In addition, add support for a "payload" string to be passed along with
each notify event.
This implementation should be significantly more efficient than the old one,
and is also more compatible with Hot Standby usage. There is not yet any
facility for HS slaves to receive notifications generated on the master,
although such a thing is possible in future.
Joachim Wieland, reviewed by Jeff Davis; also hacked on by me.
The purpose of this change is to eliminate the need for every caller
of SearchSysCache, SearchSysCacheCopy, SearchSysCacheExists,
GetSysCacheOid, and SearchSysCacheList to know the maximum number
of allowable keys for a syscache entry (currently 4). This will
make it far easier to increase the maximum number of keys in a
future release should we choose to do so, and it makes the code
shorter, too.
Design and review by Tom Lane.
Move rd_targblock, rd_fsm_nblocks, and rd_vm_nblocks from relcache to the smgr
relation entries, so that they will get reset to InvalidBlockNumber whenever
an smgr-level flush happens. Because we now send smgr invalidation messages
immediately (not at end of transaction) when a relation truncation occurs,
this ensures that other backends will reset their values before they next
access the relation. We no longer need the unreliable assumption that a
VACUUM that's doing a truncation will hold its AccessExclusive lock until
commit --- in fact, we can intentionally release that lock as soon as we've
completed the truncation. This patch therefore reverts (most of) Alvaro's
patch of 2009-11-10, as well as my marginal hacking on it yesterday. We can
also get rid of assorted no-longer-needed relcache flushes, which are far more
expensive than an smgr flush because they kill a lot more state.
In passing this patch fixes smgr_redo's failure to perform visibility-map
truncation, and cleans up some rather dubious assumptions in freespace.c and
visibilitymap.c about when rd_fsm_nblocks and rd_vm_nblocks can be out of
date.
truncating the table and transaction commit. This isn't really making
it safe, but at least there is no good reason to do free space map
cleanup within the risk window. Don't lock out cancel interrupts
until we have to, either.
VACUUM FULL INPLACE), along with a boatload of subsidiary code and complexity.
Per discussion, the use case for this method of vacuuming is no longer large
enough to justify maintaining it; not to mention that we don't wish to invest
the work that would be needed to make it play nicely with Hot Standby.
Aside from the code directly related to old-style VACUUM FULL, this commit
removes support for certain WAL record types that could only be generated
within VACUUM FULL, redirect-pointer removal in heap_page_prune, and
nontransactional generation of cache invalidation sinval messages (the last
being the sticking point for Hot Standby).
We still have to retain all code that copes with finding HEAP_MOVED_OFF and
HEAP_MOVED_IN flag bits on existing tuples. This can't be removed as long
as we want to support in-place update from pre-9.0 databases.
as per my recent proposal.
First, teach IndexBuildHeapScan to not wait for INSERT_IN_PROGRESS or
DELETE_IN_PROGRESS tuples to commit unless the index build is checking
uniqueness/exclusion constraints. If it isn't, there's no harm in just
indexing the in-doubt tuple.
Second, modify VACUUM FULL/CLUSTER to suppress reverifying
uniqueness/exclusion constraint properties while rebuilding indexes of
the target relation. This is reasonable because these commands aren't
meant to deal with corrupted-data situations. Constraint properties
will still be rechecked when an index is rebuilt by a REINDEX command.
This gets us out of the problem that new-style VACUUM FULL would often
wait for other transactions while holding exclusive lock on a system
catalog, leading to probable deadlock because those other transactions
need to look at the catalogs too. Although the real ultimate cause of
the problem is a debatable choice to release locks early after modifying
system catalogs, changing that choice would require pretty serious
analysis and is not something to be undertaken lightly or on a tight
schedule. The present patch fixes the problem in a fairly reasonable
way and should also improve the speed of VACUUM FULL/CLUSTER a little bit.
of shared or nailed system catalogs. This has two key benefits:
* The new CLUSTER-based VACUUM FULL can be applied safely to all catalogs.
* We no longer have to use an unsafe reindex-in-place approach for reindexing
shared catalogs.
CLUSTER on nailed catalogs now works too, although I left it disabled on
shared catalogs because the resulting pg_index.indisclustered update would
only be visible in one database.
Since reindexing shared system catalogs is now fully transactional and
crash-safe, the former special cases in REINDEX behavior have been removed;
shared catalogs are treated the same as non-shared.
