This however produces incorrect results if the range starts with a NULL
value (or a sequence of NULL values), in which case we forget the range
contains NULL values when adding the first non-NULL value.
This happens because the allnulls flag is used for two separate
purposes - to mark empty ranges (not representing any rows yet) and
ranges containing only NULL values.
Opclasses don't know which of these cases it is, and so don't know
whether to set hasnulls=true. Setting the flag in both cases would make
it correct, but it would also make BRIN indexes useless for queries with
IS NULL clauses. All ranges start empty (and thus allnulls=true), so all
ranges would end up with either allnulls=true or hasnulls=true.
The severity of the issue is somewhat reduced by the fact that it only
happens when adding values to an existing summary with allnulls=true.
This can happen e.g. for small tables (because a summary for the first
range exists for all BRIN indexes), or for tables with large fraction of
NULL values in the indexed columns.
Bulk summarization (e.g. during CREATE INDEX or automatic summarization)
that processes all values at once is not affected by this issue. In this
case the flags were updated in a slightly different way, not forgetting
the NULL values.
To identify empty ranges we use a new flag, stored in an unused bit in
the BRIN tuple header so the on-disk format remains the same. A matching
flag is added to BrinMemTuple, into a 3B gap after bt_placeholder.
That means there's no risk of ABI breakage, although we don't actually
pass the BrinMemTuple to any public API.
We could also skip storing index tuples for empty summaries, but then
we'd have to always process such ranges - even if there are no rows in
large parts of the table (e.g. after a bulk DELETE), it would still
require reading the pages etc. So we store them, but ignore them when
building the bitmap.
Backpatch to 11. The issue exists since BRIN indexes were introduced in
9.5, but older releases are already EOL.
Backpatch-through: 11
Reviewed-by: Justin Pryzby, Matthias van de Meent, Alvaro Herrera
Discussion: https://postgr.es/m/
402430e4-7d9d-6cf1-09ef-
464d80afff3b@enterprisedb.com
Bruce Momjian [Thu, 18 May 2023 22:53:28 +0000 (18:53 -0400)]
doc: PG 16 relnotes, add freeze and update walinspect items
Reported-by: Peter Geoghegan
Bruce Momjian [Thu, 18 May 2023 22:25:21 +0000 (18:25 -0400)]
docs: re-order some PG 16 relnotes items
Reported-by: Jonathan Katz
Bruce Momjian [Thu, 18 May 2023 21:42:47 +0000 (17:42 -0400)]
doc: more PG 16 relnote wording improvements
Tomas Vondra [Thu, 18 May 2023 11:00:31 +0000 (13:00 +0200)]
Fix handling of NULLs when merging BRIN summaries
When merging BRIN summaries, union_tuples() did not correctly update the
target hasnulls/allnulls flags. When merging all-NULL summary into a
summary without any NULL values, the result had both flags set to false
(instead of having hasnulls=true).
This happened because the code only considered the hasnulls flags,
ignoring the possibility the source summary has allnulls=true.
Discovered while investigating issues with handling empty BRIN ranges
and handling of NULL values, but it's a separate problem (has nothing to
do with empty ranges).
Fixed by considering both flags on the source summary, and updating the
hasnulls flag on the target summary.
Backpatch to 11. The bug exists since 9.5 (where BRIN indexes were
introduced), but those releases are EOL already.
Discussion: https://postgr.es/m/
9d993d0d-e431-2196-9ccc-
0554d0e60154%40enterprisedb.com
Bruce Momjian [Thu, 18 May 2023 20:33:05 +0000 (16:33 -0400)]
doc: add more sections to PG 16 release notes
Bruce Momjian [Thu, 18 May 2023 20:00:04 +0000 (16:00 -0400)]
doc: first draft of the PG 16 release notes
Jeff Davis [Thu, 18 May 2023 18:04:30 +0000 (11:04 -0700)]
Fix documentation build broken by
1e16af8ab5.
Jeff Davis [Thu, 18 May 2023 17:37:55 +0000 (10:37 -0700)]
Doc improvements for language tags and custom ICU collations.
Separate the documentation for language tags themselves from the
available collation settings which can be included in a language tag.
