Peter Eisentraut [Sun, 27 Nov 2011 20:42:32 +0000 (22:42 +0200)]
Add pg_upgrade test suite
It runs the regression tests, runs pg_upgrade on the populated
database, and compares the before and after dumps. While not actually
a cross-version upgrade, this does detect omissions and bugs in the
involved tools from time to time. It's also possible to do a
cross-version upgrade by manually supplying parameters.
Tom Lane [Sun, 27 Nov 2011 17:57:11 +0000 (12:57 -0500)]
Use the proper macro to convert a bool to a Datum.
The original coding was
var->value = (Datum) state;
which is bogus, and then in commit 2f0f7b4bce13e68394543728801ef011fd82fac6
it was "corrected" to
var->value = PointerGetDatum(state);
which is a faithful translation but still wrong.
This seems purely cosmetic, though, so no need for a back-patch.
Bruce Momjian [Sun, 27 Nov 2011 03:34:36 +0000 (22:34 -0500)]
Move pg_dump memory routines into pg_dumpmem.c/h and restore common.c
with its original functions. The previous function migration would
cause too many difficulties in back-patching.
Tom Lane [Sat, 26 Nov 2011 19:27:05 +0000 (14:27 -0500)]
Make GiST index searches smarter about queries against empty ranges.
In the cases where the result of the called proc is negated, we should
explicitly test both inputs for empty, to ensure we'll never return "true"
for an unsatisfiable query. In other cases we can rely on the called proc
to say the right thing.
Alvaro Herrera [Fri, 25 Nov 2011 20:56:55 +0000 (17:56 -0300)]
Fix unvalidated check constraints on domains, too
Same bug as reported by Thom Brown for check constraints on tables: the
constraint must be dumped separately from the domain, otherwise it is
restored before the data and thus prevents potentially-violating data
from being loaded in the first place.
Alvaro Herrera [Fri, 25 Nov 2011 15:10:46 +0000 (12:10 -0300)]
Improve logging of autovacuum I/O activity
This adds some I/O stats to the logging of autovacuum (when the
operation takes long enough that log_autovacuum_min_duration causes it
to be logged), so that it is easier to tune. Notably, it adds buffer
I/O counts (hits, misses, dirtied) and read and write rate.
Tom Lane [Fri, 25 Nov 2011 18:58:59 +0000 (13:58 -0500)]
Fix erroneous replay of GIN_UPDATE_META_PAGE WAL records.
A simple thinko in ginRedoUpdateMetapage, namely failing to increment a
loop counter, led to inserting records into the last pending-list page in
the wrong order (the opposite of that intended). So far as I can tell,
this would not upset the code that eventually flushes pending items into
the main part of the GIN index. But it did break the code that searched
the pending list for matches, resulting in transient failure to find
matching entries during index lookups, as illustrated in bug #6307 from
Maksym Boguk.
Back-patch to 8.4 where the incorrect code was introduced.
Robert Haas [Fri, 25 Nov 2011 13:02:10 +0000 (08:02 -0500)]
Move "hot" members of PGPROC into a separate PGXACT array.
This speeds up snapshot-taking and reduces ProcArrayLock contention.
Also, the PGPROC (and PGXACT) structures used by two-phase commit are
now allocated as part of the main array, rather than in a separate
array, and we keep ProcArray sorted in pointer order. These changes
are intended to minimize the number of cache lines that must be pulled
in to take a snapshot, and testing shows a substantial increase in
performance on both read and write workloads at high concurrencies.
Tom Lane [Fri, 25 Nov 2011 04:21:06 +0000 (23:21 -0500)]
Fix unsupported options in CREATE TABLE ... AS EXECUTE.
The WITH [NO] DATA option was not supported, nor the ability to specify
replacement column names; the former limitation wasn't even documented, as
per recent complaint from Naoya Anzai. Fix by moving the responsibility
for supporting these options into the executor. It actually takes less
code this way ...
catversion bump due to change in representation of IntoClause, which might
affect stored rules.
Preserve SQLSTATE when an SPI error is propagated through PL/python
exception handler. This was a regression in 9.1, when the capability
to catch specific SPI errors was added, so backpatch to 9.1.
