Fix threaded test cases on Windows not to crash in setlocale() which can be
global or local to a thread on Windows.

Author: Christian Ullrich

Our documentation hasn't really caught up with the fact that
non-exclusive backups can now be taken using pg_start_backup and
pg_stop_backup even on standbys.  Update, also correcting some
errors introduced by 52f8a59d.
Updates to the 9.6 documentation are needed as well, but that
will need a separate patch as some things are different on that
version.

David Steele, reviewed by Robert Haas and Michael Paquier

Discussion: http://postgr.es/m/d4d951b9-89c0-6bc1-b6ff-d0b2dd5a8966@pgmasters.net

Force sorting in "C" locale so that the output ordering doesn't vary,
per buildfarm.

In passing, add missing .gitignore entries.

Discussion: https://postgr.es/m/0975f4bb-5dee-c33c-b719-3ce44026d397@chrullrich.net

related to 6ce6a618Reported-by: Christoph Berg <myon@debian.org>

Author: Vinayak Pokale
Reviewed-By: Masahiko Sawada

Minor improvements for commit 1f6d515a.  We do not need the (rather
expensive) test for SRFs in the targetlist, because since v10 any
such SRFs would appear in separate ProjectSet nodes.  Also, make the
code look more like the existing cases by turning it into a simple
recursion --- the argument that there might be some performance
benefit to contorting the code seems unfounded to me, especially since
any good compiler should turn the tail-recursion into iteration anyway.

Discussion: http://postgr.es/m/CADE5jYLuugnEEUsyW6Q_4mZFYTxHxaVCQmGAsF0yiY8ZDggi-w@mail.gmail.com

Test that blessed records can be transferred through a TupleQueue and
correctly decoded by another backend.  While touching the file, make
sure that force_parallel_mode settings only cover relevant tests.

Author: Thomas Munro, editorialized by Andres Freund
Reviewed-By: Andres Freund
Discussion: https://postgr.es/m/20170823054644.efuzftxjpfi6wwqs%40alap3.anarazel.de

Commit 8c0d7baf introduced dshash with hash
and compare functions like DynaHash's, and also variants that take a user
data pointer instead of size.  Simplify the interface by merging them into
a single pair of function pointer types that take both size and a user data
pointer.

Since it is anticipated that memcmp and tag_hash behavior will be a common
requirement, provide wrapper functions dshash_memcmp and dshash_memhash that
conform to the new function types.

Author: Thomas Munro
Reviewed-By: Andres Freund
Discussion: https://postgr.es/m/20170823054644.efuzftxjpfi6wwqs%40alap3.anarazel.de

Tidy-up for commit 8c0d7baf, based on a
complaint from Andres Freund.

Author: Thomas Munro
Reviewed-By: Andres Freund
Discussion: https://postgr.es/m/20170823054644.efuzftxjpfi6wwqs%40alap3.anarazel.de

Commit 16be2fd1 added DSA_ALLOC_HUGE,
DSA_ALLOC_ZERO and DSA_ALLOC_NO_OOM which have the same numerical
values and meanings as the similarly named MCXT_... macros.  In one
place we accidentally used MCXT_ALLOC_NO_OOM when DSA_ALLOC_NO_OOM is
wanted, so tidy that up.

Author: Thomas Munro
Discussion: http://postgr.es/m/CAEepm=2AimHxVkkxnMfQvbZMkXy0uKbVa0-D38c5-qwrCm4CMQ@mail.gmail.com
Backpatch: 10, where dsa was introduced.

Set expanded output when requested through \gx in ExecQueryUsingCursor()
(used when FETCH_COUNT is set).

Discussion: https://www.postgresql.org/message-id/CB7A53AA-5645-4BDD-AB07-4D22CD9D8FF1%40gmx.net
Author: Tobias Bussmann

Remove code meant for upgrading to a particular version of PostgreSQL
9.0.  Since pg_upgrade only supports upgrading to the current major
version, this code is no longer useful.

The original value 12 was set based on RFC 5802 for SCRAM-SHA-1, but RFC
7677 for SCRAM-SHA-256 uses 16, so use that.  (This does not affect the
validity of already stored verifiers.)

