Thomas Munro [Thu, 14 Jul 2022 13:23:29 +0000 (01:23 +1200)]
Don't clobber postmaster sigmask in dsm_impl_resize.
Commit
4518c798 intended to block signals in regular backends that
allocate DSM segments, but dsm_impl_resize() is also reached by
dsm_postmaster_startup(). It's not OK to clobber the postmaster's
signal mask, so only manipulate the signal mask when under the
postmaster.
Back-patch to all releases, like
4518c798.
Discussion: https://postgr.es/m/CA%2BhUKGKNpK%3D2OMeea_AZwpLg7Bm4%3DgYWk7eDjZ5F6YbozfOf8w%40mail.gmail.com
Tom Lane [Thu, 14 Jul 2022 13:04:23 +0000 (09:04 -0400)]
Tighten up parsing logic in gen_node_support.pl.
Teach this script to handle function pointer fields honestly.
Previously they were just silently ignored, but that's not likely to
be a behavior we can accept indefinitely. This mostly entails fixing
it so that a field declaration spanning multiple lines can be parsed,
because we have a bunch of such fields that're laid out that way.
But that's a good improvement in its own right.
With that change and a minor regex adjustment, the only struct it
fails to parse in the node-defining headers is A_Const, because
of the embedded union. The path of least resistance is to move
that union declaration outside the struct.
Having done those things, we can make it error out if it finds
any within-struct syntax it doesn't understand, which seems like
a pretty important property for robustness.
This commit doesn't change the output files at all; it's just in
the way of future-proofing.
Discussion: https://postgr.es/m/
2593369.
1657759779@sss.pgh.pa.us
Thomas Munro [Tue, 12 Jul 2022 04:17:36 +0000 (16:17 +1200)]
Avoid shadowing a variable in sync.c.
It was confusing to reuse the variable name 'entry' in two scopes.
Use distinct variable names.
Reported-by: Ranier Vilela <ranier.vf@gmail.com>
Reported-by: Tom Lane <tgl@sss.pgh.pa.us>
Reported-by: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
Discussion: https://postgr.es/m/CAEudQArDrFyQ15Am3rgWBunGBVZFDb90onTS8SRiFAWHeiLiFA%40mail.gmail.com
Thomas Munro [Thu, 14 Jul 2022 10:52:13 +0000 (22:52 +1200)]
Create a distinct wait event for POSIX DSM allocation.
Previously we displayed "DSMFillZeroWrite" while in posix_fallocate(),
because we shared the same wait event for "mmap" and "posix" DSM types.
Let's introduce a new wait event "DSMAllocate", to be more accurate.
Reported-by: Andres Freund <andres@anarazel.de>
Discussion: https://postgr.es/m/
20220711174518.yldckniicknsxgzl%40awork3.anarazel.de
Thomas Munro [Thu, 14 Jul 2022 07:23:35 +0000 (19:23 +1200)]
Remove redundant ftruncate() for POSIX DSM memory.
In early releases of the DSM infrastructure, it was possible to resize
segments. That was removed in release 12 by commit
3c60d0fa. Now the
ftruncate() + posix_fallocate() sequence during DSM segment creation has
a redundant step: we're always extending from zero to the desired size,
so we might as well just call posix_fallocate().
Let's also include the remaining ftruncate() call (non-Linux POSIX
systems) in the wait event reporting, for good measure.
Discussion: https://postgr.es/m/CA%2BhUKGJSm-nq8s%2B_59zb7NbFQF-OS%3DxTnTAiGLrQpuSmU2y_1A%40mail.gmail.com
Thomas Munro [Wed, 13 Jul 2022 04:16:07 +0000 (16:16 +1200)]
Block signals while allocating DSM memory.
On Linux, we call posix_fallocate() on shm_open()'d memory to avoid
later potential SIGBUS (see commit
899bd785).
Based on field reports of systems stuck in an EINTR retry loop there,
there, we made it possible to break out of that loop via slightly odd
coding where the CHECK_FOR_INTERRUPTS() call was somewhat removed from
the loop (see commit
422952ee).
On further reflection, that was not a great choice for at least two
reasons:
1. If interrupts were held, the CHECK_FOR_INTERRUPTS() would do nothing
and the EINTR error would be surfaced to the user.
2. If EINTR was reported but neither QueryCancelPending nor
ProcDiePending was set, then we'd dutifully retry, but with a bit more
understanding of how posix_fallocate() works, it's now clear that you
can get into a loop that never terminates. posix_fallocate() is not a
function that can do some of the job and tell you about progress if it's
interrupted, it has to undo what it's done so far and report EINTR, and
if signals keep arriving faster than it can complete (cf recovery
conflict signals), you're stuck.
Therefore, for now, we'll simply block most signals to guarantee
progress. SIGQUIT is not blocked (see InitPostmasterChild()), because
its expected handler doesn't return, and unblockable signals like
SIGCONT are not expected to arrive at a high rate. For good measure,
we'll include the ftruncate() call in the blocked region, and add a
retry loop.
Back-patch to all supported releases.
Reported-by: Alvaro Herrera <alvherre@alvh.no-ip.org>
Reported-by: Nicola Contu <nicola.contu@gmail.com>
Reviewed-by: Alvaro Herrera <alvherre@alvh.no-ip.org>
Reviewed-by: Andres Freund <andres@anarazel.de>
Discussion: https://postgr.es/m/
20220701154105.jjfutmngoedgiad3%40alvherre.pgsql
John Naylor [Thu, 14 Jul 2022 02:38:06 +0000 (09:38 +0700)]
Correct some uses of e.g. and i.e. in message strings and documentation
E.g. means "for example" and i.e. means "that is". Fix a couple uses
that don't match the intended meaning.
Kyotaro Horiguchi
Reviewed by Junwang Zhao and Aleksander Alekseev, with one addition by me
Discussion: https://www.postgresql.org/message-id/flat/
20220713.180943.
589079824955875739.horikyota.ntt%40gmail.com
Michael Paquier [Thu, 14 Jul 2022 02:22:49 +0000 (11:22 +0900)]
Remove support for Visual Studio 2013
No members of the buildfarm are using this version of Visual Studio,
resulting in all the code cleaned up here as being mostly dead, and
VS2017 is the oldest version still supported.
More versions could be cut, but the gain would be minimal, while
removing only VS2013 has the advantage to remove from the core code all
the dependencies on the value defined by _MSC_VER, where compatibility
tweaks have accumulated across the years mostly around locales and
strtof(), so that's a nice isolated cleanup.
Note that this commit additionally allows a revert of
3154e16. The
versions of Visual Studio now supported range from 2015 to 2022.
Author: Michael Paquier
Reviewed-by: Juan José Santamaría Flecha, Tom Lane, Thomas Munro, Justin
Pryzby
Discussion: https://postgr.es/m/YoH2IMtxcS3ncWn+@paquier.xyz
Michael Paquier [Wed, 13 Jul 2022 23:32:18 +0000 (08:32 +0900)]
Fix output of createuser --help with --valid-until
The argument required by --valid-until, a timestamp string, was missing
in the description of --help.
Author: Shinoda, Noriyoshi
Reviewed-by: Nathan Bossart
Discussion: https://postgr.es/m/DM4PR84MB1734A6CE3839A68B59BEA599EE899@DM4PR84MB1734.NAMPRD84.PROD.OUTLOOK.COM
Tom Lane [Wed, 13 Jul 2022 18:33:24 +0000 (14:33 -0400)]
Mop up pg_upgrade's nls.mk for commit
b0a55e432.
We'll do this the hard way for today.
Discussion: https://postgr.es/m/
20220713.160853.
453362706160476128.horikyota.ntt@gmail.com
Tom Lane [Wed, 13 Jul 2022 18:29:10 +0000 (14:29 -0400)]
Revert "Use wildcards instead of manually-maintained file lists in */nls.mk."
This reverts commit
617d69141220f277170927e03a19d2f1b77aed77.
While I still think the basic idea is attractive, we need to sort
out what happens with built .c files, and there also seem to be
VPATH issues.
Tom Lane [Wed, 13 Jul 2022 17:37:10 +0000 (13:37 -0400)]
Avoid unsatisfied-external-reference errors in static inlines.
Commit
9c727360b neglected the lesson we've learned before:
protect references to backend global variables with #ifndef FRONTEND.
