Was wrong in commit 1a9388bd.

Reported-by: Tom Lane, Justin Pryzby

Discussion: https://postgr.es/m/20201102063333.GE22691@telsasoft.com

Backpatch-through: 9.5

Summarily changing the STYPE of regression-test aggregates that
depend on array_append or array_cat is an issue for the buildfarm's
cross-version-upgrade tests, because those aggregates (as defined
in the back branches) now won't load into HEAD.  Although this seems
like only a minimal risk for genuine user-defined aggregates, we
need to do something for the buildfarm.  Hence, adjust the aggregate
definitions, in both HEAD and the back branches.

Discussion: https://postgr.es/m/1401824.1604537031@sss.pgh.pa.us
Discussion: https://postgr.es/m/E1kaQ2c-0005lx-Eg@gemulon.postgresql.org

It's entirely possible to see WAL for a relation that doesn't exist in
the target anymore. That happens when the relation was dropped later.
The refactoring in commit eb00f1d4 broke that case, by sanity-checking
the file type in the target before checking the flag forwhether it
exists there at all.

I noticed this during manual testing. Modify the 001_basic.pl test so
that it covers this case.

This option was declared obsolete many years ago.

Reviewed-By: Tom Lane
Discussion: https://postgr.es/m/CABUevEyOE=9CQwZm2j=vwP5+6OLCSoxn9pBjK8gyRdkTzMfqtQ@mail.gmail.com

LLVM 12, to be released in a few months, made some breaking changes to
the Orc JIT interface. OrcV2 eventually will make it easier to support
features like concurrent JIT compilation, but this commit only allows
to compile against LLVM 12.

This commit is a bit bigger than desirable. That partially is because
the V2 interface is more granular than V1 interface, but also because
I chose to make some minor changes to < LLVM 12 code to keep the code
somewhat readable.

The LLVM 12 support will need to be backpatched. I plan to do so after
the patch stewed on the buildfarm for a few days.

Author: Andres Freund
Discussion: https://postgr.es/m/20201016011244.pmyvr3ee2gbzplq4@alap3.anarazel.de

After studying the code, NULLIF is a lot more subtle than you might
have guessed.

Discussion: https://postgr.es/m/160486028730.25500.15740897403028593550@wrigleys.postgresql.org

Remove a CHECK_FOR_INTERRUPTS() call that could never actually handle an
interrupt.  We always have a heap page buffer lock at this point.
Having a useless CHECK_FOR_INTERRUPTS() call is harmless but misleading.

It is probably possible to work around the immediate problem by moving
the CHECK_FOR_INTERRUPTS() to before the heap page buffer lock is
acquired.  That isn't enough to make the function responsive to
interrupts, though.  The index AM caller will still hold an exclusive
buffer lock of its own.

If an interactive psql session used \gset when querying a compromised
server, the attacker could execute arbitrary code as the operating
system account running psql.  Using a prefix not found among specially
treated variables, e.g. every lowercase string, precluded the attack.
Fix by issuing a warning and setting no variable for the column in
question.  Users wanting the old behavior can use a prefix and then a
meta-command like "\set HISTSIZE :prefix_HISTSIZE".  Back-patch to 9.5
(all supported versions).

Reviewed by Robert Haas.  Reported by Nick Cleaton.

Security: CVE-2020-25696

Specifically, this blocks DECLARE ... WITH HOLD and firing of deferred
triggers within index expressions and materialized view queries.  An
attacker having permission to create non-temp objects in at least one
schema could execute arbitrary SQL functions under the identity of the
bootstrap superuser.  One can work around the vulnerability by disabling
autovacuum and not manually running ANALYZE, CLUSTER, REINDEX, CREATE
INDEX, VACUUM FULL, or REFRESH MATERIALIZED VIEW.  (Don't restore from
pg_dump, since it runs some of those commands.)  Plain VACUUM (without
FULL) is safe, and all commands are fine when a trusted user owns the
target object.  Performance may degrade quickly under this workaround,
however.  Back-patch to 9.5 (all supported versions).