Discussion: https://www.postgresql.org/message-id/flat/12cc9297-7e05-932f-d863-765e5626ead4%402ndquadrant.com

The test case added by commit 1f6d515a fails on buildfarm members that
have force_parallel_mode turned on, because we currently don't report sort
performance details from worker processes back to the master.  To fix that,
just make the test table be temp rather than regular; that's a good idea
anyway to forestall any possible interference from auto-analyze.
(The restriction that workers can't access temp tables might go away
someday, but almost certainly not before the other thing gets fixed.)

Also, improve the test so that we retain as much as possible of the
EXPLAIN ANALYZE output.  This aids debugging failures, and might also
expose problems that the preceding version masked.

Discussion: http://postgr.es/m/CADE5jYLuugnEEUsyW6Q_4mZFYTxHxaVCQmGAsF0yiY8ZDggi-w@mail.gmail.com

for commit 237a0b87

Reported-by: Alvaro Herrera <alvherre@2ndquadrant.com>

Author: Thomas Munro <thomas.munro@enterprisedb.com>

This does not use the normal plural handling, because no numbers appear
in the actual message.

Clarify/correct some error messages, fix up some code comments that
confused SASL and SCRAM, and other minor fixes.  No changes in
functionality.

This was erroneously removed in
55a70a02.

Add general purpose chaining hash tables for DSA memory.  Unlike
DynaHash in shared memory mode, these hash tables can grow as
required, and cope with being mapped into different addresses in
different backends.

There is a wide range of potential users for such a hash table, though
it's very likely the interface will need to evolve as we come to
understand the needs of different kinds of users.  E.g support for
iterators and incremental resizing is planned for later commits and
the details of the callback signatures are likely to change.

Author: Thomas Munro
Reviewed-By: John Gorman, Andres Freund, Dilip Kumar, Robert Haas
Discussion:
	https://postgr.es/m/CAEepm=3d8o8XdVwYT6O=bHKsKAM2pu2D6sV1S_=4d+jStVCE7w@mail.gmail.com
	https://postgr.es/m/CAEepm=0ZtQ-SpsgCyzzYpsXS6e=kZWqk3g5Ygn3MDV7A8dabUA@mail.gmail.com

Reported-by: Basil Bourque <basil.bourque@pobox.com>

Previously, tuple descriptors were stored in chains keyed by a fixed size
array of OIDs.  That meant there were effectively two levels of collision
chain -- one inside and one outside the hash table.  Instead, let dynahash.c
look after conflicts for us by supplying a proper hash and equal function
pair.

This is a nice cleanup on its own, but also simplifies followup
changes allowing blessed TupleDescs to be shared between backends
participating in parallel query.

Author: Thomas Munro
Reviewed-By: Andres Freund
Discussion: https://postgr.es/m/CAEepm%3D34GVhOL%2BarUx56yx7OPk7%3DqpGsv3CpO54feqjAwQKm5g%40mail.gmail.com

This hash function is derived from Boost's function of the same name.

Author: Andres Freund, Thomas Munro
Discussion: https://postgr.es/m/CAEepm%3D3rdgjfxW4cKvJ0OEmya2-34B0qHNG1xV0vK7TGPJGMUQ%40mail.gmail.com
Discussion: https://postgr.es/m/20170731210844.3cwrkmsmbbpt4rjc%40alap3.anarazel.de

Users can still create them themselves.  Instead, document Unicode TR 35
collation options for ICU, so users can create all this themselves.
Reviewed-by: Peter Geoghegan <pg@bowt.ie>

Install language+region combinations even if they are not distinct from
the language's base locale.  This gives better long-term stability of
the set of predefined locales and makes the predefined locales less
implementation-dependent and more practical for users.
Reviewed-by: Peter Geoghegan <pg@bowt.ie>

Douglas Doole, reviewed by Ashutosh Bapat and by me.  Minor formatting
change by me.

Discussion: http://postgr.es/m/CADE5jYLuugnEEUsyW6Q_4mZFYTxHxaVCQmGAsF0yiY8ZDggi-w@mail.gmail.com

Periodically while the server is running, and at shutdown, write out a
list of blocks in shared buffers.  When the server reaches consistency
-- unfortunatey, we can't do it before that point without breaking
things -- reload those blocks into any still-unused shared buffers.

Mithun Cy and Robert Haas, reviewed and tested by Beena Emerson,
Amit Kapila, Jim Nasby, and Rafia Sabih.