Since there's already a place for static inlines in this file,
move the just-converted functions to that stanza. Undo the
entirely gratuitous de-macroization of RelationGetNumberOfBlocks
(that one may be okay, since it has no global variable references,
but it's also pointless).
Per buildfarm.
Tom Lane [Wed, 13 Jul 2022 16:56:42 +0000 (12:56 -0400)]
Use wildcards instead of manually-maintained file lists in */nls.mk.
The backend already used a mechanically-generated list of *.c files,
but everywhere else we had a manually-written-out list of files in
which to seek translatable messages. Commit
b0a55e432 contains the
latest in a long line of failures to update those lists. Rather than
manually fix its oversight, let's change to using "$(wildcard *.c)"
in all these nls.mk files.
Many of these files also have manual references to some *.c files
in other directories, most often src/common/. Perhaps we should try
to improve that situation too; but it's a bit less clear how, so for
now just fix the local file references.
Kyotaro Horiguchi and Tom Lane
Discussion: https://postgr.es/m/
20220713.160853.
453362706160476128.horikyota.ntt@gmail.com
Tom Lane [Wed, 13 Jul 2022 15:48:17 +0000 (11:48 -0400)]
Remove artificial restrictions on which node types have out/read funcs.
The initial version of gen_node_support.pl manually excluded most
utility statement node types from having out/read support, and
also some raw-parse-tree-only node types. That was mostly to keep
the output comparable to the old hand-maintained code. We'd like
to have out/read support for utility statements, for debugging
purposes and so that they can be included in new-style SQL functions;
so it's time to lift that restriction.
Most if not all of the previously-excluded raw-parse-tree-only node
types can appear in expression subtrees of utility statements, so
they have to be handled too.
We don't quite have full read support yet; certain custom_read_write
node types need to have their handwritten read functions implemented
before that will work.
Doing this allows us to drop the previous hack in _outQuery to not
dump the utilityStmt field in most cases, which means we no longer
need manually-maintained out/read functions for Query, so get rid
of those in favor of auto-generating them.
Fix a couple of omissions in gen_node_support.pl that are exposed
through having to handle more node types.
catversion bump forced because somebody was sloppy about the field
order in the manually-maintained Query out/read functions.
(Committers should note that almost all changes in parsenodes.h
are now grounds for a catversion bump.)
Peter Eisentraut [Wed, 13 Jul 2022 12:33:03 +0000 (14:33 +0200)]
Convert macros to static inline functions (bufmgr.h)
Reviewed-by: Amul Sul <sulamul@gmail.com>
Discussion: https://www.postgresql.org/message-id/flat/
5b558da8-99fb-0a99-83dd-
f72f05388517%40enterprisedb.com
Peter Eisentraut [Wed, 13 Jul 2022 12:33:03 +0000 (14:33 +0200)]
Convert macros to static inline functions (itemptr.h)
Reviewed-by: Amul Sul <sulamul@gmail.com>
Discussion: https://www.postgresql.org/message-id/flat/
5b558da8-99fb-0a99-83dd-
f72f05388517%40enterprisedb.com
Alvaro Herrera [Wed, 13 Jul 2022 10:10:03 +0000 (12:10 +0200)]
Plug memory leak
Commit
054325c5eeb3 created a memory leak in PQsendQueryInternal in case
an error occurs while sending the message. Repair.
Backpatch to 14, like that commit. Reported by Coverity.
Alvaro Herrera [Wed, 13 Jul 2022 09:57:25 +0000 (11:57 +0200)]
Fix flag tests in src/test/modules/test_oat_hooks
In what must have been a copy'n paste mistake, all the flag tests use
the same flag rather than a different flag each. The bug is not
suprising, considering that it's dead code; add a minimal, testimonial
line to cover it.
This is all pretty inconsequential, because this is just example code,
but it had better be correct.
Discussion: https://postgr.es/m/
20220712152059.fwli2majwgzdmh4r@alvherre.pgsql
Peter Eisentraut [Wed, 13 Jul 2022 10:21:45 +0000 (12:21 +0200)]
Allow specifying STORAGE attribute for a new table
Previously, the STORAGE specification was only available in ALTER
TABLE. This makes it available in CREATE TABLE as well.
Also make the code and the documentation for STORAGE and COMPRESSION
attributes consistent.
Author: Teodor Sigaev <teodor@sigaev.ru>
Author: Aleksander Alekseev <aleksander@timescale.com>
Reviewed-by: Peter Eisentraut <peter.eisentraut@enterprisedb.com>
Reviewed-by: wenjing zeng <wjzeng2012@gmail.com>
Reviewed-by: Matthias van de Meent <boekewurm+postgres@gmail.com>
Reviewed-by: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
Discussion: https://postgr.es/m/
de83407a-ae3d-a8e1-a788-
920eb334f25b@sigaev.ru
Peter Eisentraut [Wed, 13 Jul 2022 09:40:00 +0000 (11:40 +0200)]
Remove useless assertions
We don't need Assert(IsA(foo, String)) right before running
strVal(foo), since strVal() already does the assertion internally (via
castNode()).
Alvaro Herrera [Wed, 13 Jul 2022 08:34:54 +0000 (10:34 +0200)]
Fix XID list support some more
Read/out support in
5ca0fe5c8ad7 was missing/incomplete, per Tom Lane.
Again, as far as core is concerned, this is not only dead code but also
untested; however, third parties may come to rely on it, so the standard
features should work.
Discussion: https://postgr.es/m/
1548311.
1657636605@sss.pgh.pa.us
Peter Eisentraut [Wed, 13 Jul 2022 07:15:01 +0000 (09:15 +0200)]
Fix for make unportability
88dad06b47eb80f699211c9b0b7a1c6d9016ad19 contains a make $(shell)
construct that apparently confuses older GNU make versions (possibly
because of the # inside the shell command?). This construct, which
would allow # comments inside LINGUAS files, was adapted from gettext
recommendations, but we don't actually need that functionality, so
sidestep this whole issue by just using plain "cat".
In passing, make this code work with vpath.
Peter Eisentraut [Wed, 13 Jul 2022 05:51:43 +0000 (07:51 +0200)]
NLS: Put list of available languages into LINGUAS files
This moves the list of available languages from nls.mk into a separate
file called po/LINGUAS. Advantages:
- It keeps the parts notionally managed by programmers (nls.mk)
separate from the parts notionally managed by translators (LINGUAS).
- It's the standard practice recommended by the Gettext manual
nowadays.
- The Meson build system also supports this layout (and of course
doesn't know anything about our custom nls.mk), so this would enable
sharing the list of languages between the two build systems.
(The MSVC build system currently finds all po files by globbing, so it
is not affected by this change.)
Reviewed-by: Andres Freund <andres@anarazel.de>
Discussion: https://www.postgresql.org/message-id/flat/
557a9f5c-e871-edc7-2f58-
a4140fb65b7b@enterprisedb.com
David Rowley [Wed, 13 Jul 2022 05:01:01 +0000 (17:01 +1200)]
Small cleanup of create_list_bounds()
When checking for interleaved partitions, we mark the partition as
interleaved when;
1. we find an earlier partition index when looping over the
sorted-by-Datum indexes[] array, or;
2. we find that the NULL partition allows some non-NULL Datum value.
In the code, as it was written in
db632fbca we'll continue to check for
case 2 when we've already marked the partition as interleaved for case 1.
Here we make it so we don't bother marking the partition as interleaved
for case 2 when it's already been marked due to case 1.
Really all this saves is a useless call to bms_add_member(), but since
this code is new to PG15, it seems worth fixing it now to save anyone the
trouble of complaining at some time in the future. We have the
opportunity to improve this now before PG15 is out. This might ease some
future back-patching pain.
Per report and patch by Zhihong Yu. However, I slightly revised the
comments and altered the bms_add_member() code to match in both locations.
We already know that index is equal to boundinfo->null_index from the if
condition.
Author: Zhihong Yu
Discussion: https://postgr.es/m/CALNJ-vQbZR0pYxz9zQ5bqXVcwtGgNgVupeEpNT65HZ+yWZnc4g@mail.gmail.com
Backpatch-through: 15, same as
db632fbca.
Michael Paquier [Wed, 13 Jul 2022 03:21:20 +0000 (12:21 +0900)]
createuser: Add support for more clause types through new options
The following options are added to createuser:
* --valid-until to generate a VALID UNTIL clause for the role created.