Include tables of the available options, more details about the
effects of each option, and additional examples.
Also include an explanation of the "levels" of textual features and
how they relate to collation.
Discussion: https://postgr.es/m/
25787ec7-4c04-9a8a-d241-
4dc9be0b1ba3@postgresql.org
Reviewed-by: Jonathan S. Katz
Tom Lane [Thu, 18 May 2023 14:39:16 +0000 (10:39 -0400)]
Tweak API of new function clause_is_computable_at().
Pass it the RestrictInfo under consideration, not just the
clause_relids. This should save some trivial amount of
code at the call sites, and it gives us more flexibility
about what clause_is_computable_at() does. There's no
actual functional change here, though.
Discussion: https://postgr.es/m/
3564467.
1684352557@sss.pgh.pa.us
Jeff Davis [Wed, 17 May 2023 20:43:41 +0000 (13:43 -0700)]
ICU: check for U_STRING_NOT_TERMINATED_WARNING.
Fixes memory error in cases where the length of the language name
returned by uloc_getLanguage() is exactly ULOC_LANG_CAPACITY, in which
case the status is set to U_STRING_NOT_TERMINATED_WARNING.
Also check in call sites for other ICU functions that are expected to
return a C string to be safe (no bug is known at these other call
sites).
Reported-by: Alexander Lakhin
Discussion: https://postgr.es/m/
2098874d-c111-41e4-9063-
30bcf135226b@gmail.com
Jeff Davis [Wed, 17 May 2023 20:18:40 +0000 (13:18 -0700)]
Reduce icu_validation_level default to WARNING.
Discussion: https://postgr.es/m/
daa9f060aa2349ebc84444515efece49e7b32c5d[email protected]
Peter Eisentraut [Wed, 17 May 2023 19:33:13 +0000 (21:33 +0200)]
Fix error message wordings
The original patch for percentrepl.c
c96de2ce17 adopted the error
messages from basebackup_to_shell, but that uses terminology that
doesn't really fit with the new API naming.
Andres Freund [Wed, 17 May 2023 18:18:35 +0000 (11:18 -0700)]
Add writeback to pg_stat_io
28e626bde00 added the concept of IOOps but neglected to include writeback
operations.
ac8d53dae5 added time spent doing these I/O operations. Without
counting writeback, checkpointer write time in the log often differed
substantially from that in pg_stat_io. To fix this, add IOOp IOOP_WRITEBACK
and track writeback in pg_stat_io.
Bumps catversion.
Author: Melanie Plageman <
[email protected]>
Reviewed-by: Kyotaro Horiguchi <[email protected]>
Reported-by: Andres Freund <[email protected]>
Discussion: https://postgr.es/m/
20230419172326.dhgyo4wrrhulovt6%40awork3.anarazel.de
Andres Freund [Wed, 17 May 2023 18:17:12 +0000 (11:17 -0700)]
Update parameter name context to wb_context
For clarity of review, renaming the function parameter "context" in
ScheduleBufferTagForWriteback() and IssuePendingWritebacks() to
"wb_context" is a separate commit. The next commit adds an "io_context"
parameter and "wb_context" makes it more clear which is which.
Author: Melanie Plageman <
[email protected]>
Discussion: https://postgr.es/m/CAAKRu_acc6iL4M3hvOTeztf_ZPpsB3Pqio5aVHgZ5q=Pi3BZKg@mail.gmail.com
Andres Freund [Wed, 17 May 2023 18:17:02 +0000 (11:17 -0700)]
Use BUFFER_USAGE_LIMIT to reduce needed test table size
Using the minimum BUFFER_USAGE_LIMIT value, we can make one of the pg_stat_io
test tables smaller while still causing reuses.
Author: Melanie Plageman <
[email protected]>
Discussion: https://postgr.es/m/CAAKRu_acc6iL4M3hvOTeztf_ZPpsB3Pqio5aVHgZ5q=Pi3BZKg@mail.gmail.com
Alexander Korotkov [Wed, 17 May 2023 17:06:50 +0000 (20:06 +0300)]
Revert "Add USER SET parameter values for pg_db_role_setting"
This reverts commit
096dd80f3ccc and its fixups
beecbe8e5001,
afdd9f7f0e00,
529da086ba,
db93e739ac61.