This commit does not do anything about the recently-discussed problem of
deadlocks between VACUUM FULL/CLUSTER on a system catalog and other
concurrent queries; will address that in a separate patch. As a stopgap,
parallel_schedule has been tweaked to run vacuum.sql by itself, to avoid
such failures during the regression tests.
of old and new toast tables can be done either at the logical level (by
swapping the heaps' reltoastrelid links) or at the physical level (by swapping
the relfilenodes of the toast tables and their indexes). This is necessary
infrastructure for upcoming changes to support CLUSTER/VAC FULL on shared
system catalogs, where we cannot change reltoastrelid. The physical swap
saves a few catalog updates too.
We unfortunately have to keep the logical-level swap logic because in some
cases we will be adding or deleting a toast table, so there's no possibility
of a physical swap. However, that only happens as a consequence of schema
changes in the table, which we do not need to support for system catalogs,
so such cases aren't an obstacle for that.
In passing, refactor the cluster support functions a little bit to eliminate
unnecessarily-duplicated code; and fix the problem that while CLUSTER had
been taught to rename the final toast table at need, ALTER TABLE had not.
heap_sync() to the callers, because heap_sync() is sometimes called even
if the operation itself is WAL-logged. This eliminates the bogus unlogged
records from CLUSTER that Simon Riggs reported, patch by Fujii Masao.
the relfilenode of currently-not-relocatable system catalogs.
1. Get rid of inval.c's dependency on relfilenode, by not having it emit
smgr invalidations as a result of relcache flushes. Instead, smgr sinval
messages are sent directly from smgr.c when an actual relation delete or
truncate is done. This makes considerably more structural sense and allows
elimination of a large number of useless smgr inval messages that were
formerly sent even in cases where nothing was changing at the
physical-relation level. Note that this reintroduces the concept of
nontransactional inval messages, but that's okay --- because the messages
are sent by smgr.c, they will be sent in Hot Standby slaves, just from a
lower logical level than before.
2. Move setNewRelfilenode out of catalog/index.c, where it never logically
belonged, into relcache.c; which is a somewhat debatable choice as well but
better than before. (I considered catalog/storage.c, but that seemed too
low level.) Rename to RelationSetNewRelfilenode.
3. Cosmetic cleanups of some other relfilenode manipulations.
relations (they don't live in pg_toast). This caused an Assert failure in
assert-enabled builds. So far as I can see, in a non-assert build it would
only have messed up the checks for conflicting names, so a failure would be
quite improbable but perhaps not impossible.
When a column is renamed, we recursively rename the same column in
all descendent tables. But if one of those tables also inherits that
column from a table outside the inheritance hierarchy rooted at the
named table, we must throw an error. The previous coding correctly
prohibited the rename when the parent had inherited the column from
elsewhere, but overlooked the case where the parent was OK but a child
table also inherited the same column from a second, unrelated parent.
For now, not backpatched due to lack of complaints from the field.
KaiGai Kohei, with further changes by me.
Reviewed by Bernd Helme and Tom Lane.
We show the number of buckets, the number of batches (and also the original
number if it has changed), and the peak space used by the hash table. Minor
executor changes to track peak space used.
We need to free the OID list returned by ExecInsertIndexTuples to avoid
a query-lifespan memory leak. When many rows require rechecking, this
can be a significant leak --- it's even more than the space used for the
queued trigger events.
Dean Rasheed
default of "plpgsql". This is more reasonable than it was when the DO patch
was written, because we have since decided that plpgsql should be installed
by default. Per discussion, having a parameter for this doesn't seem useful
enough to justify the risk of application breakage if the value is changed
unexpectedly.
AbortTransaction or AbortSubTransaction, when trying to clean up after an
error that prevented (sub)transaction start from completing:
* access to TopTransactionResourceOwner that might not exist
* assert failure in AtEOXact_GUC, if AtStart_GUC not called yet
* assert failure or core dump in AfterTriggerEndSubXact, if
AfterTriggerBeginSubXact not called yet
Per testing by injecting elog(ERROR) at successive steps in StartTransaction
and StartSubTransaction. It's not clear whether all of these cases could
really occur in the field, but at least one of them is easily exposed by
simple stress testing, as per my accidental discovery yesterday.
Attributes can now have options, just as relations and tablespaces do, and
the reloptions code is used to parse, validate, and store them. For
simplicity and because these options are not performance critical, we store
them in a separate cache rather than the main relcache.