* --bypassrls/--no-bypassrls for BYPASSRLS/NOBYPASSRLS.
* -m/--member to make the new role a member of an existing role, with an
extra ROLE clause generated. The clause generated overlaps with
-g/--role, but per discussion this was the most popular choice as option
name.
* -a/--admin for the addition of an ADMIN clause.
These option names are chosen to be completely new, so as they do not
impact anybody relying on the existing option set. Tests are added for
the new options and extended a bit, while on it, to cover more patterns
where quotes are added to various elements of the query generated.
Author: Shinya Kato
Reviewed-by: Nathan Bossart, Daniel Gustafsson, Robert Haas, Kyotaro
Horiguchi, David G. Johnston, Przemysław Sztoch
Discussion: https://postgr.es/m/
69a9851035cf0f0477bcc5d742b031a3@oss.nttdata.com
David Rowley [Wed, 13 Jul 2022 03:03:47 +0000 (15:03 +1200)]
Use list_copy_head() instead of list_truncate(list_copy(...), ...)
Truncating off the end of a freshly copied List is not a very efficient
way of copying the first N elements of a List.
In many of the cases that are updated here, the pattern was only being
used to remove the final element of a List. That's about the best case
for it, but there were many instances where the truncate trimming the List
down much further.
4cc832f94 added list_copy_head(), so let's use it in cases where it's
useful.
Author: David Rowley
Discussion: https://postgr.es/m/
1986787.
1657666922%40sss.pgh.pa.us
Michael Paquier [Wed, 13 Jul 2022 02:29:02 +0000 (11:29 +0900)]
createuser: Cleanup and fix internal option ordering
This utility supports 23 options that are not really ordered in the
code, making the addition of new things more complicated than necessary.
This cleanup is in preparation for a patch to add even more options.
Discussion: https://postgr.es/m/
69a9851035cf0f0477bcc5d742b031a3@oss.nttdata.com
David Rowley [Wed, 13 Jul 2022 02:02:20 +0000 (14:02 +1200)]
Tidy up code in get_cheapest_group_keys_order()
There are a few things that we could do a little better within
get_cheapest_group_keys_order():
1. We should be using list_free() rather than pfree() on a List.
2. We should use for_each_from() instead of manually coding a for loop to
skip the first n elements of a List
3. list_truncate(list_copy(...), n) is not a great way to copy the first n
elements of a list. Let's invent list_copy_head() for that. That way we
don't need to copy the entire list just to truncate it directly
afterwards.
4. We can simplify finding the cheapest cost by setting the cheapest cost
variable to DBL_MAX. That allows us to skip special-casing the initial
iteration of the loop.
Author: David Rowley
Discussion: https://postgr.es/m/CAApHDvrGyL3ft8waEkncG9y5HDMu5TFFJB1paoTC8zi9YK97Nw@mail.gmail.com
Backpatch-through: 15, where get_cheapest_group_keys_order was added.
Tom Lane [Tue, 12 Jul 2022 21:05:46 +0000 (17:05 -0400)]
Fix ECPG's handling of type names that match SQL keywords.
Previously, ECPG could only cope with variable declarations whose
type names either weren't any SQL keyword, or were at least partially
reserved. If you tried to use something in the unreserved_keyword
category, you got a syntax error.
This is pretty awful, not only because it says right on the tin that
those words are not reserved, but because the set of such keywords
tends to grow over time. Thus, an ECPG program that was just fine
last year could fail when recompiled with a newer SQL grammar.
We had to work around this recently when STRING became a keyword,
but it's time for an actual fix instead of a band-aid.
To fix, borrow a trick from C parsers and make the lexer's behavior
change when it sees a word that is known as a typedef. This is not
free of downsides: if you try to use such a name as a SQL keyword
in EXEC SQL later in the program, it won't be recognized as a SQL
keyword, leading to a syntax error there instead. So in a real
sense this is just trading one hazard for another. But there is an
important difference: with this, whether your ECPG program works
depends only on what typedef names and SQL commands are used in the
program text. If it compiles today it'll still compile next year,
even if more words have become SQL keywords.
Discussion: https://postgr.es/m/
3661437.
1653855582@sss.pgh.pa.us
Tom Lane [Tue, 12 Jul 2022 20:30:36 +0000 (16:30 -0400)]
Invent qsort_interruptible().
Justin Pryzby reported that some scenarios could cause gathering
of extended statistics to spend many seconds in an un-cancelable
qsort() operation. To fix, invent qsort_interruptible(), which is
just like qsort_arg() except that it will also do CHECK_FOR_INTERRUPTS
every so often. This bloats the backend by a couple of kB, which
seems like a good investment. (We considered just enabling
CHECK_FOR_INTERRUPTS in the existing qsort and qsort_arg functions,
but there are some callers for which that'd demonstrably be unsafe.
Opt-in seems like a better way.)
For now, just apply qsort_interruptible() in statistics collection.
There's probably more places where it could be useful, but we can
always change other call sites as we find problems.
Back-patch to v14. Before that we didn't have extended stats on
expressions, so that the problem was less severe. Also, this patch
depends on the sort_template infrastructure introduced in v14.
Tom Lane and Justin Pryzby
Discussion: https://postgr.es/m/
20220509000108.GQ28830@telsasoft.com
Tom Lane [Tue, 12 Jul 2022 19:37:39 +0000 (15:37 -0400)]
Improve error reporting from validate_exec().
validate_exec() didn't guarantee to set errno to something appropriate
after a failure, leading to callers not being able to print an on-point
message. Improve that.
Noted by Kyotaro Horiguchi, though this isn't exactly his proposal.
Discussion: https://postgr.es/m/
20220615.131403.
1791191615590453058.horikyota.ntt@gmail.com
Tom Lane [Tue, 12 Jul 2022 19:17:44 +0000 (15:17 -0400)]
Remove trailing newlines in pg_upgrade's message strings.
pg_upgrade does not use common/logging.c, which is unfortunate
but changing it to do so seems like more work than is justified.
However, we really need to make it work more like common/logging.c
in one respect: the latter expects supplied message strings to not
end with a newline, instead adding one internally. As it stands,
pg_upgrade's logging facilities expect a caller-supplied newline
in some cases and not others, which is already an invitation to bugs,
but the inconsistency with our other frontend code makes it worse.
There are already several places with missing or extra newlines,
and it's inevitable that there won't be more if we let this stand.
Hence, run around and get rid of all trailing newlines in message
strings, and add an Assert that there's not one, similar to the
existing Assert in common/logging.c. Adjust the logging functions
to supply a newline at the right places.
(Some of these strings also have a *leading* newline, which would
be a good thing to get rid of too; but this patch doesn't attempt
that.)
There are some consequent minor changes in output. The ones that
aren't outright bug fixes are generally removal of extra blank
lines that the original coding intentionally inserted. It didn't
seem worth being bug-compatible with that.
Patch by me, reviewed by Kyotaro Horiguchi and Peter Eisentraut
Discussion: https://postgr.es/m/113191.
1655233060@sss.pgh.pa.us
Tom Lane [Tue, 12 Jul 2022 15:22:52 +0000 (11:22 -0400)]
Add defenses against unexpected changes in the NodeTag enum list.
Having different build systems producing different contents of the
NodeTag enum would be catastrophic for extension ABI stability.
But that ordering depends on the order in which gen_node_support.pl
processes its input files. It seems too fragile to let the Makefiles,
MSVC build scripts, and soon meson build scripts all set this order
independently. As a klugy but serviceable solution, put a canonical
copy of the file list into gen_node_support.pl itself, and check that
against the files given on the command line.
Also, while it's fine to add and delete node tags during development,
we must not let the assigned NodeTag values change unexpectedly in
stable branches. Add a cross-check that can be enabled when a branch
is forked off (or later, but that is a time when we're unlikely to
miss doing it). It just checks that the last auto-assigned number
doesn't change, which is simplistic but will catch the most likely
sorts of mistakes.
From time to time we do need to add a node tag in a stable branch.
To support doing that without changing the branch's auto-assigned
tag numbers, invent pg_node_attr(nodetag_number(VALUE)) which can
be used to give such a node a hand-assigned tag above the last
auto-assigned one.
Discussion: https://postgr.es/m/
1249010.
1657574337@sss.pgh.pa.us
Tom Lane [Tue, 12 Jul 2022 14:46:58 +0000 (10:46 -0400)]
Invent nodetag_only attribute for Nodes.