Mika Eloranta, with some editing by Jan UrbaĆski.
Bruce Momjian [Thu, 24 Nov 2011 03:45:50 +0000 (22:45 -0500)]
Add pg_upgrade ENABLE_SAME_CATVERSION_UPGRADES macro for testing to
allow upgrades of the same catalog version. (Doesn't work for
tablespaces, as indicated by C comment.)
Tom Lane [Wed, 23 Nov 2011 22:13:02 +0000 (17:13 -0500)]
Adjust range_adjacent to support different canonicalization rules.
The original coding would not work for discrete ranges in which the
canonicalization rule is to produce symmetric boundaries (either [] or ()
style), as noted by Jeff Davis. Florian Pflug pointed out that we could
fix that by invoking the canonicalization function to see if the range
"between" the two given ranges normalizes to empty. This implementation
of Florian's idea is a tad slower than the original code, but only in the
case where there actually is a canonicalization function --- if not, it's
essentially the same logic as before.
Tom Lane [Wed, 23 Nov 2011 17:45:49 +0000 (12:45 -0500)]
Creator of a range type must have permission to call support functions.
Since range types can be created by non-superusers, we need to consider
their permissions. Ideally we'd check this when the type is used, not
when it's created, but that seems like much more trouble than it's worth.
The existing restriction that the support functions be immutable already
prevents most cases where an unauthorized call to a function might be
thought a security issue, and the fact that the user has no access to
the results of the system's calls to subtype_diff closes off the other
plausible reason for concern. So this check is basically pro-forma,
but let's make it anyway.
Tom Lane [Wed, 23 Nov 2011 05:03:22 +0000 (00:03 -0500)]
Remove user-selectable ANALYZE option for range types.
It's not clear that a per-datatype typanalyze function would be any more
useful than a generic typanalyze for ranges. What *is* clear is that
letting unprivileged users select typanalyze functions is a crash risk or
worse. So remove the option from CREATE TYPE AS RANGE, and instead put in
a generic typanalyze function for ranges. The generic function does
nothing as yet, but hopefully we'll improve that before 9.2 release.
Tom Lane [Wed, 23 Nov 2011 01:45:05 +0000 (20:45 -0500)]
Remove zero- and one-argument range constructor functions.
Per discussion, the zero-argument forms aren't really worth the catalog
space (just write 'empty' instead). The one-argument forms have some use,
but they also have a serious problem with looking too much like functional
cast notation; to the point where in many real use-cases, the parser would
misinterpret what was wanted.
Committing this as a separate patch, with the thought that we might want
to revert part or all of it if we can think of some way around the cast
ambiguity.
Tom Lane [Tue, 22 Nov 2011 22:45:02 +0000 (17:45 -0500)]
Improve implementation of range-contains-element tests.
Implement these tests directly instead of constructing a singleton range
and then applying range-contains. This saves a range serialize/deserialize
cycle as well as a couple of redundant bound-comparison steps, and adds
very little code on net.
Remove elem_contained_by_range from the GiST opclass: it doesn't belong
there because there is no way to use it in an index clause (where the
indexed column would have to be on the left). Its commutator is in the
opclass, and that's what counts.
Robert Haas [Tue, 22 Nov 2011 21:16:26 +0000 (16:16 -0500)]
Check for INSERT privileges in SELECT INTO / CREATE TABLE AS.
In the normal course of events, this matters only if ALTER DEFAULT
PRIVILEGES has been used to revoke default INSERT permission. Whether
or not the new behavior is more or less likely to be what the user wants
when dealing only with the built-in privilege facilities is arguable,
but it's clearly better when using a loadable module such as sepgsql
that may use the hook in ExecCheckRTPerms to enforce additional
permissions checks.
Tom Lane [Tue, 22 Nov 2011 21:05:49 +0000 (16:05 -0500)]
Still more review for range-types patch.
Per discussion, relax the range input/construction rules so that the
only hard error is lower bound > upper bound. Cases where the lower
bound is <= upper bound, but the range nonetheless normalizes to empty,
are now permitted.
Fix core dump in range_adjacent when bounds are infinite. Marginal
cleanup of regression test cases, some more code commenting.