Thanks to Alex Hunsaker for the review.
that would've been WAL-logged if archiving was enabled. If we encounter
such records in archive recovery anyway, we know that some data is
missing from the log. A WARNING is emitted in that case.
Original patch by Fujii Masao, with changes by me.
name already exists, so we'd get an error message about a "type" instead
of about a "relation", because the composite type code shares code with
relation creation.
pg_constraint before searching pg_trigger. This allows saner handling of
corner cases; in particular we now say "constraint is not deferrable"
rather than "constraint does not exist" when the command is applied to
a constraint that's inherently non-deferrable. Per a gripe several months
ago from hubert depesz lubaczewski.
To make this work without breaking user-defined constraint triggers,
we have to add entries for them to pg_constraint. However, in return
we can remove the pgconstrname column from pg_constraint, which represents
a fairly sizable space savings. I also replaced the tgisconstraint column
with tgisinternal; the old meaning of tgisconstraint can now be had by
testing for nonzero tgconstraint, while there is no other way to get
the old meaning of nonzero tgconstraint, namely that the trigger was
internally generated rather than being user-created.
In passing, fix an old misstatement in the docs and comments, namely that
pg_trigger.tgdeferrable is exactly redundant with pg_constraint.condeferrable.
Actually, we mark RI action triggers as nondeferrable even when they belong to
a nominally deferrable FK constraint. The SET CONSTRAINTS code now relies on
that instead of hard-coding a list of exception OIDs.
against concurrent reconnection. Failure during testing showed issue
was possible, even though earlier analysis seemed to indicate it
would not be required. Use LockSharedObjectForSession() before
ResolveRecoveryConflictWithDatabase() and hold lock until end of
processing for that WAL record. Simple approach to avoid introducing
further bugs at this stage of development on an improbable issue.
parse analysis phase, rather than at execution time. This makes parameter
handling work the same as it does in ordinary plannable queries, and in
particular fixes the incompatibility that Pavel pointed out with plpgsql's
new handling of variable references. plancache.c gets a little bit
grottier, but the alternatives seem worse.
This includes two new kinds of postmaster processes, walsenders and
walreceiver. Walreceiver is responsible for connecting to the primary server
and streaming WAL to disk, while walsender runs in the primary server and
streams WAL from disk to the client.
Documentation still needs work, but the basics are there. We will probably
pull the replication section to a new chapter later on, as well as the
sections describing file-based replication. But let's do that as a separate
patch, so that it's easier to see what has been added/changed. This patch
also adds a new section to the chapter about FE/BE protocol, documenting the
protocol used by walsender/walreceivxer.
Bump catalog version because of two new functions,
pg_last_xlog_receive_location() and pg_last_xlog_replay_location(), for
monitoring the progress of replication.
Fujii Masao, with additional hacking by me
of this are to centralise the conflict code to allow further change,
as well as to allow passing through the full reason for the conflict
through to the conflicting backends. Backend state alters how we
can handle different types of conflict so this is now required.
As originally suggested by Heikki, no longer optional.
can upgrade clusters without renaming the tablespace directories. New
directory structure format is, e.g.:
$PGDATA/pg_tblspc/20981/PG_8.5_201001061/719849/83292814
Previously we only cancelled sessions that were in-transaction.
Simple fix is to just cancel all sessions without waiting. Doing
it this way avoids complicating common code paths, which would
not be worth the trouble to cover this rare case.
Problem report and fix by Andres Freund, edited somewhat by me
VACUUM FULL was renamed to VACUUM FULL INPLACE. Also added a new
option -i, --inplace for vacuumdb to perform FULL INPLACE vacuuming.
Since the new VACUUM FULL uses CLUSTER infrastructure, we cannot
use it for system tables. VACUUM FULL for system tables always
fall back into VACUUM FULL INPLACE silently.
Itagaki Takahiro, reviewed by Jeff Davis and Simon Riggs.
This patch only supports seq_page_cost and random_page_cost as parameters,
but it provides the infrastructure to scalably support many more.
In particular, we may want to add support for effective_io_concurrency,
but I'm leaving that as future work for now.
Thanks to Tom Lane for design help and Alvaro Herrera for the review.
decisions about when to auto-analyze.
The previous code depended on n_live_tuples + n_dead_tuples - last_anl_tuples,
where all three of these numbers could be bad estimates from ANALYZE itself.