Reviewed by Robert Haas.  Reported by Etienne Stalmans.

Security: CVE-2020-25695

Since this script just runs vacuumdb anyway, remove the script and
replace the instructions to run it with instructions to run vacuumdb
directly.

Reviewed-By: Michael Paquier
Discussion: https://postgr.es/m/CABUevEwg5LDFzthhxzSj7sZGMiVsZe0VVNbzzwTQOHJ=rN7+5A@mail.gmail.com

Appears to have been a copy/paste error in the original commit that
moved the messages to fe_utils/.

Author: Tang, Haiying <tanghy.fnst@cn.fujitsu.com>
Backpatch-through: 13
Discussion: https://postgr.es/m/3321cbcea76d4d2c8320a05c19b9304a@G08CNEXMBPEKD05.g08.fujitsu.local

The pg_settings view does not display customized options until
the extension module that defines them has been loaded. This commit
add the note about that behavior, into the docs.

Author: John Naylor
Reviewed-by: Tom Lane, Fujii Masao
Discussion: https://postgr.es/m/CAFBsxsGsBZsG=cLM0Op5HFb2Ks6SzJrOc_eRO_jcKSNuqFRKnQ@mail.gmail.com

Coverity complained about a defect in commit 257836a7:

  Calling "parsePGArray" without checking return value (as is
  done elsewhere 11 out of 13 times).

Fix, and also check for empty strings explicitly (NULL as represented by
PQgetvalue()).  That worked correctly before only because parsePGArray()
happens to set *nitems = 0 when it fails on an empty string.  Also
convert a sanity check assertion to an error to be more paranoid, and
pgindent a nearby line.
Reported-by: Michael Paquier <michael@paquier.xyz>

This back-patches commit 20d3fe90 into the v12 and v13 branches.
At the time I thought that commit was not fixing any observable
bug, but Bertrand Drouvot showed otherwise: adding a dropped column
to the previously-considered scenario crashes v12 and v13, unless the
dropped column happens to be an integer.  That is, of course, because
the tupdesc we derive from the plan output tlist fails to describe
the dropped column accurately, so that we'll do the wrong thing with
a tuple in which that column isn't NULL.

There is no bug in pre-v12 branches because they already did use
the table's real tuple descriptor for any trigger-returned tuple.
It seems that this set of bugs can be blamed on the changes that
removed es_trig_tuple_slot, though I've not attempted to pin that
down precisely.

Although there's no code change needed in HEAD, update the test case
to include a dropped column there too.

Discussion: https://postgr.es/m/db5d97c8-f48a-51e2-7b08-b73d5434d425@amazon.com
Discussion: https://postgr.es/m/16644-5da7ef98a7ac4545@postgresql.org

Commit 257836a7 included an assertion that a version lookup routine is
not trying to look up "C" or "POSIX", but that case is reachable with
the user-facing SQL function pg_collation_actual_version().  Remove the
assertion.

fix for 6be725e7

The description of how LP_DEAD bit setting by index scans works
following commit 2ed5b87f was rather unclear.  Clean that up a bit.

Also refer to LP_DEAD bit setting within _bt_check_unique() at the start
of the same section.  This mechanism may actually be more important than
the generic kill_prior_tuple mechanism that the section focuses on, so
it at least deserves to be mentioned in passing.

* Avoid pointlessly highlighting that an index vacuum was executed by a
  parallel worker; user doesn't care.

* Don't give the impression that a non-concurrent reindex of an invalid
  index on a TOAST table would work, because it wouldn't.

* Add a "translator:" comment for a mysterious message.

Discussion: https://postgr.es/m/20201107034943.GA16596@alvherre.pgsqlReviewed-by: Michael Paquier <michael@paquier.xyz>

A few client tools duplicate error messages already provided by libpq.