Simon Riggs [Tue, 22 Nov 2011 09:48:06 +0000 (09:48 +0000)]
Continue to allow VACUUM to mark last block of index dirty
even when there is no work to do. Further analysis required.
Revert of patch c1458cc495ff800cd176a1c2e56d8b62680d9b71
Tom Lane [Mon, 21 Nov 2011 21:19:53 +0000 (16:19 -0500)]
More code review for rangetypes patch.
Fix up some infelicitous coding in DefineRange, and add some missing error
checks. Rearrange operator strategy number assignments for GiST anyrange
opclass so that they don't make such a mess of opr_sanity's table of
operator names associated with different strategy numbers. Assign
hopefully-temporary selectivity estimators to range operators that didn't
have one --- poor as the estimates are, they're still a lot better than the
default 0.5 estimate, and they'll shut up the opr_sanity test that wants to
see selectivity estimators on all built-in operators.
Tom Lane [Mon, 21 Nov 2011 16:24:39 +0000 (11:24 -0500)]
Fix citext upgrade script to update derived copies of pg_type.typcollation.
If the existing citext type has not merely been created, but used in any
tables, then the upgrade script wasn't doing enough. We have to update
attcollation for each citext table column, and indcollation for each citext
index column, as well. Per report from Rudolf van der Leeden.
Tom Lane [Sat, 19 Nov 2011 05:35:29 +0000 (00:35 -0500)]
Avoid floating-point underflow while tracking buffer allocation rate.
When the system is idle for awhile after activity, the "smoothed_alloc"
state variable in BgBufferSync converges slowly to zero. With standard
IEEE float arithmetic this results in several iterations with denormalized
values, which causes kernel traps and annoying log messages on some
poorly-designed platforms. There's no real need to track such small values
of smoothed_alloc, so we can prevent the kernel traps by forcing it to zero
as soon as it's too small to be interesting for our purposes. This issue
is purely cosmetic, since the iterations don't happen fast enough for the
kernel traps to pose any meaningful performance problem, but still it seems
worth shutting up the log messages.
The kernel log messages were previously reported by a number of people,
but kudos to Greg Matthews for tracking down exactly where they were coming
from.
Simon Riggs [Fri, 18 Nov 2011 16:06:53 +0000 (16:06 +0000)]
Avoid marking buffer dirty when VACUUM has no work to do.
When wal_level = 'hot_standby' we touched the last page of the
relation during a VACUUM, even if nothing else had happened.
That would alter the LSN of the last block and set the mtime
of the relation file unnecessarily. Noted by Thom Brown.
Bruce Momjian [Fri, 18 Nov 2011 02:59:04 +0000 (21:59 -0500)]
Remove scandir() requirement in pg_upgrade; instead just use readdir()
--- we were not using the scandir pattern filtering anyway. This also
removes the scandir requirement in configure.
Robert Haas [Fri, 18 Nov 2011 02:31:29 +0000 (21:31 -0500)]
Further consolidation of DROP statement handling.
This gets rid of an impressive amount of duplicative code, with only
minimal behavior changes. DROP FOREIGN DATA WRAPPER now requires object
ownership rather than superuser privileges, matching the documentation
we already have. We also eliminate the historical warning about dropping
a built-in function as unuseful. All operations are now performed in the
same order for all object types handled by dropcmds.c.
Tom Lane [Thu, 17 Nov 2011 23:56:33 +0000 (18:56 -0500)]
Declare range inclusion operators as taking anyelement not anynonarray.
Use of anynonarray was a crude hack to get around ambiguity versus the
array inclusion operators of the same names. My previous patch to extend
the parser's type resolution heuristics makes that unnecessary, so use
the more general declaration instead. This eliminates a wart that these
operators couldn't be used with ranges over arrays, which are otherwise
supported just fine.
Also, mark range_before and range_after as commutator operators,
per discussion with Jeff Davis.
Tom Lane [Thu, 17 Nov 2011 23:28:41 +0000 (18:28 -0500)]
Extend the unknowns-are-same-as-known-inputs type resolution heuristic.