Even worse, in the presence of a steady flow of HOT updates and matching
HOT-tuple reclamations, auto-analyze might never trigger at all, even if all
three numbers are exactly right, because n_dead_tuples could hold steady.
To fix, replace last_anl_tuples with an accurately tracked count of the total
number of committed tuple inserts + updates + deletes since the last ANALYZE
on the table. This can still be compared to the same threshold as before, but
it's much more trustworthy than the old computation. Tracking this requires
one more intra-transaction counter per modified table within backends, but no
additional memory space in the stats collector. There probably isn't any
measurable speed difference; if anything it might be a bit faster than before,
since I was able to eliminate some per-tuple arithmetic operations in favor of
adding sums once per (sub)transaction.
Also, simplify the logic around pgstat vacuum and analyze reporting messages
by not trying to fold VACUUM ANALYZE into a single pgstat message.
The original thought behind this patch was to allow scheduling of analyzes
on parent tables by artificially inflating their changes_since_analyze count.
I've left that for a separate patch since this change seems to stand on its
own merit.
find_inheritance_children(). This is a complete no-op in databases without
any inheritance. In databases where there are just a few entries in
pg_inherits, it could conceivably be a small loss. However, in databases with
many inheritance parents, it can be a big win.
and teach ANALYZE to compute such stats for tables that have subclasses.
Per my proposal of yesterday.
autovacuum still needs to be taught about running ANALYZE on parent tables
when their subclasses change, but the feature is useful even without that.
PL/pgSQL function within an exception handler. Make sure we use the right
resource owner when we create the tuplestore to hold returned tuples.
Simplify tuplestore API so that the caller doesn't need to be in the right
memory context when calling tuplestore_put* functions. tuplestore.c
automatically switches to the memory context used when the tuplestore was
created. Tuplesort was already modified like this earlier. This patch also
removes the now useless MemoryContextSwitch calls from callers.
Report by Aleksei on pgsql-bugs on Dec 22 2009. Backpatch to 8.1, like
the previous patch that broke this.
Modify pg_dump --binary-upgrade and add backend support routines to
support the preservation of pg_type oids when doing a binary upgrade.
This allows user-defined composite types and arrays to be binary
upgraded.
their underlying table columns. That code was not bright enough to cope with
collision situations (ie, new name conflicts with some other column of the
index). Since there is no functional reason to do this at all, trying to
upgrade the logic to be bulletproof doesn't seem worth the trouble.
This change means that both the index name and the column names of an index
are set when it's created, and won't be automatically changed when the
underlying table columns are renamed. Neatnik DBAs are still free to rename
them manually, of course.
CREATE FOREIGN DATA WRAPPER. Arguably it wasn't a bug because the
documentation said that it's passed the catalog ID or zero, but surely
we should provide it when it's known. And there isn't currently any
scenario where it's not known, and I can't imagine having one in the
future either, so better remove the "or zero" escape hatch and always
pass a valid catalog ID. Backpatch to 8.4.
Martin Pihlak
Index expression columns are now named after the FigureColname result for
their expressions, rather than always being "pg_expression_N". Digits are
appended to this name if needed to make the column name unique within the
index. (That happens for regular columns too, thus fixing the old problem
that CREATE INDEX fooi ON foo (f1, f1) fails. Before exclusion indexes
there was no real reason to do such a thing, but now maybe there is.)
Default names for indexes and associated constraints now include the column
names of all their columns, not only the first one as in previous practice.
(Of course, this will be truncated as needed to fit in NAMEDATALEN. Also,
pkey indexes retain the historical behavior of not naming specific columns
at all.)
An example of the results:
regression=# create table foo (f1 int, f2 text,
regression(# exclude (f1 with =, lower(f2) with =));
NOTICE: CREATE TABLE / EXCLUDE will create implicit index "foo_f1_lower_exclusion" for table "foo"
CREATE TABLE
regression=# \d foo_f1_lower_exclusion
Index "public.foo_f1_lower_exclusion"
Column | Type | Definition
--------+---------+------------
f1 | integer | f1
lower | text | lower(f2)
btree, for table "public.foo"
and composite types, which are the only relkinds for which pg_dump support
exists for dumping column comments. There is no obvious usefulness for
comments on columns of sequences or toast tables; and while comments on
index columns might have some value, it's not worth the risk of compatibility
problems due to possible changes in the algorithm for assigning names to
index columns. Per discussion.
In consequence, remove now-dead code for copying such comments in CREATE TABLE
LIKE.