This allows explaining gen_node_support.pl's handling of execnodes.h
and some other input files as being a shortcut for explicit marking
of all their node declarations as pg_node_attr(nodetag_only).
I foresee that someday we might need to be more fine-grained about
that, and this change provides the infrastructure needed to do so.
For now, it just allows removal of the script's klugy special case
for CallContext and InlineCodeBlock.
Discussion: https://postgr.es/m/75063.
1657410615@sss.pgh.pa.us
Robert Haas [Tue, 12 Jul 2022 14:26:48 +0000 (10:26 -0400)]
Rename some functions to mention Relation instead of RelFileLocator.
This is definitely shorter, and hopefully clearer.
Kyotaro Horiguchi, reviewed by Dilip Kumar and by me
Discussion: http://postgr.es/m/
20220707.174436.
1885393789789795413.horikyota.ntt@gmail.com
Alvaro Herrera [Tue, 12 Jul 2022 14:11:04 +0000 (16:11 +0200)]
Add copy/equal support for XID lists
Commit
f10a025cfe97 added support for List to store Xids, but didn't
handle the new type in all cases. Add some obviously necessary pieces.
As far as I am aware, this is all dead code as far as core code is
concerned, but it seems unacceptable not to have it in case third-party
code wants to rely on this type of list. (Some parts of the List API
remain unimplemented, but that can be fixed as and when needed -- see
lack of list_intersection_oid, list_deduplicate_int as precedents.)
Discussion: https://postgr.es/m/
20220708164534.nbejhgt4ajz35p65@alvherre.pgsql
John Naylor [Tue, 12 Jul 2022 04:13:41 +0000 (11:13 +0700)]
Fix out-of-bounds read in json_lex_string
Commit
3838fa269 added a lookahead loop to allow building strings multiple
bytes at a time. This loop could exit because it reached the end of input,
yet did not check for that before checking if we reached the end of a
valid string. To fix, put the end of string check back in the outer loop.
Per Valgrind animal skink
Fujii Masao [Tue, 12 Jul 2022 00:18:02 +0000 (09:18 +0900)]
Support TRUNCATE triggers on foreign tables.
Now some foreign data wrappers support TRUNCATE command.
So it's useful to support TRUNCATE triggers on foreign tables for
audit logging or for preventing undesired truncation.
Author: Yugo Nagata
Reviewed-by: Fujii Masao, Ian Lawrence Barwick
Discussion: https://postgr.es/m/
20220630193848.
5b02e0d6076b86617a915682@sraoss.co.jp
Thomas Munro [Mon, 11 Jul 2022 22:57:52 +0000 (10:57 +1200)]
Doc: Acknowledge historically supported CPUs and OSes.
PostgreSQL/POSTGRES has run on a huge range of CPUs and OSes. As we're
dropping some of the earliest systems the project was founded on, let's
provide a place to remember them.
Discussion: https://postgr.es/m/959917.
1657522169%40sss.pgh.pa.us
Thomas Munro [Mon, 11 Jul 2022 22:53:45 +0000 (10:53 +1200)]
Further tidy-up for old CPU architectures.
Further to commit
92d70b77, let's drop the code we carry for the
following untested architectures: M68K, M88K, M32R, SuperH. We have no
idea if anything actually works there, and surely as vintage hardware
and microcontrollers they would be underpowered for modern purposes.
We could always consider re-adding SuperH based on evidence of usage and
build farm support, if someone shows up to provide it.
While here, SPARC is usually written in all caps.
Suggested-by: Tom Lane <tgl@sss.pgh.pa.us>
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Reviewed-by: Robert Haas <robertmhaas@gmail.com> (the idea, not the patch)
Discussion: https://postgr.es/m/959917.
1657522169%40sss.pgh.pa.us
Jeff Davis [Mon, 11 Jul 2022 19:29:33 +0000 (12:29 -0700)]
Provide log_status_format(), useful for an emit_log_hook.
Refactor so that log_line_prefix() is a thin wrapper over a new
function log_status_format(), and move the implementation to the
latter. Export log_status_format() so that it can be used by an
emit_log_hook.
Discussion: https://postgr.es/m/
39c8197652f4d3050aedafae79fa5af31096505f.camel%40j-davis.com
Reviewed-by: Michael Paquier, Alvaro Herrera
Tom Lane [Mon, 11 Jul 2022 17:38:40 +0000 (13:38 -0400)]
Rationalize order of input files for gen_node_support.pl.
Per a question from Andres Freund. While here, also make the
list of nodetag-only files easier to compare to the full list
of input files.
Discussion: https://postgr.es/m/
20220710214622.haiektrjzisob6rl@awork3.anarazel.de
Robert Haas [Mon, 11 Jul 2022 17:32:23 +0000 (13:32 -0400)]
Fix mistake in comment.
Kyotaro Horiguchi
Discussion: http://postgr.es/m/
20220708.145951.
382076151410075693.horikyota.ntt@gmail.com
Peter Eisentraut [Mon, 11 Jul 2022 05:20:35 +0000 (07:20 +0200)]
Convert macros to static inline functions (bufpage.h)
Remove PageIsValid() and PageSizeIsValid(), which weren't used and
seem unnecessary.
Some code using these formerly-macros needs some adjustments because
it was previously playing loose with the Page vs. PageHeader types,
which is no longer possible with the functions instead of macros.
Reviewed-by: Amul Sul <sulamul@gmail.com>
Discussion: https://www.postgresql.org/message-id/flat/
5b558da8-99fb-0a99-83dd-
f72f05388517%40enterprisedb.com
Thomas Munro [Mon, 11 Jul 2022 02:47:16 +0000 (14:47 +1200)]
Fix lock assertions in dshash.c.
dshash.c previously maintained flags to be able to assert that you
didn't hold any partition lock. These flags could get out of sync with
reality in error scenarios.
Get rid of all that, and make assertions about the locks themselves
instead. Since LWLockHeldByMe() loops internally, we don't want to put
that inside another loop over all partition locks. Introduce a new
debugging-only interface LWLockAnyHeldByMe() to avoid that.
This problem was noted by Tom and Andres while reviewing changes to
support the new shared memory stats system, and later showed up in
reality while working on commit
389869af.
Back-patch to 11, where dshash.c arrived.
Reported-by: Tom Lane <tgl@sss.pgh.pa.us>
Reported-by: Andres Freund <andres@anarazel.de>
Reviewed-by: Kyotaro HORIGUCHI <horiguchi.kyotaro@lab.ntt.co.jp>
Reviewed-by: Zhihong Yu <zyu@yugabyte.com>
Reviewed-by: Andres Freund <andres@anarazel.de>
Discussion: https://postgr.es/m/
20220311012712.botrpsikaufzteyt@alap3.anarazel.de
Discussion: https://postgr.es/m/CA%2BhUKGJ31Wce6HJ7xnVTKWjFUWQZPBngxfJVx4q0E98pDr3kAw%40mail.gmail.com
John Naylor [Fri, 1 Jul 2022 10:28:20 +0000 (17:28 +0700)]
Build de-escaped JSON strings in larger chunks during lexing
During COPY BINARY with large JSONB blobs, it was found that half
the time was spent parsing JSON, with much of that spent in separate
appendStringInfoChar() calls for each input byte.
Add lookahead loop to json_lex_string() to allow batching multiple bytes
via appendBinaryStringInfo(). Also use this same logic when de-escaping
is not done, to avoid code duplication.
Report and proof of concept patch by Jelte Fennema, reworked by Andres
Freund and John Naylor
Discussion: https://www.postgresql.org/message-id/CAGECzQQuXbies_nKgSiYifZUjBk6nOf2%3DTSXqRjj2BhUh8CTeA%40mail.gmail.com
Discussion: https://www.postgresql.org/message-id/flat/PR3PR83MB0476F098CBCF68AF7A1CA89FF7B49@PR3PR83MB0476.EURPRD83.prod.outlook.com
Michael Paquier [Mon, 11 Jul 2022 03:47:52 +0000 (12:47 +0900)]
Fix two portability issues with the tests of test_oat_hooks
This addresses two issues in the tests of test_oat_hooks:
- The role regress_test_user was being left behind, preventing the test
to succeed on repeated runs. It makes sense to leave some objects
behind to have more coverage for pg_upgrade (as does test_pg_dump), but
the role dropped here does not own any objects so there is no reason to
keep it.