Catversion is bumped.
Discussion: https://postgr.es/m/
d46f9265-ff3c-6743-2278-
6772598233c2%40pgmasters.net
Alvaro Herrera [Wed, 17 May 2023 17:13:08 +0000 (19:13 +0200)]
pg_dump: Error message improvements
Remove spurious semicolon from one error message, and print the
offending value of a parameter reported as invalid in another.
Alvaro Herrera [Wed, 17 May 2023 16:55:51 +0000 (18:55 +0200)]
pg_dump: Have _EndLO report errno after CFH->write_func() failure
Other callers of that function do things this way, but this one didn't
get the memo.
Tom Lane [Wed, 17 May 2023 15:46:15 +0000 (11:46 -0400)]
Track tlist_vinfo.varnullingrels even in non-Assert builds.
Oversight in commit
867be9c07 (which should get reverted
along with that, if we ever do revert it). Per buildfarm.
Tom Lane [Wed, 17 May 2023 15:13:52 +0000 (11:13 -0400)]
Fix some issues with improper placement of outer join clauses.
After applying outer-join identity 3 in the forward direction,
it was possible for the planner to mistakenly apply a qual clause
from above the two outer joins at the now-lower join level.
This can give the wrong answer, since a value that would get nulled
by the now-upper join might not yet be null.
To fix, when we perform such a transformation, consider that the
now-lower join hasn't really completed the outer join it's nominally
responsible for and thus its relid set should not include that OJ's
relid (nor should its output Vars have that nullingrel bit set).
Instead we add those bits when the now-upper join is performed.
The existing rules for qual placement then suffice to prevent
higher qual clauses from dropping below the now-upper join.
There are a few complications from needing to consider transitive
closures in case multiple pushdowns have happened, but all in all
it's not a very complex patch.
This is all new logic (from
2489d76c4) so no need to back-patch.
The added test cases all have the same results as in v15.
Tom Lane and Richard Guo
Discussion: https://postgr.es/m/
0b819232-4b50-f245-1c7d-
c8c61bf41827@postgrespro.ru
Tom Lane [Wed, 17 May 2023 14:50:35 +0000 (10:50 -0400)]
Convert nullingrels match checks from Asserts to test-and-elog.
It seems like the code that these checks are backstopping may have
a few bugs left in it. Use a test-and-elog so that the tests are
performed even in non-assert builds, and so that we get something
more informative than "server closed the connection" on failure.
Committed separately with the idea that eventually we'll revert
this. It might be awhile though.
Discussion: https://postgr.es/m/
3014965.
1684293045@sss.pgh.pa.us
Tomas Vondra [Wed, 17 May 2023 14:49:31 +0000 (16:49 +0200)]
Advance input pointer when LZ4 compressing data
LZ4File_write() did not advance the input pointer on subsequent invocations of
LZ4F_compressUpdate(). As a result the generated compressed output would be a
compressed version of the same input chunk.
Tests failed to catch this error because the data would comfortably fit
within the default buffer size, as a single chunk. Tests have been added
to provide adequate coverage of multi-chunk compression.
WriteDataToArchiveLZ4() which is also using LZ4F_compressUpdate() did
not suffer from this omission.
Author: Georgios Kokolatos <
[email protected]>
Reported-by: Michael Paquier <[email protected]>
Discussion: https://postgr.es/m/ZFhCyn4Gm2eu60rB%40paquier.xyz
Tomas Vondra [Wed, 17 May 2023 14:35:17 +0000 (16:35 +0200)]
Null-terminate the output buffer of LZ4Stream_gets
LZ4Stream_gets did not null-terminate its output buffer. The callers expected
the buffer to be null-terminated and passed it around to functions such as
sscanf with unintended consequences.