Rewrite or adjust various comments for clarity. Remove one bogus comment that
doesn't reflect what the code actually does. Improve the description of the
lo_compat_privileges option.
Enabled by recovery_connections = on (default) and forcing archive recovery using a recovery.conf. Recovery processing now emulates the original transactions as they are replayed, providing full locking and MVCC behaviour for read only queries. Recovery must enter consistent state before connections are allowed, so there is a delay, typically short, before connections succeed. Replay of recovering transactions can conflict and in some cases deadlock with queries during recovery; these result in query cancellation after max_standby_delay seconds have expired. Infrastructure changes have minor effects on normal running, though introduce four new types of WAL record.
New test mode "make standbycheck" allows regression tests of static command behaviour on a standby server while in recovery. Typical and extreme dynamic behaviours have been checked via code inspection and manual testing. Few port specific behaviours have been utilised, though primary testing has been on Linux only so far.
This commit is the basic patch. Additional changes will follow in this release to enhance some aspects of behaviour, notably improved handling of conflicts, deadlock detection and query cancellation. Changes to VACUUM FULL are also required.
Simon Riggs, with significant and lengthy review by Heikki Linnakangas, including streamlined redesign of snapshot creation and two-phase commit.
Important contributions from Florian Pflug, Mark Kirkwood, Merlin Moncure, Greg Stark, Gianni Ciolli, Gabriele Bartolini, Hannu Krosing, Robert Haas, Tatsuo Ishii, Hiroyuki Yamada plus support and feedback from many other community members.
ExplainSeparatePlans() was busted for both JSON and YAML output - the present
code is a holdover from the original version of my machine-readable explain
patch, which didn't have the grouping_stack machinery. Also, fix an odd
distribution of labor between ExplainBeginGroup() and ExplainYAMLLineStarting()
when marking lists with "- ", with each providing one character. This broke
the output format for multi-query statements. Also, fix ExplainDummyGroup()
for the YAML output format.
Along the way, make the YAML format use escape_yaml() in situations where the
JSON format uses escape_json(). Right now, it doesn't matter because all the
values are known not to need escaping, but it seems safer this way. Finally,
I added some comments to better explain what the YAML output format is doing.
Greg Sabino Mullane reported the issues with multi-query statements.
Analysis and remaining cleanups by me.
This patch also removes buffer-usage statistics from the track_counts
output, since this (or the global server statistics) is deemed to be a better
interface to this information.
Itagaki Takahiro, reviewed by Euler Taveira de Oliveira.
Without these functions, anyone outside of explain.c can't actually use
ExplainPrintPlan, because the ExplainState won't be initialized properly.
The user-visible result of this was a crash when using auto_explain with
the JSON output format.
Report by Euler Taveira de Oliveira. Analysis by Tom Lane. Patch by me.
an allegedly immutable index function. It was previously recognized that
we had to prevent such a function from executing SET/RESET ROLE/SESSION
AUTHORIZATION, or it could trivially obtain the privileges of the session
user. However, since there is in general no privilege checking for changes
of session-local state, it is also possible for such a function to change
settings in a way that might subvert later operations in the same session.
Examples include changing search_path to cause an unexpected function to
be called, or replacing an existing prepared statement with another one
that will execute a function of the attacker's choosing.
The present patch secures VACUUM, ANALYZE, and CREATE INDEX/REINDEX against
these threats, which are the same places previously deemed to need protection
against the SET ROLE issue. GUC changes are still allowed, since there are
many useful cases for that, but we prevent security problems by forcing a
rollback of any GUC change after completing the operation. Other cases are
handled by throwing an error if any change is attempted; these include temp
table creation, closing a cursor, and creating or deleting a prepared
statement. (In 7.4, the infrastructure to roll back GUC changes doesn't
exist, so we settle for rejecting changes of "search_path" in these contexts.)
Original report and patch by Gurjeet Singh, additional analysis by
Tom Lane.
Security: CVE-2009-4136
support any indexable commutative operator, not just equality. Two rows
violate the exclusion constraint if "row1.col OP row2.col" is TRUE for
each of the columns in the constraint.
Jeff Davis, reviewed by Robert Haas
checked to determine whether the trigger should be fired.
For BEFORE triggers this is mostly a matter of spec compliance; but for AFTER
triggers it can provide a noticeable performance improvement, since queuing of
a deferred trigger event and re-fetching of the row(s) at end of statement can
be short-circuited if the trigger does not need to be fired.