- GRANT SET ON PARAMETER is issued, creating an entry in
pg_parameter_acl without cleaning up the entry created. This causes
an overlap with unsafe_tests as both use work_mem, making the latter
fail. This commit adds an extra REVOKE SET ON PARAMETER to clean the
contents of pg_parameter_acl, switching to maintenance_work_mem rather
than work_mem to avoid an overlap between both tests.
The tests of test_oat_hooks cannot use installcheck yet as these are
proving to be unstable with caching and the namespace search hooks, so
the issues fixed here cannot be reached yet, but they would be once the
hook issue is addressed and installcheck is allowed again in
test_oat_hooks.
Discussion: https://postgr.es/m/YrpVkADAY0knF6vM@paquier.xyz
Backpatch-through: 15
Michael Paquier [Mon, 11 Jul 2022 02:20:15 +0000 (11:20 +0900)]
Improve error message with JSON_SERIALIZE()
The error message introduced in
3c633f3 can share the same format string
with an existing message used for JSON(), reducing the translation
effort.
Author: Kyotaro Horiguchi
Discussion: https://postgr.es/m/
20220708.154135.
2123613118233840495.horikyota.ntt@gmail.com
Backpatch-through: 15
Michael Paquier [Mon, 11 Jul 2022 02:07:33 +0000 (11:07 +0900)]
Improve two comments related to a boolean DefElem's value
The original comments mentioned a "parameter" as something not defined
in a fast-exit path to assume a true status. This is rather confusing
as the parameter DefElem is defined, and the intention is to check if
its value is defined. This improves both comments to mention the value
assigned to the DefElem's value instead, so as future patches are able
to catch the tweak if this code pattern gets copied around more.
Author: Peter Smith
Discussion: https://postgr.es/m/CAHut+Pv0yWynWTmp4o34s0d98xVubys9fy=p0YXsZ5_sUcNnMw@mail.gmail.com
Michael Paquier [Mon, 11 Jul 2022 01:56:17 +0000 (10:56 +0900)]
doc: Fix inconsistent quotes in some jsonb fields
Single quotes are not allowed in json internals, double quotes are.
Reported-by: Eric Mutta
Discussion: https://postgr.es/m/
165715362165.665.
3875113264927503997@wrigleys.postgresql.org
Backpatch-through: 14
Thomas Munro [Sun, 10 Jul 2022 23:50:41 +0000 (11:50 +1200)]
Tidy up claimed supported CPUs and OSes.
* Remove arbitrary mention of certain endianness and bitness variants;
it's enough to say that applicable variants are expected to work.
* List RISC-V (known to work, being tested).
* List SuperH and M88K (code exists, unknown status, like M68K).
* De-list VAX and remove code (known not to work).
* Remove stray trace of Alpha (support was removed years ago).
* List illumos, DragonFlyBSD (known to work, being tested).
* No need to single Windows out by listing a specific version, when we
don't do that for other OSes; it's enough to say that we support
current versions of the listed OSes (when 16 ships, that'll be
Windows 10+).
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Reviewed-by: Greg Stark <stark@mit.edu>
Discussion: https://postgr.es/m/CA%2BhUKGKk7NZO1UnJM0PyixcZPpCGqjBXW_0bzFZpJBGAf84XKg%40mail.gmail.com
Thomas Munro [Sun, 10 Jul 2022 04:30:03 +0000 (16:30 +1200)]
Fix \watch's interaction with libedit on ^C.
When you hit ^C, the terminal driver in Unix-like systems echoes "^C" as
well as sending an interrupt signal (depending on stty settings). At
least libedit (but maybe also libreadline) is then confused about the
current cursor location, and corrupts the display if you try to scroll
back. Fix, by moving to a new line before the next prompt is displayed.
Back-patch to all supported released.
Author: Pavel Stehule <pavel.stehule@gmail.com>
Reported-by: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://postgr.es/m/
3278793.
1626198638%40sss.pgh.pa.us
Tom Lane [Sat, 9 Jul 2022 23:25:53 +0000 (19:25 -0400)]
Exclude nodetags.h from headerscheck and cpluspluscheck.
Since this file contains just a fragment of an enum declaration,
it can't be compiled on its own.
Tom Lane [Sat, 9 Jul 2022 19:15:05 +0000 (15:15 -0400)]
Make assorted quality-of-life improvements in gen_node_support.pl.
Fix incorrect reporting of the location of errors (such as bogus
node attributes). Add header comments to the generated files,
containing copyright notices and reminders that they are generated
files, as we do in other file-generating scripts. Arrange to not
leave a clutter of temporary files when the script detects an error.
Discussion: https://postgr.es/m/
3843645.
1657385930@sss.pgh.pa.us
Tom Lane [Sat, 9 Jul 2022 19:10:15 +0000 (15:10 -0400)]
Doc: rearrange high-level commentary about node support coverage.
copyfuncs.c and friends no longer seem like great places to put
high-level remarks about what's covered and what isn't. Move that
material to backend/nodes/README and other more-prominent places.
Add back (versions of) some remarks that disappeared in
2be87f092.
Discussion: https://postgr.es/m/
3843645.
1657385930@sss.pgh.pa.us
Tom Lane [Sat, 9 Jul 2022 17:58:06 +0000 (13:58 -0400)]
Mark Scan as an abstract node type, too.
On further review, this one is never instantiated either.
Tom Lane [Sat, 9 Jul 2022 17:35:37 +0000 (13:35 -0400)]
Mark PlanState as an abstract node type.
In the same vein as commit
251154beb, make it clear that we never
instantiate PlanState.
Also mark MemoryContextData as abstract. This has no effect right now,
since memnodes.h isn't one of the files fed to gen_node_support.pl.
But it seems like good documentation and future-proofing.
Peter Eisentraut [Sat, 9 Jul 2022 13:06:01 +0000 (15:06 +0200)]
Remove code sections obsoleted by node support automation
This removes the code sections that were ifdef'ed out by
964d01ae90c314eb31132c2e7712d5d9fc237331.
Peter Eisentraut [Sat, 9 Jul 2022 11:22:46 +0000 (13:22 +0200)]
Move a comment
Move a comment from the to-be-deleted section of nodes.h to where it
might still be useful.
Peter Eisentraut [Sat, 9 Jul 2022 08:20:54 +0000 (10:20 +0200)]
Add generated nodes\nodetags.h to clean.bat
Peter Eisentraut [Sat, 9 Jul 2022 08:20:23 +0000 (10:20 +0200)]
Install generated nodes/nodetags.h
Peter Eisentraut [Sat, 9 Jul 2022 07:44:09 +0000 (09:44 +0200)]
Fix vpath build
Peter Eisentraut [Sat, 9 Jul 2022 06:52:19 +0000 (08:52 +0200)]
Automatically generate node support functions
Add a script to automatically generate the node support functions
(copy, equal, out, and read, as well as the node tags enum) from the
struct definitions.
For each of the four node support files, it creates two include files,
e.g., copyfuncs.funcs.c and copyfuncs.switch.c, to include in the main
file. All the scaffolding of the main file stays in place.
I have tried to mostly make the coverage of the output match what is
currently there. For example, one could now do out/read coverage of
utility statement nodes, but I have manually excluded those for now.
The reason is mainly that it's easier to diff the before and after,
and adding a bunch of stuff like this might require a separate
analysis and review.
Subtyping (TidScan -> Scan) is supported.
For the hard cases, you can just write a manual function and exclude
generating one. For the not so hard cases, there is a way of
annotating struct fields to get special behaviors. For example,
pg_node_attr(equal_ignore) has the field ignored in equal functions.
(In this patch, I have only ifdef'ed out the code to could be removed,
mainly so that it won't constantly have merge conflicts. It will be
deleted in a separate patch. All the code comments that are worth
keeping from those sections have already been moved to the header
files where the structs are defined.)
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://www.postgresql.org/message-id/flat/
c1097590-a6a4-486a-64b1-
e1f9cc0533ce%40enterprisedb.com
Bruce Momjian [Sat, 9 Jul 2022 00:23:35 +0000 (20:23 -0400)]
doc: add examples for array_length() and jsonb_array_length()
The examples show the output of array_length() and jsonb_array_length()
for empty arrays.
