We'd glossed over most of this complexity for years, but it's hard
to avoid writing it all down now, so that we can explain what happens
when there's no "posixrules" file in the IANA time zone database.
That was at best a tiny minority situation till now, but it's likely
to become quite common in the future, so we'd better explain it.

Nonetheless, we don't really encourage people to use POSIX zone specs;
picking a named zone is almost always what you really want, unless
perhaps you're stuck with an out-of-date zone database.  Therefore,
let's shove all this detail into an appendix.

Patch by me; thanks to Robert Haas for help with some awkward wording.

Discussion: https://postgr.es/m/1390.1562258309@sss.pgh.pa.us

Advancing a replication slot did not recompute the oldest xmin and LSN
values across replication slots, preventing resource removal like
segments not recycled at checkpoint time.  The original commit that
introduced the slot advancing in 9c7d06d6 never did the update of those
oldest values, and b0afdcad removed this code.

This commit adds a TAP test to check segment recycling with advancing
for physical slots, enforcing an extra segment switch before advancing
to check if the segment gets correctly recycled after a checkpoint.

Reported-by: Andres Freund
Reviewed-by: Alexey Kondratov, Kyptaro Horiguchi
Discussion: https://postgr.es/m/20200609171904.kpltxxvjzislidks@alap3.anarazel.de
Backpatch-through: 11

The previous implementation returned 1 for all negative numbers, which
is not sensible under any definition.

Discussion: https://www.postgresql.org/message-id/flat/6ce1df0e-86a3-e544-743a-f357ff663f68%402ndquadrant.com

Move from int4 to numeric test.  (They were originally int4 functions,
but were reimplemented for numeric in
04a4821a.)  Add some tests for edge
cases.

Discussion: https://www.postgresql.org/message-id/flat/6ce1df0e-86a3-e544-743a-f357ff663f68%402ndquadrant.com

testtablespace is an extra path used as tablespace location in the main
regression test suite, computed from --outputdir as defined by the
caller of pg_regress (current directory if undefined).

This special handling was introduced as of f10589e5 to be specific to
MSVC, as we let pg_regress' Makefile handle this cleanup in other
environments.  This moves the cleanup to the MSVC script running
regression tests instead where needed: check, installcheck and
upgradecheck.  I have also checked this patch on MSVC with repeated runs
of each target.

Author: Kyotaro Horiguchi, Michael Paquier
Discussion: https://postgr.es/m/20200219.142519.437573253063431435.horikyota.ntt@gmail.com

This absorbs a leap-second-related bug fix in localtime.c, and
teaches zic to handle an expiration marker in the leapseconds file.
Neither are of any interest to us (for the foreseeable future
anyway), but we need to stay more or less in sync with upstream.

Also adjust some over-eager changes in the README from commit 95733841.
I have no intention of making changes that require C99 in this code,
until such time as all the live back branches require C99.  Otherwise
back-patching will get too exciting.

For the same reason, absorb assorted whitespace and other cosmetic
changes from HEAD into the back branches; mostly this reflects use of
improved versions of pgindent.

All in all then, quite a boring update.  But I figured I'd get it
done while I was looking at this code.

Oversight in commit 0d861bbb.

Once the counter goes negative we ended up with spinlocks that errored
out on first use (due to check in tas_sema).

Author: Andres Freund
Reviewed-By: Robert Haas
Discussion: https://postgr.es/m/20200606023103.avzrctgv7476xj7i@alap3.anarazel.de
Backpatch: 9.5-

On platforms without support for 64bit atomic operations where we also
cannot rely on 64bit reads to have single copy atomicity, such atomics
are implemented using a spinlock based fallback. That means it's not
safe to even read such atomics from within a signal handler (since the
signal handler might run when the spinlock already is held).

To avoid this issue defer global barrier processing out of the signal
handler. Instead of checking local / shared barrier generation to
determine whether to set ProcSignalBarrierPending, introduce
PROCSIGNAL_BARRIER and always set ProcSignalBarrierPending when
receiving such a signal. Additionally avoid redundant work in
ProcessProcSignalBarrier if ProcSignalBarrierPending is unnecessarily.

Also do a small amount of other polishing.

Author: Andres Freund
Reviewed-By: Robert Haas
Discussion: https://postgr.es/m/20200609193723.eu5ilsjxwdpyxhgz@alap3.anarazel.de
Backpatch: 13-, where the code was introduced.

Don't use fread(), since that doesn't necessarily set errno. We could
use read() instead, but it's even better to use pg_pread(), which
allows us to avoid some extra calls to seek to the desired location in
the file.