Takahiro Itagaki, reviewed by KaiGai Kohei.
strength of database passwords, and create a sample implementation of
such a hook as a new contrib module "passwordcheck".
Laurenz Albe, reviewed by Takahiro Itagaki
adopted for EXPLAIN. This will allow additional options to be implemented
in future without having to make them fully-reserved keywords. The old syntax
remains available for existing options, however.
Itagaki Takahiro
Per discussion, this should result in defaulting to SQL_ASCII encoding.
The original coding could not support that because it conflated selection
of SQL_ASCII encoding with not being able to determine the encoding.
Adjust pg_get_encoding_from_locale()'s API to distinguish these cases,
and fix callers appropriately. Only initdb actually changes behavior,
since the other callers were perfectly content to consider these cases
equivalent.
Per bug #5178 from Boh Yap. Not going to bother back-patching, since
no one has complained before and there's an easy workaround (namely,
specify the encoding you want).
Apple has fixed that bug in 10.6.2, and we should encourage users to
update to that version rather than trusting this cosmetic patch.
As was recently noted by Stephen Tyler, this patch was only masking
the problem in the context of DROP TABLESPACE, but the failure could
occur in other places such as pg_xlog cleanup.
In VACUUM FULL, an interrupt after the initial transaction has been recorded
as committed can cause postmaster to restart with the following error message:
PANIC: cannot abort transaction NNNN, it was already committed
This problem has been reported many times.
In lazy VACUUM, an interrupt after the table has been truncated by
lazy_truncate_heap causes other backends' relcache to still point to the
removed pages; this can cause future INSERT and UPDATE queries to error out
with the following error message:
could not read block XX of relation 1663/NNN/MMMM: read only 0 of 8192 bytes
The window to this race condition is extremely narrow, but it has been seen in
the wild involving a cancelled autovacuum process.
The solution for both problems is to inhibit interrupts in both operations
until after the respective transactions have been committed. It's not a
complete solution, because the transaction could theoretically be aborted by
some other error, but at least fixes the most common causes of both problems.
it works just as well to have them be ordinary identifiers, and this gets rid
of a number of ugly special cases. Plus we aren't interfering with non-rule
usage of these names.
catversion bump because the names change internally in stored rules.
As proof of concept, modify plpgsql to use the hooks. plpgsql is still
inserting $n symbols textually, but the "back end" of the parsing process now
goes through the ParamRef hook instead of using a fixed parameter-type array,
and then execution only fetches actually-referenced parameters, using a hook
added to ParamListInfo.
Although there's a lot left to be done in plpgsql, this already cures the
"if (TG_OP = 'INSERT' and NEW.foo ...)" problem, as illustrated by the
changed regression test.
that it's called within an AfterTriggerBeginQuery/AfterTriggerEndQuery pair.
The RI cascade triggers suppress that overhead on the assumption that they
are always run non-deferred, so it's possible to violate the condition if
someone mistakenly changes pg_trigger to mark such a trigger deferred.
We don't really care about supporting that, but throwing an error instead
of crashing seems desirable. Per report from Marcelo Costa.
a lot of strange behaviors that occurred in join cases. We now identify the
"current" row for every joined relation in UPDATE, DELETE, and SELECT FOR
UPDATE/SHARE queries. If an EvalPlanQual recheck is necessary, we jam the
appropriate row into each scan node in the rechecking plan, forcing it to emit
only that one row. The former behavior could rescan the whole of each joined
relation for each recheck, which was terrible for performance, and what's much
worse could result in duplicated output tuples.
Also, the original implementation of EvalPlanQual could not re-use the recheck
execution tree --- it had to go through a full executor init and shutdown for
every row to be tested. To avoid this overhead, I've associated a special
runtime Param with each LockRows or ModifyTable plan node, and arranged to
make every scan node below such a node depend on that Param. Thus, by
signaling a change in that Param, the EPQ machinery can just rescan the
already-built test plan.
This patch also adds a prohibition on set-returning functions in the
targetlist of SELECT FOR UPDATE/SHARE. This is needed to avoid the
duplicate-output-tuple problem. It seems fairly reasonable since the
other restrictions on SELECT FOR UPDATE are meant to ensure that there
is a unique correspondence between source tuples and result tuples,
which an output SRF destroys as much as anything else does.
are named in the UPDATE's SET list.