For a very long time, one of the parser's heuristics for resolving
ambiguous operator calls has been to assume that unknown-type literals are
of the same type as the other input (if it's known). However, this was
only used in the first step of quickly checking for an exact-types match,
and thus did not help in resolving matches that require coercion, such as
matches to polymorphic operators. As we add more polymorphic operators,
this becomes more of a problem. This patch adds another use of the same
heuristic as a last-ditch check before failing to resolve an ambiguous
operator or function call. In particular this will let us define the range
inclusion operator in a less limited way (to come in a follow-on patch).
Robert Haas [Thu, 17 Nov 2011 19:20:13 +0000 (14:20 -0500)]
Remove ancient downcasing code from procedural language operations.
A very long time ago, language names were specified as literals rather
than identifiers, so this code was added to do case-folding. But that
style has ben deprecated for many years so this isn't needed any more.
Language names will still be downcased when specified as unquoted
identifiers, but quoted identifiers or the old style using string
literals will be left as-is.
Robert Haas [Thu, 17 Nov 2011 17:41:37 +0000 (12:41 -0500)]
Restructure get_object_address() so it's safe against concurrent DDL.
This gives a much better error message when the object of interest is
concurrently dropped and avoids needlessly failing when the object of
interest is concurrently dropped and recreated. It also improves the
behavior of two concurrent DROP IF EXISTS operations targeted at the
same object; as before, one will drop the object, but now the other
will emit the usual NOTICE indicating that the object does not exist,
instead of rolling back. As a fringe benefit, it's also slightly
less code.
Tom Lane [Wed, 16 Nov 2011 23:21:34 +0000 (18:21 -0500)]
Code review for range-types catalog entries.
Fix assorted infelicities, such as dependency on OIDs that aren't
hardwired, as well as outright misdeclaration of daterange_canonical(),
which resulted in crashes if you invoked it directly. Add some more
regression tests to try to catch similar mistakes in future.
Tom Lane [Tue, 15 Nov 2011 18:05:45 +0000 (13:05 -0500)]
Restructure function-internal caching in the range type code.
Move the responsibility for caching specialized information about range
types into the type cache, so that the catalog lookups only have to occur
once per session. Rearrange APIs a bit so that fn_extra caching is
actually effective in the GiST support code. (Use of OidFunctionCallN is
bad enough for performance in itself, but it also prevents the function
from exploiting fn_extra caching.)
The range I/O functions are still not very bright about caching repeated
lookups, but that seems like material for a separate patch.
Also, avoid unnecessary use of memcpy to fetch/store the range type OID and
flags, and don't use the full range_deserialize machinery when all we need
to see is the flags value.
Also fix API error in range_gist_penalty --- it was failing to set *penalty
for any case involving an empty range.
Tom Lane [Tue, 15 Nov 2011 02:42:04 +0000 (21:42 -0500)]
Fix alignment and toasting bugs in range types.
A range type whose element type has 'd' alignment must have 'd' alignment
itself, else there is no guarantee that the element value can be used
in-place. (Because range_deserialize uses att_align_pointer which forcibly
aligns the given pointer, violations of this rule did not lead to SIGBUS
but rather to garbage data being extracted, as in one of the added
regression test cases.)
Also, you can't put a toast pointer inside a range datum, since the
referenced value could disappear with the range datum still present.
For consistency with the handling of arrays and records, I also forced
decompression of in-line-compressed bound values. It would work to store
them as-is, but our policy is to avoid situations that might result in
double compression.
Add assorted regression tests for this, and bump catversion because of
fixes to built-in pg_type entries.
Also some marginal cleanup of inconsistent/unnecessary error checks.
Tom Lane [Mon, 14 Nov 2011 20:34:39 +0000 (15:34 -0500)]
Return NULL instead of throwing error when desired bound is not available.
Change range_lower and range_upper to return NULL rather than throwing an
error when the input range is empty or the relevant bound is infinite. Per
discussion, throwing an error seems likely to be unduly hard to work with.
Also, this is more consistent with the behavior of the constructors, which
treat NULL as meaning an infinite bound.
Tom Lane [Mon, 14 Nov 2011 20:15:53 +0000 (15:15 -0500)]
Return FALSE instead of throwing error for comparisons with empty ranges.
Change range_before, range_after, range_adjacent to return false rather
than throwing an error when one or both input ranges are empty.