Discussion: https://postgr.es/m/CAKFQuwaoBmRuWdMLzLHDCFDJDX3wvfQ7egAF0bpik_BFgG1KWg@mail.gmail.com
Author: David G. Johnston
Backpatch-through: 13
Bruce Momjian [Fri, 8 Jul 2022 22:36:27 +0000 (18:36 -0400)]
doc: add pg_prewarm example
Discussion: https://postgr.es/m/
20220618085541.ezxdaljlpo6x7msc@home-desktop
Author: Dong Wook Lee
Backpatch-through: 11
Alexander Korotkov [Fri, 8 Jul 2022 18:51:26 +0000 (21:51 +0300)]
Add missing inequality searches to rbtree
PostgreSQL contains the implementation of the red-black tree. The red-black
tree is the ordered data structure, and one of its advantages is the ability
to do inequality searches. This commit adds rbt_find_less() and
rbt_find_great() functions implementing these searches. While these searches
aren't yet used in the core code, they might be useful for extensions.
Discussion: https://postgr.es/m/CAGRrpzYE8-7GCoaPjOiL9T_HY605MRax-2jgTtLq236uksZ1Sw%40mail.gmail.com
Author: Steve Chavez, Alexander Korotkov
Reviewed-by: Alexander Korotkov
Alexander Korotkov [Fri, 8 Jul 2022 18:51:00 +0000 (21:51 +0300)]
Use C99 designator in the rbtree sentinel definition
This change should improve the code readability.
Discussion: https://postgr.es/m/CAGRrpzYE8-7GCoaPjOiL9T_HY605MRax-2jgTtLq236uksZ1Sw%40mail.gmail.com
Author: Steve Chavez, Alexander Korotkov
Reviewed-by: Alexander Korotkov
Robert Haas [Fri, 8 Jul 2022 14:15:19 +0000 (10:15 -0400)]
Preserve relfilenode of pg_largeobject and its index across pg_upgrade.
Commit
9a974cbcba005256a19991203583a94b4f9a21a9 did this for user
tables, but pg_upgrade treats pg_largeobject as a user table, and so
needs the same treatment. Without this fix, if you rewrite the
pg_largeobject table and then perform an upgrade with pg_upgrade, the
table will apparently be empty on the new cluster, while all of your
objects will end up with an orphaned file.
With this fix, instead of the old cluster's pg_largeobject files ending
up orphaned, the original files fro the new cluster do. That's mostly
harmless because we expect the table to be empty, but we might want
to arrange to remove the as part of the upgrade. Since we're still
debating the best way of doing that, I (rhaas) have decided to postpone
dealing with that problem and get the basic fix committed.
Justin Pryzby, reviewed by Shruthi Gowda and by me.
Discussion: http://postgr.es/m/
20220701231413.GI13040@telsasoft.com
Peter Eisentraut [Fri, 8 Jul 2022 09:03:45 +0000 (11:03 +0200)]
Adjust node serialization tag of A_Expr for consistency
Changed from AEXPR to A_EXPR for consistency.
Discussion: https://www.postgresql.org/message-id/
2592455.
1657140387%40sss.pgh.pa.us
Peter Eisentraut [Fri, 8 Jul 2022 08:38:02 +0000 (10:38 +0200)]
Remove T_Join and T_Plan
These are abstract node types that don't need to have a node tag
defined.
Discussion: https://www.postgresql.org/message-id/
2592455.
1657140387%40sss.pgh.pa.us
John Naylor [Fri, 8 Jul 2022 07:53:20 +0000 (14:53 +0700)]
Simplify json lexing state
Instead of updating the length as we go, use a const pointer to end of
the input, which we know already at the start.
This simplifies the coding and may improve performance slightly, but
the real motivation for doing this is to make further changes in this
area easier to reason about.
Discussion: https://www.postgresql.org/message-id/CAFBsxsGhaR2KQ5eisaK%3D6Vm60t%3DaxhD8Ckj1qFoCH1pktZi%2B2w%40mail.gmail.com
Peter Eisentraut [Fri, 8 Jul 2022 07:22:27 +0000 (09:22 +0200)]
Reformat some more node comments
Reformat some more comments in node field definitions to avoid long
lines. Similar to
835d476fd21bcfb60b055941dee8c3d9559af14c, based on
additional per-field annotations that will be required.
Discussion: https://www.postgresql.org/message-id/
c5906b07-220a-a3d4-8ff3-
8ee593009424@enterprisedb.com
Thomas Munro [Thu, 7 Jul 2022 23:21:46 +0000 (11:21 +1200)]
Remove HP/Intel Itanium support.
This CPU architecture has been discontinued. We already removed HP-UX
support, we never supported Windows/Itanium, and the open source
operating systems that a vintage hardware owner might hope to run have
all either ended Itanium support or never fully released support (NetBSD
may eventually). The extra code we carry for this rare ISA is now
untested. It seems like a good time to remove it.
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Reviewed-by: Andres Freund <andres@anarazel.de>
Discussion: https://postgr.es/m/
1415825.
1656893299%40sss.pgh.pa.us
Thomas Munro [Thu, 7 Jul 2022 23:17:47 +0000 (11:17 +1200)]
Remove HP-UX port.
HP-UX hardware is no longer produced, build farm coverage recently
ended, and there are no known active maintainers targeting this OS.
Since there is a major rewrite of the build system in the pipeline for
PostgreSQL 16, and that requires development, testing and maintainance
for each OS and tool chain, it seems like a good time to drop support
for:
* HP-UX, the operating system.
* HP aCC, the HP-UX native compiler.
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Reviewed-by: Andres Freund <andres@anarazel.de>
Reviewed-by: Peter Eisentraut <peter.eisentraut@enterprisedb.com>
Discussion: https://postgr.es/m/
1415825.
1656893299%40sss.pgh.pa.us
Andrew Dunstan [Thu, 7 Jul 2022 21:40:02 +0000 (17:40 -0400)]
Only allow returning string types or bytea from json_serialize
These are documented to be the allowed types for the RETURNING clause,
but the restriction was not being enforced, which caused a segfault if
another type was specified. Add some testing for this.
Per report from a.kozhemyakin
Backpatch to release 15.
Tom Lane [Thu, 7 Jul 2022 15:23:40 +0000 (11:23 -0400)]
Remove stray references to lefttree/righttree in the executor.
The general convention in the executor is to refer to child plans
and planstates via the outerPlan[State] and innerPlan[State]
macros, but a few places didn't do it like that. For consistency
and readability, convert all the stragglers to use the macros.
(See also commit
40f42d2a3, which did some similar cleanup a few
years ago, but missed these cases.)
Richard Guo
Discussion: https://postgr.es/m/CAMbWs4-vYhh1xsa_veah4PUed2Xq=Ed_YH3=Mqt5A3Y=EgfCEg@mail.gmail.com
Fujii Masao [Thu, 7 Jul 2022 13:37:54 +0000 (22:37 +0900)]
Add checkpoint and REDO LSN to log_checkpoints message.
It is useful for debugging purposes to report the checkpoint LSN and
REDO LSN in log_checkpoints message. It can give more context while
analyzing checkpoint-related issues. pg_controldata reports the last
checkpoint LSN and REDO LSN, but having this information alongside
the log message helps analyze issues that happened previously,
connect the dots and identify the root cause.
Author: Bharath Rupireddy, Kyotaro Horiguchi
Reviewed-by: Michael Paquier, Julien Rouhaud, Nathan Bossart, Fujii Masao, Greg Stark
Discussion: https://postgr.es/m/CALj2ACWt6kqriAHrO+AJj+OmP=suwbktHT5JoYAn-nqZe2gd2g@mail.gmail.com
Dean Rasheed [Thu, 7 Jul 2022 12:08:08 +0000 (13:08 +0100)]
Fix alias matching in transformLockingClause().
When locking a specific named relation for a FOR [KEY] UPDATE/SHARE
clause, transformLockingClause() finds the relation to lock by
scanning the rangetable for an RTE with a matching eref->aliasname.
However, it failed to account for the visibility rules of a join RTE.
If a join RTE doesn't have a user-supplied alias, it will have a
generated eref->aliasname of "unnamed_join" that is not visible as a
relation name in the parse namespace. Such an RTE needs to be skipped,
otherwise it might be found in preference to a regular base relation
with a user-supplied alias of "unnamed_join", preventing it from being
locked.
In addition, if a join RTE doesn't have a user-supplied alias, but
does have a join_using_alias, then the RTE needs to be matched using
that alias rather than the generated eref->aliasname, otherwise a
misleading "relation not found" error will be reported rather than a
"join cannot be locked" error.