Also, advertise a wait event while reading from a file, as we do for
most other places where we're reading data from files.

Patch by me, reviewed by Hamid Akhtar.

Discussion: http://postgr.es/m/CA+TgmobBw-3573vMosGj06r72ajHsYeKtksT_oTxH8XvTL7DxA@mail.gmail.com

Merge two calls to sendDir() that are exactly the same except for
the fifth argument. Adjust comments to match.

Also, don't bother checking whether tblspc_map_file is NULL. We
initialize it in all cases, so it can't be.

Patch by me, reviewed by Amit Kapila and Kyotaro Horiguchi.

Discussion: http://postgr.es/m/CA+TgmoYq+59SJ2zBbP891ngWPA9fymOqntqYcweSDYXS2a620A@mail.gmail.com

Commit 72d422a5 made xlog.c call
sendTablespace() with the 'sizeonly' argument set to true, which
required basebackup.c to export sendTablespace(). However, that's
kind of ugly, so instead defer the call to sendTablespace() until
basebackup.c regains control. That way, it can still be a static
function.

Patch by me, reviewed by Amit Kapila and Kyotaro Horiguchi.

Discussion: http://postgr.es/m/CA+TgmoYq+59SJ2zBbP891ngWPA9fymOqntqYcweSDYXS2a620A@mail.gmail.com

Add a separate enum for use in the locking APIs, which were the only
user.

Discussion: https://www.postgresql.org/message-id/flat/a6f91ead-0ce4-2a34-062b-7ab9813ea308%402ndquadrant.com

The synopsis for PGTYPESinterval_free() used the wrong name.

Discussion: https://postgr.es/m/159231203030.679.3061023914894071953@wrigleys.postgresql.org

Per POSIX this case should produce +0, but buildfarm member fossa
(with icc (ICC) 19.0.5.281 20190815) is reporting -0.  icc has a
boatload of unsafe floating-point optimizations, with a corresponding
boatload of not-too-well-documented compiler switches, and it seems our
default use of "-mp1" isn't whacking it hard enough to keep it from
misoptimizing the stanza in dpow() that checks whether y is odd.
There's nothing wrong with that code (seeing that no other buildfarm
member has trouble with it), so I'm content to blame this on the
compiler.  But without access to the compiler I'm not going to guess at
what switches might be needed to fix it.  For now, tweak the test case
so it will accept either -0 or +0 as a correct answer.

Discussion: https://postgr.es/m/E1jkyFX-0005RR-1Q@gemulon.postgresql.org

Broken by move of fe_archive.c to fe_utils.

I failed to notice that we don't really need to check for y being an
integer in the code path where x = -inf; we already did.

Also make some further cosmetic rearrangements in that spot in hopes
of dodging the seeming compiler bug that buildfarm member fossa is
hitting.  And be consistent about declaring variables as "float8"
not "double", since the pre-existing variables in this function are
like that.

Discussion: https://postgr.es/m/E1jkyFX-0005RR-1Q@gemulon.postgresql.org

It now either returns after it wrote all the data you gave it, or raises
an error.  Not done in back-branches, because it might cause problems
for external code.

Discussion: https://postgr.es/m/CA%2BhUKGJE04G%3D8TLK0DLypT_27D9dR8F1RQgNp0jK6qR0tZGWOw%40mail.gmail.com

Convert buffile.c error handling to use ereport.  This fixes cases where
I/O errors were indistinguishable from EOF or not reported.  Also remove
"%m" from error messages where errno would be bogus.  While we're
modifying those strings, add block numbers and short read byte counts
where appropriate.

Back-patch to all supported releases.
Reported-by: Amit Khandekar <amitdkhan.pg@gmail.com>
Reviewed-by: Melanie Plageman <melanieplageman@gmail.com>
Reviewed-by: Alvaro Herrera <alvherre@2ndquadrant.com>
Reviewed-by: Robert Haas <robertmhaas@gmail.com>
Reviewed-by: Ibrar Ahmed <ibrar.ahmad@gmail.com>
Reviewed-by: Michael Paquier <michael@paquier.xyz>
Discussion: https://postgr.es/m/CA%2BhUKGJE04G%3D8TLK0DLypT_27D9dR8F1RQgNp0jK6qR0tZGWOw%40mail.gmail.com

This has existed for a very long time, equivalent to the ! and !!
operators, but it was never documented.
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://www.postgresql.org/message-id/flat/6ce1df0e-86a3-e544-743a-f357ff663f68%402ndquadrant.com

Non-zero vacuum_defer_cleanup_age values cause pg_upgrade freezing of
the system catalogs to be incomplete, or do nothing.  This will cause
the upgrade to fail in confusing ways.