Author: Georgios Kokolatos <
[email protected]>
Reported-by: Alexander Lakhin <[email protected]>
Discussion: https://postgr.es/m/
94ae9bca-5ebb-1e68-bb7b-
4f32e89fefbe@gmail.com
Michael Paquier [Wed, 17 May 2023 01:19:17 +0000 (10:19 +0900)]
Add back SQLValueFunction for SQL keywords
This is equivalent to a revert of
f193883 and
fb32748, with the addition
that the declaration of the SQLValueFunction node needs to gain a couple
of node_attr for query jumbling. The performance impact of removing the
function call inlining is proving to be too huge for some workloads
where these are used. A worst-case test case of involving only simple
SELECT queries with a SQL keyword is proving to lead to a reduction of
10% in TPS via pgbench and prepared queries on a high-end machine.
None of the tests I ran back for this set of changes saw such a huge
gap, but Alexander Lakhin and Andres Freund have found that this can be
noticeable. Keeping the older performance would mean to do more
inlining in the executor when using COERCE_SQL_SYNTAX for a function
expression, similarly to what SQLValueFunction does. This requires more
redesign work and there is little time until 16beta1 is released, so for
now reverting the change is the best way forward, bringing back the
previous performance.
Bump catalog version.
Reported-by: Alexander Lakhin
Discussion: https://postgr.es/m/
b32bed1b-0746-9b20-1472-
4bdc9ca66d52@gmail.com
Tom Lane [Tue, 16 May 2023 14:53:42 +0000 (10:53 -0400)]
Ensure Soundex difference() function handles empty input sanely.
fuzzystrmatch's difference() function assumes that _soundex()
always initializes its output buffer fully. This was not so for
the case of a string containing no alphabetic characters, resulting
in unstable output and Valgrind complaints.
Fix by using memset() to fill the whole buffer in the early-exit
case. Also make some cosmetic improvements (I didn't care for the
random switches between "instr[0]" and "*instr" notation).
Report and diagnosis by Alexander Lakhin (bug #17935).
Back-patch to all supported branches.
Discussion: https://postgr.es/m/17935-
b99316aa79c18513@postgresql.org
Alvaro Herrera [Tue, 16 May 2023 09:50:08 +0000 (11:50 +0200)]
libpq: Error message improvement
Move a variable name out of the translatable message, to make it
identical to others.
Alvaro Herrera [Tue, 16 May 2023 09:47:25 +0000 (11:47 +0200)]
Mark internal messages as no longer translatable
The problem that these messages protect against can only occur because
a corrupted hash spill file was written, i.e., a Postgres bug. There's
no reason to have them as translatable.
Backpatch to 15, where these messages were changed by commit
c4649cce39a4.
Reviewed-by: Daniel Gustafsson <[email protected]>
Discussion: https://postgr.es/m/
20230510175407[email protected]
Daniel Gustafsson [Tue, 16 May 2023 08:49:48 +0000 (10:49 +0200)]
Fix reported runtime for single tests in pg_regress
Commit
558fff0adf got the order of the parameters to test_status_failed
mixed up which resulted in the runtime being reported as 0 ms. Fix by
changing the order to the correct one. No backpatching is needed since
this has not been shipped in a release yet.
Discussion: https://postgr.es/m/
0134C9EC-5F6B-4EAC-B2D5-
BB4249BEBD4D@yesql.se
Michael Paquier [Tue, 16 May 2023 07:51:44 +0000 (16:51 +0900)]
doc: Fix ordering of entries in wait event table for I/O type
Issue spotted while reviewing a different patch touching this area.
Discussion: https://postgr.es/m/
[email protected]
Peter Eisentraut [Tue, 16 May 2023 06:59:34 +0000 (08:59 +0200)]
libpq: Error message improvement
Peter Eisentraut [Tue, 16 May 2023 04:13:59 +0000 (06:13 +0200)]
psql: Adjust capitalization of table heading
for consistency with surrounding headings
Michael Paquier [Mon, 15 May 2023 07:02:07 +0000 (16:02 +0900)]
doc: Fix incorrect version list for Windows SDK
References to SDK 8.1a should have been cleaned up in
495ed0e, as only
version 10 and above are supported with Windows 10 and newer versions.