Backpatch all the way, except for the second part which only goes back
to 14, where JOIN USING aliases were added.
Dean Rasheed, reviewed by Tom Lane.
Discussion: https://postgr.es/m/CAEZATCUY_KOBnqxbTSPf=7fz9HWPnZ5Xgb9SwYzZ8rFXe7nb=w@mail.gmail.com
Etsuro Fujita [Thu, 7 Jul 2022 07:25:00 +0000 (16:25 +0900)]
postgres_fdw: Fix grammar.
Oversight in commit
4036bcbbb; back-patch to v15 where that appeared.
Peter Eisentraut [Thu, 7 Jul 2022 06:30:52 +0000 (08:30 +0200)]
Convert macros to static inline functions (block.h)
Remove BlockIdIsValid(), which wasn't used and is unnecessary.
Remove BlockIdCopy(), which wasn't used and can be done by struct
assignment.
(BlockIdEquals() also isn't used, but seems reasonable to keep
around.)
Reviewed-by: Amul Sul <sulamul@gmail.com>
Discussion: https://www.postgresql.org/message-id/flat/
5b558da8-99fb-0a99-83dd-
f72f05388517%40enterprisedb.com
Michael Paquier [Thu, 7 Jul 2022 04:25:45 +0000 (13:25 +0900)]
Make Windows 10 the minimal runtime requirement for WIN32
This commit bumps the runtime value of _WIN32_WINNT to be 0x0A00 for any
builds on Windows. Hence, this makes Windows 10 the minimal requirement
when running PostgreSQL under WIN32, be it for builds of Cygwin, MinGW
or Visual Studio.
The previous minimal runtime version was either Windows Vista when
building with at least Visual Studio 2015 or Windows XP for the rest.
Windows 10 is the most modern version supported by Microsoft, and per
discussion, as we don't have buildfarm members that run older versions
anymore, this is the minimal supported version that suits better for our
needs. This will actually make easier the development of some patches,
two being async I/O and large page handling by avoiding a lot of
compatibility gotchas, on platforms that have most likely few users
anyway.
It is possible to remove MIN_WINNT in win32.h and the macros
IsWindowsXXXOrGreater() that were used in the code at runtime to check
which version of Windows was getting used. The change in pg_locale.c
comes from Juan. Note that all my tests passed, and that the CI is
green. The buildfarm will quickly tell if this needs more adjustments.
Author: Michael Paquier, Juan José Santamaría Flecha
Reviewed-by: Thomas Munro
Discussion: https://postgr.es/m/Yo7tHKD8VCkeNi71@paquier.xyz
Amit Kapila [Thu, 7 Jul 2022 03:30:27 +0000 (09:00 +0530)]
Re-order disable_on_error in tab-complete.
By convention, the tab-complete subscription parameters are listed in the
COMPLETE_WITH lists in alphabetical order, but when the "disable_on_error"
parameter was introduced this was not done.
This patch just tidies that up.
Reported-by: Peter Smith
Author: Peter Smith
Reviewed-by: Euler Taveira, Takamichi Osumi
Backpatch-through: 15, where it was introduced
Discussion: https://postgr.es/m/CAHut+PucvKZgg_eJzUW--iL6DXHg1Jwj6F09tQziE3kUF67uLg@mail.gmail.com
Michael Paquier [Thu, 7 Jul 2022 01:13:01 +0000 (10:13 +0900)]
Clean up some includes and comments in TAP test scripts
A few tests included File::Path::rmtree without using it, and a comment
related to the segment size for replication slot limits was wrong.
Author: Pavel Borisov, Bharath Rupireddy
Reviewed-by: Maxim Orlov
Discussion: https://postgr.es/m/CALj2ACU4-aNLX=DrUM8F7QDwynJKzYRiqOj_33NhnGbhDs5-kQ@mail.gmail.com
Andres Freund [Wed, 6 Jul 2022 22:50:38 +0000 (15:50 -0700)]
pgstat: slru: remove outdated comment
That comment might have been true at some point during development, but
definitely isn't anymore.
Reported-By: Melanie Plageman <melanieplageman@gmail.com>
Backpatch: 15-
Tom Lane [Wed, 6 Jul 2022 21:26:27 +0000 (17:26 -0400)]
Fix wrong field order in _readMergeWhenClause().
We hadn't noticed this because it's dead code: there is no
situation where we read raw parse trees from text format.
So maybe the right fix is to remove the function altogether,
but I'll forbear for now; it's not the only dead code in
readfuncs.c, I think.
Noted while comparing existing code to the results of
Peter's auto-generation script.
David Rowley [Wed, 6 Jul 2022 20:14:00 +0000 (08:14 +1200)]
Overload index_form_tuple to allow the memory context to be supplied
40af10b57 changed things so we make use of a generation memory context for
storing tuples to be sorted by tuplesort.c. That change does not play
nicely with the changes made in
9f03ca915 (back in 2014). That commit
changed things so that index_form_tuple() is called while switched into
the tuplestore's tuplecontext. In order to fetch the tuple from the index,
index_form_tuple() must do various memory allocations which are unrelated
to the storage of the final returned tuple. Although all of these
allocations are pfree'd, the fact that we now use a generation context
means that the memory for these pfree'd allocations won't be used again by
any other allocation due to generation.c's lack of freelists. This could
result in sorts used for building indexes exceeding maintenance_work_mem
by a very large amount.
Here we fix it so we no longer allocate anything apart from the tuple
itself into the generation context by adding a new version of
index_form_tuple named index_form_tuple_context, which can be called to
specify the MemoryContext to allocate the tuple into.
Discussion: https://postgr.es/m/CAApHDvrHQkiFRHiGiAS-LMOvJN-eK-s762=tVzBz8ZqUea-a_A@mail.gmail.com
Backpatch-through: 15, where
40af10b57 was added.
Robert Haas [Wed, 6 Jul 2022 17:05:51 +0000 (13:05 -0400)]
pg_stat_statements: Fix test that assumes wal_records = rows.
It's not very robust to assume that each inserted row will produce
exactly one WAL record and that no other WAL records will be generated
in the process, because for example a particular transaction could
always be the one that has to extend clog.
Because these tests are not run by 'make installcheck' but only by
'make check', it may be that in our current testing infrastructure
this can't be hit, but it doesn't seem like a good idea to rely on
that, since unrelated changes to the regression tests or the way
write-ahead logging is done could easily cause it to start happening,
and debugging such failures is a pain.
Adjust the regression test to be less sensitive.
Anton Melnikov, reviewed by Julien Rouhaud
Discussion: http://postgr.es/m/
1ccd00d9-1723-6b68-ae56-
655aab00d406@inbox.ru
Andres Freund [Wed, 6 Jul 2022 15:53:25 +0000 (08:53 -0700)]
pgstat: drop subscription stats without slot as well, fix comment
There's no reason anymore to only drop subscription stats if associated with a
slot, now that stats drops are transactional. And since there's now no other
cleanup of stats, this would lead to stats for slot-less subscriptions to get
leaked (however most slot-less subs won't have stats).
Additionally, the comment referring to autovacuum cleaning up stats was
clearly outdated.
Author: Masahiko Sawada <sawada.mshk@gmail.com>
Discussion: https://postgr.es/m/CAD21AoAwiby3HeJE7vJe16Gr75RFfJ640dyHqvsiUhyKJTXPtw@mail.gmail.com
Backpatch: 15-
Robert Haas [Wed, 6 Jul 2022 15:39:09 +0000 (11:39 -0400)]
Change internal RelFileNode references to RelFileNumber or RelFileLocator.
We have been using the term RelFileNode to refer to either (1) the
integer that is used to name the sequence of files for a certain relation
within the directory set aside for that tablespace/database combination;
or (2) that value plus the OIDs of the tablespace and database; or
occasionally (3) the whole series of files created for a relation
based on those values. Using the same name for more than one thing is
confusing.
Replace RelFileNode with RelFileNumber when we're talking about just the
single number, i.e. (1) from above, and with RelFileLocator when we're
talking about all the things that are needed to locate a relation's files
on disk, i.e. (2) from above. In the places where we refer to (3) as
a relfilenode, instead refer to "relation storage".
Since there is a ton of SQL code in the world that knows about
pg_class.relfilenode, don't change the name of that column, or of other
SQL-facing things that derive their name from it.