Discussion: http://postgr.es/m/CAD__OugubOs1Vy7kgF6xTjmEqTR4CrGAv8w+ZbaY_+MZeitukw@mail.gmail.com

It appeared in a conditional that excludes AIX, Cygwin and MinGW.  Give
ICU support a chance to work on those platforms.  Back-patch to v10,
where ICU support was introduced.

TupleDesc's attributes were already stored in contiguous memory after the
struct.  Go one step further and get rid of the array of pointers to
attributes so that they can be stored in shared memory mapped at different
addresses in each backend.  This won't work for TupleDescs with contraints
and defaults, since those point to other objects, but for many purposes
only attributes are needed.

Author: Thomas Munro
Reviewed-By: Andres Freund
Discussion: https://postgr.es/m/CAEepm=0ZtQ-SpsgCyzzYpsXS6e=kZWqk3g5Ygn3MDV7A8dabUA@mail.gmail.com

This is a mechanical change in preparation for a later commit that
will change the layout of TupleDesc.  Introducing a macro to abstract
the details of where attributes are stored will allow us to change
that in separate step and revise it in future.

Author: Thomas Munro, editorialized by Andres Freund
Reviewed-By: Andres Freund
Discussion: https://postgr.es/m/CAEepm=0ZtQ-SpsgCyzzYpsXS6e=kZWqk3g5Ygn3MDV7A8dabUA@mail.gmail.com

Commit 3eb9a5e7 unintentionally introduced an ordering dependency
into restore_toc_entries_prefork().  The existing coding of
reduce_dependencies() contains a check to skip moving a TOC entry
to the ready_list if it wasn't initially in the pending_list.
This used to suffice to prevent reduce_dependencies() from trying to
move anything into the ready_list during restore_toc_entries_prefork(),
because the pending_list stayed empty throughout that phase; but it no
longer does.  The problem doesn't manifest unless the TOC has been
reordered by SortTocFromFile, which is how I missed it in testing.

To fix, just add a test for ready_list == NULL, converting the call
with NULL from a poor man's sanity check into an explicit command
not to touch TOC items' list membership.  Clarify some of the comments
around this; in particular, note the primary purpose of the check for
pending_list membership, which is to ensure that we can't try to restore
the same item twice, in case a TOC list forces it to be restored before
its dependency count goes to zero.

Per report from Fabrízio de Royes Mello.  Back-patch to 9.3, like the
previous commit.

Discussion: https://postgr.es/m/CAFcNs+pjuv0JL_x4+=71TPUPjdLHOXA4YfT32myj_OrrZb4ohA@mail.gmail.com

It would create the comment referring to the keyword-less parent
locale.  This was broken in ddb5fdc0.

Add a new EState member es_leaf_result_relations, so that the trigger
code knows about ResultRelInfos created by tuple routing.  Also make
sure ExplainPrintTriggers knows about partition-related
ResultRelInfos.

Etsuro Fujita, reviewed by Amit Langote

Discussion: http://postgr.es/m/57163e18-8e56-da83-337a-22f2c0008051@lab.ntt.co.jp

That code patch was good as far as it went, but the associated test case
has exposed fundamental brain damage in the parallel scan mechanism,
which is going to take nontrivial work to correct.  In the interests of
getting the buildfarm back to green so that unrelated work can proceed,
let's temporarily remove the test case.

Instead, lock them in the caller using find_all_inheritors so that
they get locked in the standard order, minimizing deadlock risks.

Also in RelationGetPartitionDispatchInfo, avoid opening tables which
are not partitioned; there's no need.

Amit Langote, reviewed by Ashutosh Bapat and Amit Khandekar

Discussion: http://postgr.es/m/91b36fa1-c197-b72f-ca6e-56c593bae68c@lab.ntt.co.jp

Move some logic that is currently in ATExecAttachPartition to
separate functions to facilitate future code reuse.

Ashutosh Bapat and Jeevan Ladhe

Discussion: http://postgr.es/m/CA+Tgmobbnamyvii0pRdg9pp_jLHSUvq7u5SiRrVV0tEFFU58Tg@mail.gmail.com

Per discussion with Tom Lane.

Discussion: http://postgr.es/m/28589.1502902172@sss.pgh.pa.us