Discussion: https://www.postgresql.org/message-id/flat/3e937641-88a1-e697-612e-99bba4b8e5e4%40enterprisedb.com

The buildfarm thinks this leads to memory stomps, though annoyingly
I can't duplicate that here.  The existing code in strings.pgc is
doing something that doesn't seem to be sanctioned at all really
by the documentation, but I'm disinclined to try to make that nicer
right now.  Let's just declare some more output variables in hopes
of working around it.

These were broken in multiple ways:

* The xbstart and xhstart lexer actions neglected to set
"state_before_str_start" before transitioning to the xb/xh states,
thus possibly resulting in "internal error: unreachable state" later.

* The test for valid string contents at the end of xb state was flat out
wrong, as it accounted incorrectly for the "b" prefix that the xbstart
action had injected.  Meanwhile, the xh state had no such check at all.

* The generated literal value failed to include any quote marks.

* The grammar did the wrong thing anyway, typically ignoring the
literal value and emitting something else, since BCONST and XCONST
tokens were handled randomly differently from SCONST tokens.

The first of these problems is evidently an oversight in commit
7f380c59, but the others seem to be very ancient.  The lack of
complaints shows that ECPG users aren't using these syntaxes much
(although I do vaguely remember one previous complaint).

As written, this patch is dependent on 7f380c59, so it can't go
back further than v13.  Given the shortage of complaints, I'm not
excited about adapting the patch to prior branches.

Report and patch by Shenhao Wang (test case adjusted by me)

Discussion: https://postgr.es/m/d6402f1bacb74ecba22ef715dbba17fd@G08CNEXMBPEKD06.g08.fujitsu.local

Move the system catalog index declarations from catalog/indexing.h to
the respective parent tables' catalog/pg_*.h files.  The original
reason for having it split was that the old genbki system produced the
output in the order of the catalog files it read, so all the indexing
stuff needed to come separately.  But this is no longer the case, and
keeping it together makes more sense.
Reviewed-by: John Naylor <john.naylor@enterprisedb.com>
Discussion: https://www.postgresql.org/message-id/flat/c7cc82d6-f976-75d6-2e3e-b03d2cab26bb@2ndquadrant.com

Move the system catalog toast table declarations from
catalog/toasting.h to the respective parent tables' catalog/pg_*.h
files.  The original reason for having it split was that the old
genbki system produced the output in the order of the catalog files it
read, so all the toasting stuff needed to come separately.  But this
is no longer the case, and keeping it together makes more sense.
Reviewed-by: John Naylor <john.naylor@enterprisedb.com>
Discussion: https://www.postgresql.org/message-id/flat/c7cc82d6-f976-75d6-2e3e-b03d2cab26bb@2ndquadrant.com

The list of indexes was being leaked when asked for an index that
doesn't have an index partition in the table partition.  Not a common
case admittedly --and in most cases where it occurs, caller throws an
error anyway-- but worth fixing for cleanliness and in case any
third-party code is calling this function.

While at it, remove use of lfirst_oid() to obtain a value we already
have.

Author: Justin Pryzby <pryzby@telsasoft.com>
Reviewed-by: Michael Paquier <michael@paquier.xyz>
Discussion: https://postgr.es/m/20201105203606.GF22691@telsasoft.com

This should have been done in the initial commit that made
unix_socket_directories a list as of c9b0cbe9.  This change allows to
support correctly the case of ALTER SYSTEM, where it is possible to
specify multiple paths as a list, like the following pattern where
flattening is applied to each item:
ALTER SYSTEM SET unix_socket_directories = '/path1', '/path2';

Any parameters specified in postgresql.conf are parsed the same way, so
there is no compatibility change.  pg_dump has a hardcoded list of
parameters marked with GUC_LIST_QUOTE, that gets its routine update.
These are reordered alphabetically for clarity.