Note: the schema of pg_trigger has not actually changed; we've just started
to use a column that was there all along. catversion bumped anyway so that
this commit is included in the history of potentially interesting changes
to system catalog contents.
Itagaki Takahiro
execMain.c and into a new plan node type LockRows. Like the recent change
to put table updating into a ModifyTable plan node, this increases planning
flexibility by allowing the operations to occur below the top level of the
plan tree. It's necessary in any case to restore the previous behavior of
having FOR UPDATE locking occur before ModifyTable does.
This partially refactors EvalPlanQual to allow multiple rows-under-test
to be inserted into the EPQ machinery before starting an EPQ test query.
That isn't sufficient to fix EPQ's general bogosity in the face of plans
that return multiple rows per test row, though. Since this patch is
mostly about getting some plan node infrastructure in place and not about
fixing ten-year-old bugs, I will leave EPQ improvements for another day.
Another behavioral change that we could now think about is doing FOR UPDATE
before LIMIT, but that too seems like it should be treated as a followon
patch.
They are now handled by a new plan node type called ModifyTable, which is
placed at the top of the plan tree. In itself this change doesn't do much,
except perhaps make the handling of RETURNING lists and inherited UPDATEs a
tad less klugy. But it is necessary preparation for the intended extension of
allowing RETURNING queries inside WITH.
Marko Tiikkaja
Create a new catalog pg_db_role_setting where they are now stored, and better
encapsulate the code that deals with settings into its realm. The old
datconfig and rolconfig columns are removed.
psql has gained a \drds command to display the settings.
Backwards compatibility warning: while the backwards-compatible system views
still have the config columns, they no longer completely represent the
configuration for a user or database.
Catalog version bumped.
Partially revert the previous patch I installed and replace it with a more
general fix: any time a snapshot is pushed as Active, we need to ensure that it
will not be modified in the future. This means that if the same snapshot is
used as CurrentSnapshot, it needs to be copied separately. This affects
serializable transactions only, because CurrentSnapshot has already been copied
by RegisterSnapshot and so PushActiveSnapshot does not think it needs another
copy. However, CommandCounterIncrement would modify CurrentSnapshot, whereas
ActiveSnapshots must not have their command counters incremented.
I say "partially" because the regression test I added for the previous bug
has been kept.
(This restores 8.3 behavior, because before snapmgr.c existed, any snapshot set
as Active was copied.)
Per bug report from Stuart Bishop in
6bc73d4c0910042358k3d1adff3qa36f8df75198ecea@mail.gmail.com
inheritance parent tables are compared using equal(), instead of doing
strcmp() on the nodeToString representation. The old implementation was
always a tad cheesy, and it finally fails completely as of 8.4, now that the
node tree might contain syntax location information. equal() knows it's
supposed to ignore those fields, but strcmp() hardly can. Per recent
report from Scott Ribe.
the privileges that will be applied to subsequently-created objects.
Such adjustments are always per owning role, and can be restricted to objects
created in particular schemas too. A notable benefit is that users can
override the traditional default privilege settings, eg, the PUBLIC EXECUTE
privilege traditionally granted by default for functions.
Petr Jelinek
The old coding was using a regular snapshot, referenced elsewhere, that was
subject to having its command counter updated. Fix by creating a private copy
of the snapshot exclusively for the cursor.
Backpatch to 8.4, which is when the bug was introduced during the snapshot
management rewrite.
hand-assigned rowtype OIDs, even when they are not "bootstrapped" catalogs
that have handmade type rows in pg_type.h. Give pg_database such an OID.
Restore the availability of C macros for the rowtype OIDs of the bootstrapped
catalogs. (These macros are now in the individual catalogs' .h files,
though, not in pg_type.h.)
This commit doesn't do anything especially useful by itself, but it's
necessary infrastructure for reverting some ill-considered changes in
relcache.c.
to create a function for it.
Procedural languages now have an additional entry point, namely a function
to execute an inline code block. This seemed a better design than trying
to hide the transient-ness of the code from the PL. As of this patch, only
plpgsql has an inline handler, but probably people will soon write handlers
for the other standard PLs.
In passing, remove the long-dead LANCOMPILER option of CREATE LANGUAGE.
Petr Jelinek
This is intentionally similar to the recently revised syntax for EXPLAIN
options, ie, (name value, ...). The old syntax is still supported for
backwards compatibility, but we intend that any options added in future
will be provided only in the new syntax.