Reported-by: Thomas Munro
Discussion: https://postgr.es/m/CA+hUKG++qE9uVjQaTTD7oaC8a2T3h8K50=Eqyx9uUZvOHa__ww@mail.gmail.com
Peter Eisentraut [Mon, 15 May 2023 05:07:09 +0000 (07:07 +0200)]
Fix incorrect TAP test ordering
Using a test function before a possible skip_all is incorrect. If the
skip_all is called, the test output will become incorrect and the test
file will fail.
a4f23f9b3c introduced a new test before skip_all. After discussion,
this doesn't really need to be a test. Instead, we just bail out if
the condition is not satisfied.
Discussion: https://www.postgresql.org/message-id/
af5567a1-aea6-fbdb-7e4b-
d1e23a43c43b@enterprisedb.com
Peter Eisentraut [Mon, 15 May 2023 04:27:40 +0000 (06:27 +0200)]
Fix whitespace inconsistencies
Thomas Munro [Sun, 14 May 2023 22:45:19 +0000 (10:45 +1200)]
Fix wal_writer_flush_after initializer value.
Commit
a73952b7956 (new in 16) required default values in guc_table.c
and C variable initializers to match. This one only matched when
XLOG_BLCKSZ == 8kB. Fix by using the same expression in both places
with a new DEFAULT_XXX macro, as done for other GUCs.
Reviewed-by: Andres Freund <[email protected]>
Discussion: https://postgr.es/m/CA+hUKGLNmLV=VrT==5MqnbARgx2ifRSFtdd8ofdfrdSLL3yv5A@mail.gmail.com
Thomas Munro [Sun, 14 May 2023 22:31:14 +0000 (10:31 +1200)]
Rename io_direct to debug_io_direct.
Give the new GUC introduced by
d4e71df6 a name that is clearly not
intended for mainstream use quite yet.
Future proposals would drop the prefix only after adding infrastructure
to make it efficient. Having the switch in the tree sooner is good
because it might lead to new discoveries about the hazards awaiting us
on a wide range of systems, but that name was too enticing and could
lead to cross-version confusion in future, per complaints from Noah and
Justin.
Suggested-by: Noah Misch <[email protected]>
Reviewed-by: Noah Misch <[email protected]>
Reviewed-by: Justin Pryzby <[email protected]> (the idea, not the patch)
Reviewed-by: Tom Lane <[email protected]> (ditto)
Discussion: https://postgr.es/m/
20230430041106.GA2268796%40rfd.leadboat.com
Nathan Bossart [Fri, 12 May 2023 20:50:46 +0000 (13:50 -0700)]
Improve error message for pg_create_subscription.
c3afe8cf5a updated this error message, but it didn't use the new
style established in
de4d456b40.
Reviewed-by: Tom Lane
Discussion: https://postgr.es/m/
20230512203721.GA2644063%40nathanxps13.home
Tom Lane [Fri, 12 May 2023 20:11:14 +0000 (16:11 -0400)]
Tighten usage of PSQL_WATCH_PAGER.
Don't use PSQL_WATCH_PAGER when stdin/stdout are not a terminal.
This corresponds to the restrictions on when other commands will
use [PSQL_]PAGER. There isn't a lot of sense in trying to use a
pager in non-interactive cases, and doing so allows an environment
setting to break our tests.
Also, ignore PSQL_WATCH_PAGER if it is set but empty or all-blank,
for the same reasons we ignore such settings of [PSQL_]PAGER (see
commit
18f8f784c).
No documentation change is really needed, since there is nothing
suggesting that these constraints on [PSQL_]PAGER didn't already
apply to PSQL_WATCH_PAGER too. But I rearranged the text
a little to make it read more naturally (IMHO anyway).
Per report from Pavel Stehule. Back-patch to v15 where
PSQL_WATCH_PAGER was introduced.
Discussion: https://postgr.es/m/CAFj8pRDTwFzmEWdA-gdAcUh0ZnxUioSfTMre71WyB_wNJy-8gw@mail.gmail.com
Peter Eisentraut [Fri, 12 May 2023 07:45:50 +0000 (09:45 +0200)]
initdb: Set collversion for standard collation UNICODE
Since the behavior of the UNICODE collation can change with new
ICU/Unicode versions, we need to apply the versioning mechanism to it.