On the other hand, do adjust closely-related internal terminology. For
example, the structure member names dbNode and spcNode appear to be
derived from the fact that the structure itself was called RelFileNode,
so change those to dbOid and spcOid. Likewise, various variables with
names like rnode and relnode get renamed appropriately, according to
how they're being used in context.
Hopefully, this is clearer than before. It is also preparation for
future patches that intend to widen the relfilenumber fields from its
current width of 32 bits. Variables that store a relfilenumber are now
declared as type RelFileNumber rather than type Oid; right now, these
are the same, but that can now more easily be changed.
Dilip Kumar, per an idea from me. Reviewed also by Andres Freund.
I fixed some whitespace issues, changed a couple of words in a
comment, and made one other minor correction.
Discussion: http://postgr.es/m/CA+TgmoamOtXbVAQf9hWFzonUo6bhhjS6toZQd7HZ-pmojtAmag@mail.gmail.com
Discussion: http://postgr.es/m/CA+Tgmobp7+7kmi4gkq7Y+4AM9fTvL+O1oQ4-5gFTT+6Ng-dQ=g@mail.gmail.com
Discussion: http://postgr.es/m/CAFiTN-vTe79M8uDH1yprOU64MNFE+R3ODRuA+JWf27JbhY4hJw@mail.gmail.com
Andres Freund [Wed, 6 Jul 2022 15:11:39 +0000 (08:11 -0700)]
Remove LLVM_CONFIG from Makefile.global.in
LLVM_CONFIG is only used inside configure.
Reviewed-By: Peter Eisentraut <peter.eisentraut@enterprisedb.com>
Discussion: https://postgr.es/m/
0e81e45c-c9a5-e95b-2782-
ab2dfec8bf57@enterprisedb.com
Peter Eisentraut [Wed, 6 Jul 2022 08:32:40 +0000 (10:32 +0200)]
Tiny capitalization adjustment in pkg-config files
This makes the output match that produced by meson (and also matches
other packages and pkg-config's own documentation better).
David Rowley [Wed, 6 Jul 2022 07:40:32 +0000 (19:40 +1200)]
Remove size increase in ExprEvalStep caused by hashed saops
50e17ad28 increased the size of ExprEvalStep from 64 bytes up to 88 bytes.
Lots of effort was spent during the development of the current expression
evaluation code to make an instance of this struct as small as possible.
Making this struct larger than needed reduces CPU cache efficiency during
expression evaluation which causes noticeable slowdowns during query
execution.
In order to reduce the size of the struct, here we remove the fn_addr
field. The values from this field can be obtained via fcinfo, just with
some extra pointer dereferencing. The extra indirection does not seem to
cause any noticeable slowdowns.
Various other fields have been moved into the ScalarArrayOpExprHashTable
struct. These fields are only used when the ScalarArrayOpExprHashTable
pointer has already been dereferenced, so no additional pointer
dereferences occur for these. Here we also make hash_fcinfo_data the last
field in ScalarArrayOpExprHashTable so that we can avoid a further pointer
dereference to get the FunctionCallInfoBaseData. This also saves a call to
palloc().
50e17ad28 was added in 14, but it's too late to adjust the size of the
ExprEvalStep in that version, so here we just backpatch to 15, which is
currently in beta.
Author: Andres Freund, David Rowley
Discussion: https://postgr.es/m/
20220616233130.rparivafipt6doj3@alap3.anarazel.de
Backpatch-through: 15
Peter Eisentraut [Wed, 6 Jul 2022 06:28:02 +0000 (08:28 +0200)]
Refactor sending of DataRow messages in replication protocol
Some routines open-coded the construction of DataRow messages. Use
TupOutputState struct and associated functions instead, which was
already done in some places.
SendTimeLineHistory() is a bit more complicated and isn't converted by
this.
Reviewed-by: Nathan Bossart <nathandbossart@gmail.com>
Discussion: https://www.postgresql.org/message-id/flat/
7e4fdbdc-699c-4cd0-115d-
fb78a957fc22@enterprisedb.com
Peter Eisentraut [Wed, 6 Jul 2022 05:30:16 +0000 (07:30 +0200)]
Unify DLSUFFIX on Darwin
macOS has traditionally used extension .dylib for shared libraries
(used at build time) and .so for dynamically loaded modules (used by
dlopen()). This complicates the build system a bit. Also, Meson uses
.dylib for both, so it would be worth unifying this in order to be
able to get equal build output.
There doesn't appear to be any reason to use any particular extension
for dlopened modules, since dlopen() will accept anything and
PostgreSQL is well-factored to be able to deal with any extension.
Other software packages that I have handy appear to be about 50/50
split on which extension they use for their plugins. So it seems
possible to change this safely.
Discussion: https://www.postgresql.org/message-id/flat/
bcc45f78-e3c3-8fb3-7c42-
5371b48b5266%40enterprisedb.com
Peter Eisentraut [Wed, 6 Jul 2022 05:16:46 +0000 (07:16 +0200)]
Reorder ecpg_config.h.in alphabetically
This makes it easier to maintain the order if we automate the
generation of this file.
Michael Paquier [Wed, 6 Jul 2022 00:55:30 +0000 (09:55 +0900)]
autho_explain: Add GUC to log query parameters
auto_explain.log_parameter_max_length is a new GUC part of the
extension, similar to the corresponding core setting, that controls the
inclusion of query parameters in the logged explain output.
More tests are added to check the behavior of this new parameter: when
parameters logged in full (the default of -1), when disabled (value of
0) and when partially truncated (value different than the two others).
Author: Dagfinn Ilmari Mannsåker
Discussion: https://postgr.es/m/87ee09mohb.fsf@wibble.ilmari.org
Tom Lane [Tue, 5 Jul 2022 22:23:19 +0000 (18:23 -0400)]
Tighten pg_upgrade's new check for non-upgradable anyarray usages.
We only need to reject cases when the aggregate or operator is
itself declared with a polymorphic type. Per buildfarm.
Discussion: https://postgr.es/m/
3383880.QJadu78ljV@vejsadalnx
Andres Freund [Fri, 17 Jun 2022 19:48:34 +0000 (12:48 -0700)]
pgstat: reduce timer overhead by leaving timer running.
Previously the timer was enabled whenever there were any pending stats after
executing a statement, just to then be disabled again when not idle
anymore. That lead to an increase in GetCurrentTimestamp() calls from within
timeout.c compared to 14.
To avoid that increase, leave the timer enabled until stats are reported,
rather than until idle. The timer is only disabled once the pending stats have
been reported.
For me this fixes the increase in GetCurrentTimestamp() calls, there now are
fewer calls in 15 than in 14, in the previously slowed down workload.
While at it, also update assertion in pgstat_report_stat() to be more precise.
Author: Andres Freund <andres@anarazel.de>
Discussion: https://postgr.es/m/
20220616233130.rparivafipt6doj3@alap3.anarazel.de
Backpatch: 15-
Andres Freund [Fri, 17 Jun 2022 01:33:42 +0000 (18:33 -0700)]
expression eval: Fix EEOP_JSON_CONSTRUCTOR and EEOP_JSONEXPR size.
The new expression step types increased the size of ExprEvalStep by ~4 for all
types of expression steps, slowing down expression evaluation noticeably. Move
them out of line.
There's other issues with these expression steps, but addressing them is
largely independent of this aspect.
Author: Andres Freund <andres@anarazel.de>
Reviewed-By: Andrew Dunstan <andrew@dunslane.net>
Discussion: https://postgr.es/m/
20220616233130.rparivafipt6doj3@alap3.anarazel.de
Backpatch: 15-
Andres Freund [Tue, 5 Jul 2022 18:01:10 +0000 (11:01 -0700)]
Revert 019_replslot_limit.pl related debugging aids.
This reverts most of
91c0570a791,
f28bf667f60,
fe0972ee5e6,
afdeff10526. The
only thing left is the retry loop in 019_replslot_limit.pl that avoids
spurious failures by retrying a couple times.
We haven't seen any hard evidence that this is caused by anything but slow
process shutdown. We did not find any cases where walsenders did not vanish
after waiting for longer. Therefore there's no reason for this debugging code
to remain.
Discussion: https://postgr.es/m/
20220530190155.47wr3x2prdwyciah@alap3.anarazel.de
Backpatch: 15-