The original definition is unnecessarily difficult to use, and also can
result in undesirable planner failures since the planner could try to
compare an empty range to something else while deriving statistical
estimates. (This was, in fact, the cause of repeatable regression test
failures on buildfarm member jaguar, as well as intermittent failures
elsewhere.)
Also tweak rangetypes regression test to not drop all the objects it
creates, so that the final state of the regression database contains
some rangetype objects for pg_dump testing.
Tom Lane [Mon, 14 Nov 2011 18:59:34 +0000 (13:59 -0500)]
Fix copyright notices, other minor editing in new range-types code.
No functional changes in this commit (except I could not resist the
temptation to re-word a couple of error messages). This is just manual
cleanup after pgindent to make the code look reasonably like other PG
code, in preparation for more detailed code review to come.
Simon Riggs [Sun, 13 Nov 2011 09:00:57 +0000 (09:00 +0000)]
Wakeup WALWriter as needed for asynchronous commit performance.
Previously we waited for wal_writer_delay before flushing WAL. Now
we also wake WALWriter as soon as a WAL buffer page has filled.
Significant effect observed on performance of asynchronous commits
by Robert Haas, attributed to the ability to set hint bits on tuples
earlier and so reducing contention caused by clog lookups.
Tom Lane [Sat, 12 Nov 2011 23:49:09 +0000 (18:49 -0500)]
In plpgsql, allow foreign tables to define row types.
This seems to have been just an oversight in previous foreign-table work.
A quick grep didn't turn up any other places where RELKIND_FOREIGN_TABLE
was obviously omitted.
One change noted by Alexander Soudakov, the other by me.
Back-patch to 9.1.
Robert Haas [Sat, 12 Nov 2011 06:22:45 +0000 (01:22 -0500)]
Avoid retaining multiple relation locks in RangeVarGetRelid.
If it turns out we've locked the wrong OID, release the old lock. In
most cases, it's pretty harmless to retain the extra lock, but this
seems tidier and avoids using lock table slots unnecessarily.
Tom Lane [Thu, 10 Nov 2011 23:36:49 +0000 (18:36 -0500)]
Throw nice error if server is too old to support psql's \ef or \sf command.
Previously, you'd get "function pg_catalog.pg_get_functiondef(integer) does
not exist", which is at best rather unprofessional-looking. Back-patch
to 8.4 where \ef was introduced.
Robert Haas [Thu, 10 Nov 2011 22:54:27 +0000 (17:54 -0500)]
Revert removal of trace_userlocks, because userlocks aren't gone.
This reverts commit 0180bd6180511875db046bf8ddcaa633a2952dfd.
contrib/userlock is gone, but user-level locking still exists,
and is exposed via the pg_advisory* family of functions.
Tom Lane [Thu, 10 Nov 2011 21:08:14 +0000 (16:08 -0500)]
Avoid platform-dependent infinite loop in pg_dump.
If malloc(0) returns NULL, the binary search in findSecLabels() will
probably go into an infinite loop when there are no security labels,
because NULL-1 is greater than NULL after wraparound.
(We've seen this pathology before ... I wonder whether there's a way to
detect the class of bugs automatically?)
Diagnosis and patch by Steve Singer, cosmetic adjustments by me
Forgot to call RestoreBkpBlocks() in the redo-function, as pointed out by
Simon Riggs. In redo of a regular heap insert, it's taken care of in
heap_redo(), but this new record type uses the heap2 RM, and heap2_redo()
does not take care of that for you.
Also, failed to reset the vmbuffer and all_visibile_cleared local variables
after switching to a new buffer.
It used to be cleaned in maintainer-clean, but that is inconsistent
with other cleaning of NLS files in nls-global.mk, and it's also wrong
overall, because it's not part of the distribution tarball, which is
the base definition of the maintainer-clean target.
This greatly reduces the WAL volume, especially when the table is narrow.
The overhead of locking the heap page is also reduced. Reduced WAL traffic
also makes it scale a lot better, if you run multiple COPY processes at
the same time.
Tom Lane [Wed, 9 Nov 2011 04:05:14 +0000 (23:05 -0500)]
Fix random discrepancies between parallel_schedule and serial_schedule.