We do this with an UPDATE command in initdb; this is similar to how we
put the collation version into pg_database already.
Reported-by: Daniel Verite <[email protected]>
Discussion: https://www.postgresql.org/message-id/
49417853-7bdd-4b23-a4e9-
04c7aff33821@manitou-mail.org
Tom Lane [Thu, 11 May 2023 17:44:25 +0000 (13:44 -0400)]
Undo faulty attempt at not relying on RINFO_IS_PUSHED_DOWN.
I've had a bee in my bonnet for some time about getting rid of
RestrictInfo.is_pushed_down, because it's squishily defined and
requires not-inexpensive extra tests to use (cf RINFO_IS_PUSHED_DOWN).
In commit
2489d76c4, I tried to make remove_rel_from_query() not
depend on that macro; but the replacement test is buggy,
as exposed by a report from Rushabh Lathia and Robert Haas.
That change was pretty incidental to the main goal of
2489d76c4,
so let's just revert it for now. (Getting rid of is_pushed_down
is still far away, anyway.)
Discussion: https://postgr.es/m/CA+TgmoYco=hmg+iX1CW9Y1_CzNoSL81J03wUG-d2_3=rue+L2A@mail.gmail.com
Peter Geoghegan [Wed, 10 May 2023 17:49:50 +0000 (10:49 -0700)]
Doc: Fix link to fillfactor reloption.
Fix a link from the "Heap-Only Tuples" documentation section.
Previously, its "fillfactor" link pointed to the "CREATE TABLE"
command's documentation. Now the link directly points to the fillfactor
storage parameter documentation (which is about half way into the
"CREATE TABLE" sect1).
Oversight in commit
115464bb.
Backpatch: 12-, the first version with a usable reloption link.
Alvaro Herrera [Wed, 10 May 2023 16:26:10 +0000 (18:26 +0200)]
Fix publication syntax error message
There was some odd wording in corner-case gram.y error messages "some
error ... at or near", which appears to have been modeled after "syntax
error" messages. However, they don't work that way, and they're just
wrong. They're also uncovered by tests. Remove the trailing words,
and also add tests.
They were introduced with
5a2832465fd8; backpatch to 15.
Author: Álvaro Herrera <
[email protected]>
Peter Eisentraut [Wed, 10 May 2023 11:51:51 +0000 (13:51 +0200)]
Add missing gettext triggers
due to the changes in commit
dac048f71e
Michael Paquier [Wed, 10 May 2023 02:24:30 +0000 (11:24 +0900)]
Fix assertion failure when updating stats_fetch_consistency in a transaction
An update of the GUC stats_fetch_consistency in a transaction would be
able to trigger an assertion when doing cache->snapshot. In this case,
when retrieving a pgstat entry after the switch, a new snapshot would be
rebuilt, confusing pgstat_build_snapshot() because a snapshot is already
cached with an unexpected mode ("cache").
In order to fix this problem, this commit adds a flag to force a
snapshot clear each time this GUC is changed. Some tests are added to
check, while on it.
Some optimizations in avoiding the snapshot clear should be possible
depending on what is cached and the current GUC value, I guess, but this
solution is simple, and ensures that the state of the cache is updated
each time a new pgstat entry is fetched, hence being consistent with the
level wanted by the client that has set the GUC.
Note that cache->none and snapshot->none would not cause issues, as
fetching a pgstat entry would be retrieved from shared memory on the
second attempt, however a snapshot would still be cached. Similarly,
none->snapshot and none->cache would build a new snapshot on the second
fetch attempt. Finally, snapshot->cache would cache a new snapshot on
the second attempt.
Reported-by: Alexander Lakhin
Author: Kyotaro Horiguchi
Discussion: https://postgr.es/m/17804-
2a118cd046f2d0e5@postgresql.org
backpatch-through: 15
Michael Paquier [Wed, 10 May 2023 01:19:57 +0000 (10:19 +0900)]
Document values of stats_fetch_consistency in postgresql.conf.sample
Issue noted while looking at a patch related to that.
Discussion: https://postgr.es/m/ZE9LiFc7JdNHokz/@paquier.xyz