Author: Ian Lawrence Barwick
Reviewed-by: Peter Eisentraunt, Tom Lane, Michael Paquier
Discussion: https://postgr.es/m/CAB8KJ=iMOtNY6_sUwV=LQVCJ2zgYHBDyNzVfvE5GN3WQ3v9kQg@mail.gmail.com

The table generation script would incorrectly complain in the
recomposition sorting when matching code points.  This would not have
caused the generation of an incorrect table.  Note that this condition
is not reachable yet, but could have been reached with future updates.

pg_bswap.h does not need to be included in the frontend.x

Author: John Naylor
Discussion: https://postgr.es/m/CAFBsxsGWmExpvv=61vtDKCs7+kBbhkwBDL2Ph9CacziFKnV_yw@mail.gmail.com

brin_form_tuple failed to consider the values may be toasted, inserting
the toast pointer into the index. This may easily result in index
corruption, as the toast data may be deleted and cleaned up by vacuum.
The cleanup however does not care about indexes, leaving invalid toast
pointers behind, which triggers errors like this:

  ERROR:  missing chunk number 0 for toast value 16433 in pg_toast_16426

A less severe consequence are inconsistent failures due to the index row
being too large, depending on whether brin_form_tuple operated on plain
or toasted version of the row. For example

    CREATE TABLE t (val TEXT);
    INSERT INTO t VALUES ('... long value ...')
    CREATE INDEX idx ON t USING brin (val);

would likely succeed, as the row would likely include toast pointer.
Switching the order of INSERT and CREATE INDEX would likely fail:

    ERROR:  index row size 8712 exceeds maximum 8152 for index "idx"

because this happens before the row values are toasted.

The bug exists since PostgreSQL 9.5 where BRIN indexes were introduced.
So backpatch all the way back.

Author: Tomas Vondra
Reviewed-by: Alvaro Herrera
Backpatch-through: 9.5
Discussion: https://postgr.es/m/20201001184133.oq5uq75sb45pu3aw@development
Discussion: https://postgr.es/m/20201104010544.zexj52mlldagzowv%40development

Revert 59ab4ac3, as well as the followup fix 33862cb9c, in all
branches.  We need to think a bit harder about what the behavior
of LOCK TABLE on views should be, and there's no time for that
before next week's releases.  We'll take another crack at this
later.

Discussion: https://postgr.es/m/16703-e348f58aab3cf6cc@postgresql.org

Revert 403a3d91, as well as the followup fix 7f423503, in all
branches.  We need to think a bit harder about what the behavior
of LOCK TABLE on views should be, and there's no time for that
before next week's releases.  We'll take another crack at this
later.

Discussion: https://postgr.es/m/16703-e348f58aab3cf6cc@postgresql.org

Commit 1e53fe0e changed background processes so that they use
standard SIGHUP handler. Like that, this commit makes autoprewarm leader
process also use standard SIGHUP and SIGTERM handlers, to simplify the code.

Author: Bharath Rupireddy
Reviewed-by: Kyotaro Horiguchi, Fujii Masao
Discussion: https://postgr.es/m/CALj2ACXPorUqePswDtOeM_s82v9RW32E1fYmOPZ5NuE+TWKj_A@mail.gmail.com

Currently only OpenSSL requires this initialization, but in the future
other SSL implementations are likely to need it as well. Abstracting
this functionality out into a separate function makes this cleaner and
more clear, and also removes the dependency on OpenSSL headers from
fork_process.c.

OpenSSL is special in that we need to initialize this random number
generator even if we're not going to use it directly, until we drop
support for everything prior to OpenSSL 1.1.1. (And of course also if we
actually use it). All other implementations are left empty at this time,
but more are expected to be added in the future.

Author: Daniel Gustafsson <daniel@yesql.se>, Michael Paquier <michael@paquier.xyz>
Reviewed-By: Magnus Hagander <magnus@hagander.net>
Discussion: https://postgr.es/m/F6291C3C-747C-4C93-BCE0-28BB420B1FF5@yesql.se

There is no outright bug here but it is better to be consistent with the
usage at other places in the same file. In the passing, fix a wrong
assertion in pgstat_recv_replslot.