Reported-by: Laurenz Albe

Discussion: https://postgr.es/m/7d6f6c22ba05ce0c526e9e8b7bfa8105e7da45e6.camel@cybertec.at

Backpatch-through: 9.5

Buildfarm results for commit e532b1d5 reveal the error in my thinking
about the unexpected-EDOM case.  I'd supposed this was no longer really
a live issue, but it seems the fix for glibc's bug #3866 is not all that
old, and we still have at least one buildfarm animal (lapwing) with the
bug.  Hence, resurrect essentially the previous logic (but, I hope, less
opaquely presented), and explain what it is we're really doing here.

Also, blindly try to fix fossa's failure by tweaking the logic that
figures out whether y is an odd integer when x is -inf.  This smells
a whole lot like a compiler bug, but I lack access to icc to try to
pin it down.  Maybe doing division instead of multiplication will
dodge the issue.

Discussion: https://postgr.es/m/E1jkU7H-00024V-NZ@gemulon.postgresql.org

Introduce TAR_BLOCK_SIZE and replace many instances of 512 with
the new constant. Introduce function tarPaddingBytesRequired
and use it to replace numerous repetitions of (x + 511) & ~511.

Add preprocessor guards against multiple inclusion to pgtar.h.

Reformat the prototype for tarCreateHeader so it doesn't extend
beyond 80 characters.

Discussion: http://postgr.es/m/CA+TgmobWbfReO9-XFk8urR1K4wTNwqoHx_v56t7=T8KaiEoKNw@mail.gmail.com

Buildfarm results for commit decbe2bf show that AIX and illumos
have non-POSIX-compliant pow() functions, as do ancient NetBSD
and HPUX releases.  While it's dubious how much we should care
about the latter two platforms, the former two are probably enough
reason to put in manual handling of infinite-input cases.  Hence,
do so, and clean up the post-pow() error handling to reflect its
now-more-limited scope.  (Notably, while we no longer expect to
ever see EDOM from pow(), report it as a domain error if we do.
The former coding had the net effect of expensively converting the
error to ERANGE, which seems highly questionable: if pow() wanted
to report ERANGE, it would have done so.)

Patch by me; thanks to Michael Paquier for review.

Discussion: https://postgr.es/m/E1jkU7H-00024V-NZ@gemulon.postgresql.org

Timestamp can only be an int64 since b9d092c9, and support for WITH OIDS
has been removed as of 578b2297.

Author: Justin Pryzby
Discussion: https://postgr.es/m/20200612023709.GC14879@telsasoft.com

Recently broken by d9fa17aa.

Take some untranslatable things out of the message and replace by
format placeholders, to reduce translatable strings and reduce
translation mistakes.

Oversight in cc072641.

Reported-by: Tom Lane
Discussion: https://postgr.es/m/1098356.1592196242@sss.pgh.pa.us

Our documentation failed to point out that REPEATABLE READ is really
snapshot isolation, which might be important to some users.  Point to
the standard reference paper for this complicated topic.

Likewise, add a reference to the VLDB paper about PostgreSQL SSI, for
technical information about our SSI implementation and how it compares
to S2PL.

While here, add a note about catalog access using a lower isolation
level, per recent user complaint.

Back-patch to all releases.
Reported-by: Kyle Kingsbury <aphyr@jepsen.io>
Reviewed-by: Andres Freund <andres@anarazel.de>
Reviewed-by: Peter Geoghegan <pg@bowt.ie>
Reviewed-by: Tatsuo Ishii <ishii@sraoss.co.jp>
Discussion: https://postgr.es/m/db7b729d-0226-d162-a126-8a8ab2dc4443%40jepsen.io
Discussion: https://postgr.es/m/16454-9408996bb1750faf%40postgresql.org

Previously, these functions tended to throw underflow errors for
negative-infinity exponents.  The correct thing per POSIX is to
return 0, so let's do that instead.  (Note that the SQL standard
is silent on such issues, as it lacks the concepts of either Inf
or NaN; so our practice is to follow POSIX whenever a corresponding
C-library function exists.)

Also, add a bunch of test cases verifying that exp() and power()
actually do follow POSIX for Inf and NaN inputs.  While this patch
should guarantee that exp() passes the tests, power() will not unless
the platform's pow(3) is fully POSIX-compliant.  I already know that
gaur fails some of the tests, and I am suspicious that the Windows
animals will too; the extent of compliance of other old platforms
remains to be seen.  We might choose to drop failing test cases, or
to work harder at overriding pow(3) for these cases, but first let's
see just how good or bad the situation is.