In particular, my previous patch expected the create_index test to run
before the inherit test; but this was only true in the serial schedule.
Rearrange this portion of the schedules to be more consistent.
Tom Lane [Wed, 9 Nov 2011 02:14:21 +0000 (21:14 -0500)]
Wrap appendrel member outputs in PlaceHolderVars in additional cases.
Add PlaceHolderVar wrappers as needed to make UNION ALL sub-select output
expressions appear non-constant and distinct from each other. This makes
the world safe for add_child_rel_equivalences to do what it does. Before,
it was possible for that function to add identical expressions to different
EquivalenceClasses, which logically should imply merging such ECs, which
would be wrong; or to improperly add a constant to an EquivalenceClass,
drastically changing its behavior. Per report from Teodor Sigaev.
The only currently known consequence of this bug is "MergeAppend child's
targetlist doesn't match MergeAppend" planner failures in 9.1 and later.
I am suspicious that there may be other failure modes that could affect
older release branches; but in the absence of any hard evidence, I'll
refrain from back-patching further than 9.1.
Make DatumGetInetP() unpack inet datums with a 1-byte header, and add
a new macro, DatumGetInetPP(), that does not. This brings these macros
in line with other DatumGet*P() macros.
Backpatch to 8.3, where 1-byte header varlenas were introduced.
Adjust range type docs for some last-minute changes I made to the patch.
non_empty(anyrange) function was removed, empty(anyrange) was renamed to
isempty(anyrange), and !? operators were removed.
Robert Haas [Tue, 8 Nov 2011 02:47:45 +0000 (21:47 -0500)]
Remove hstore's text => text operator.
Since PostgreSQL 9.0, we've emitted a warning message when an operator
named => is created, because the SQL standard now reserves that token
for another use. But we've also shipped such an operator with hstore.
Use of the function hstore(text, text) has been recommended in
preference to =>(text, text). Per discussion, it's now time to take
the next step and stop shipping the operator. This will allow us to
prohibit the use of => as an operator name in a future release if and
when we wish to support the SQL standard use of this token.
The release notes should mention this incompatibility.
Patch by me, reviewed by David Wheeler, Dimitri Fontaine and Tom Lane.
Robert Haas [Tue, 8 Nov 2011 02:39:40 +0000 (21:39 -0500)]
Make VACUUM avoid waiting for a cleanup lock, where possible.
In a regular VACUUM, it's OK to skip pages for which a cleanup lock
isn't immediately available; the next VACUUM will deal with them. If
we're scanning the entire relation to advance relfrozenxid, we might
need to wait, but only if there are tuples on the page that actually
require freezing. These changes should greatly reduce the incidence
of of vacuum processes getting "stuck".
Tom Lane [Mon, 7 Nov 2011 16:48:53 +0000 (11:48 -0500)]
Fix assorted bugs in contrib/unaccent's configuration file parsing.
Make it use t_isspace() to identify whitespace, rather than relying on
sscanf which is known to get it wrong on some platform/locale combinations.
Get rid of fixed-size buffers. Make it actually continue to parse the file
after ignoring a line with untranslatable characters, as was obviously
intended.
The first of these issues is per gripe from J Smith, though not exactly
either of his proposed patches.
Tom Lane [Mon, 7 Nov 2011 01:12:20 +0000 (20:12 -0500)]
On second thought, we'd better just drop these tests altogether.
Further experimentation reveals that my previous change didn't fix the
issue entirely: these tests would still fail at the spring-forward DST
transition. There doesn't seem to be any great value in testing this
specific issue for both timestamp and timestamptz, so just lose the
latter tests.
Tom Lane [Sun, 6 Nov 2011 23:20:26 +0000 (18:20 -0500)]
Un-break horology regression test.
Adjust ill-considered timezone-dependent tests added in commit 8a3d33c8e6c681d512f79af4a521ee0c02befcef so that they won't fail on DST
transition days. Per all-pink buildfarm.
Oops, forgot to fix the catversion when I committed the range types patch.
It was inadvertently changed to 201111111, which is a wrong date. Change it
to current date, and remove the comment that was supposed to remind me to
fix it before committing.