Author: Kyotaro Horiguchi
Reviewed-by: Sawada Masahiko and Amit Kapila
Discussion: https://postgr.es/m/20201104.175523.1704166915688949637.horikyota.ntt@gmail.com

wal_consistency_checking indicated an inconsistency in certain cases
involving nbtree page deletion.  The underlying issue is that there was
a minor difference between the page image produced after a REDO routine
ran and the corresponding page image following original execution.

This harmless inconsistency has been around forever.  We more or less
expect total consistency among even deleted nbtree pages these days,
though, so this won't do anymore.

To fix, tweak the REDO routine to match original execution.

Oversight in commit f47b5e13.

LOCK TABLE has complained about "infinite recursion" when applied
to a self-referential view, ever since we made it recurse into views
in v11.  However, that breaks pg_dump's new assumption that it's
okay to lock every relation.  There doesn't seem to be any good
reason to throw an error: if we just abandon the recursion, we've
still satisfied the requirement of locking every referenced relation.

Per bug #16703 from Andrew Bille (via Alexander Lakhin).

Discussion: https://postgr.es/m/16703-e348f58aab3cf6cc@postgresql.org

Logic for counting heap TIDs from posting list tuples (added by commit
0d861bbb) was faulty.  It didn't count any TIDs/index tuples in the
event of no callback being set.  This meant that we incorrectly counted
no index tuples in clean-up only VACUUMs, which could lead to
pg_class.reltuples being spuriously set to 0 in affected indexes.

To fix, go back to counting items from the page in cases where there is
no callback.  This approach isn't very accurate, but it works well
enough in practice while avoiding the expense of accessing every index
tuple during cleanup-only VACUUMs.

Author: Peter Geoghegan <pg@bowt.ie>
Reported-By: Jehan-Guillaume de Rorthais <jgdr@dalibo.com>
https://postgr.es/m/20201023174451.69e358f1@firost
Backpatch: 13-, where nbtree deduplication was introduced

Commit dee663f7 intended to drop any queued up fsync requests before
unlinking segment files, but missed a code path.  Fix, by centralizing
the forget-and-unlink code into a single function.
Reported-by: Tomas Vondra <tomas.vondra@2ndquadrant.com>
Discussion: https://postgr.es/m/20201104013205.icogbi773przyny5%40development

float4_div and float8_div correctly produced zero for zero divided
by infinity, but threw an underflow error for nonzero finite values
divided by infinity.  This seems wrong; at the very least it's
inconsistent with the behavior recently implemented for numeric
infinities.  Remove the error and allow zero to be returned.

This patch also removes a useless isinf() test from the overflow
checks in these functions (non-Inf divided by Inf can't produce Inf).

Extracted from a larger patch; this seems significant outside the
context of geometric operators, so it deserves its own commit.

Kyotaro Horiguchi

Discussion: https://postgr.es/m/CAGf+fX70rWFOk5cd00uMfa__0yP+vtQg5ck7c2Onb-Yczp0URA@mail.gmail.com

Convert array_append, array_prepend, array_cat, array_position,
array_positions, array_remove, array_replace, and width_bucket
to use anycompatiblearray.  This is a simple extension of commit
5c292e6b to hit some other places where there's a pretty obvious
gain in usability from doing so.

Ideally we'd also modify other functions taking multiple old-style
polymorphic arguments.  But most of the remainder are tied into one
or more operator classes, making any such change a much larger can of
worms than I desire to open right now.

Discussion: https://postgr.es/m/77675130-89da-dab1-51dd-492c93dcf5d1@postgresfriends.org

This allows use of a "default" expression that doesn't slavishly
match the data column's type.  Formerly you got something like
"function lag(numeric, integer, integer) does not exist", which
is not just unhelpful but actively misleading.

The SQL spec suggests that the default should be coerced to the data
column's type, but this implementation instead chooses the common
supertype, which seems at least as reasonable.

(Note: I took the opportunity to run "make reformat-dat-files" on
pg_proc.dat, so this commit includes some cosmetic changes to
recently-added entries that aren't related to lead/lag.)

Vik Fearing

Discussion: https://postgr.es/m/77675130-89da-dab1-51dd-492c93dcf5d1@postgresfriends.org