Discussion: https://postgr.es/m/582552.1591917752@sss.pgh.pa.us

The variants for time and timetz had zero test coverage, the variant
for interval only very little.  This adds practically full coverage
for those functions.
Reviewed-by: Vik Fearing <vik@postgresfriends.org>
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://www.postgresql.org/message-id/flat/c3306ac7-fcae-a1b8-1e30-6a379d605bcb%402ndquadrant.com

This patch removes the hardcoded check for superuser privileges when
executing replication origin functions.  Instead, execution is revoked
from public, meaning that those functions can be executed by a superuser
and that access to them can be granted.

Author: Martín Marqués
Reviewed-by: Kyotaro Horiguchi, Michael Paquier, Masahiko Sawada
Discussion: https:/postgr.es/m/CAPdiE1xJMZOKQL3dgHMUrPqysZkgwzSMXETfKkHYnBAB7-0VRQ@mail.gmail.com

var_samp(numeric) and stddev_samp(numeric) disagreed with their float
cousins about what to do for a single non-null input value that is NaN.
The float versions return NULL on the grounds that the calculation is
only defined for more than one non-null input, which seems like the
right answer.  But the numeric versions returned NaN, as a result of
dealing with edge cases in the wrong order.  Fix that.  The patch
also gets rid of an insignificant memory leak in such cases.

This inconsistency is of long standing, but on the whole it seems best
not to back-patch the change into stable branches; nobody's complained
and it's such an obscure point that nobody's likely to complain.
(Note that v13 and v12 now contain test cases that will notice if we
accidentally back-patch this behavior change in future.)

Report and patch by me; thanks to Dean Rasheed for review.

Discussion: https://postgr.es/m/353062.1591898766@sss.pgh.pa.us

When there is just one non-null input value, and it is infinity or NaN,
aggregates such as stddev_pop and covar_pop should produce a NaN
result, because the calculation is not well-defined.  They used to do
so, but since we adopted Youngs-Cramer aggregation in commit e954a727,
they produced zero instead.  That's an oversight, so fix it.  Add tests
exercising these edge cases.

Affected aggregates are

 var_pop(double precision)
 stddev_pop(double precision)
 var_pop(real)
 stddev_pop(real)
 regr_sxx(double precision,double precision)
 regr_syy(double precision,double precision)
 regr_sxy(double precision,double precision)
 regr_r2(double precision,double precision)
 regr_slope(double precision,double precision)
 regr_intercept(double precision,double precision)
 covar_pop(double precision,double precision)
 corr(double precision,double precision)

Back-patch to v12 where the behavior change was accidentally introduced.

Report and patch by me; thanks to Dean Rasheed for review.

Discussion: https://postgr.es/m/353062.1591898766@sss.pgh.pa.us

Reported-By: Tom Lane
Discussion: https://postgr.es/m/841649.1592065060@sss.pgh.pa.us

Make use of the general object support already used by COMMENT, DROP,
and SECURITY LABEL.

Discussion: https://www.postgresql.org/message-id/flat/163c00a5-f634-ca52-fc7c-0e53deda8735%402ndquadrant.com

Unify the grammar of COMMENT, DROP, and SECURITY LABEL further.  They
all effectively just take an object address for later processing, so
we can make the grammar more generalized.  Some extra checking about
which object types are supported can be done later in the statement
execution.

Discussion: https://www.postgresql.org/message-id/flat/163c00a5-f634-ca52-fc7c-0e53deda8735%402ndquadrant.com

Using --outputdir with a custom output repository has never created by
default the sql/ and expected/ paths generated with contents from
respectively input/ and output/ if they don't exist, while the base
output directory gets created if it does not exist.  If sql/ and
expected/ are not present, pg_regress would fail with the path missing,
requiring test scripts to create those extra paths by themselves.  This
commit changes pg_regress so as both get created by default if they do
not exist, removing the need for external test scripts to do so.

This cleans up two code paths in the tree for pg_upgrade tests in MSVC
and environments able to use test.sh.  sql/ and expected/ were created
as part of each test script, but this is not needed anymore as
pg_regress handles the work now.

Author: Roman Zharkov, Daniel Gustafsson
Reviewed-by: Michael Paquier, Tom Lane
Discussion: https://postgr.es/m/16484-4d89e9cc11241996@postgresql.org

This adds two tests for --extra-float-digits and --rows-per-insert,
similar to what exists for --compress.

Author: Dong Wook Lee
Discussion: https://postgr.es/m/CAAcByaJsgrB-qc-ALb0mALprRGLAdmcBap7SZxO4kCAU-JEHcQ@mail.gmail.com