Robert Haas, Emmanuel Cecchet
If Apple doesn't fix that reasonably soon, we'll have to consider
back-patching a workaround; but for now, just hack it in HEAD so that
we can get buildfarm reports on HEAD from OS X machines.
Per Jan Otto.
functions.
This extends the previous patch that forbade SETting these variables inside
security-definer functions. RESET is equally a security hole, since it
would allow regaining privileges of the caller; furthermore it can trigger
Assert failures and perhaps other internal errors, since the code is not
expecting these variables to change in such contexts. The previous patch
did not cover this case because assign hooks don't really have enough
information, so move the responsibility for preventing this into guc.c.
Problem discovered by Heikki Linnakangas.
Security: no CVE assigned yet, extends CVE-2007-6600
own database's datfrozenxid, if the current value is old enough to be
forcing autovacuums or warning messages. This ensures that a bogus
value is replaced as soon as possible. Per a comment from Heikki.
Recent commits have removed the various uses it was supporting. It was a
performance bottleneck, according to bug report #4919 by Lauris Ulmanis; seems
it slowed down user creation after a billion users.
XID) in checkpoint records. This eliminates the need to recompute the value
from scratch during database startup, which is one of the two remaining
reasons for the flatfile code to exist. It should also simplify life for
hot-standby operation.
To avoid bloating the checkpoint records unreasonably, I switched from
tracking the oldest database by name to tracking it by OID. This turns
out to save cycles in general (everywhere but the warning-generating
paths, which we hardly care about) and also helps us deal with the case
that the oldest database got dropped instead of being vacuumed. The prior
coding might go for a long time without updating the wrap limit in that case,
which is bad because it might result in a lot of useless autovacuum activity.
"all tuples visible" flag in heap page headers. The flag update *must*
be applied before calling XLogInsert, but heap_update and the tuple
moving routines in VACUUM FULL were ignoring this rule. A crash and
replay could therefore leave the flag incorrectly set, causing rows
to appear visible in seqscans when they should not be. This might explain
recent reports of data corruption from Jeff Ross and others.
In passing, do a bit of editorialization on comments in visibilitymap.c.
or previously truncated in the current (sub)transaction. This is safe since
if the (sub)transaction later rolls back, we'd just discard the rel's current
physical file anyway. This avoids unreasonable growth in the number of
transient files when a relation is repeatedly truncated. Per a performance
gripe a couple weeks ago from Todd Cook.
does match some unique index on the referenced table, but that index is
only deferrably unique. We were doing this nicely for the
default-to-primary-key case, but were being lazy for the other case.
Dean Rasheed
to access a Relation entry it had just closed. I happened to be testing with
CLOBBER_CACHE_ALWAYS, which made this a guaranteed core dump (at least on
machines where sprintf %s isn't forgiving of a NULL pointer). It's probably
quite unlikely that it would fail in the field, but a bug is a bug. Fix by
moving the relation_close call down past the logging action.
Both hex format and the traditional "escape" format are automatically
handled on input. The output format is selected by the new GUC variable
bytea_output.
As committed, bytea_output defaults to HEX, which is an *incompatible
change*. We will keep it this way for awhile for testing purposes, but
should consider whether to switch to the more backwards-compatible
default of ESCAPE before 8.5 is released.
Peter Eisentraut
This patch gets us out from under the Unix limitation of two user-defined
signal types. We already had done something similar for signals directed to
the postmaster process; this adds multiplexing for signals directed to
backends and auxiliary processes (so long as they're connected to shared
memory).
As proof of concept, replace the former usage of SIGUSR1 and SIGUSR2
for backends with use of the multiplexing mechanism. There are still some
hard-wired definitions of SIGUSR1 and SIGUSR2 for other process types,
but getting rid of those doesn't seem interesting at the moment.
Fujii Masao
This was foreseen to be a good idea long ago, but nobody had got round
to doing it. The recent patch for deferred unique constraints made
transformConstraintAttrs() ugly enough that I decided it was time.
This change will also greatly simplify parsing of deferred CHECK constraints,
if anyone ever gets around to implementing that.
While at it, add a location field to Constraint, and use that to provide
an error cursor for some of the constraint-related error messages.
The current implementation fires an AFTER ROW trigger for each tuple that
looks like it might be non-unique according to the index contents at the
time of insertion. This works well as long as there aren't many conflicts,
but won't scale to massive unique-key reassignments. Improving that case
is a TODO item.
Dean Rasheed
conindid is the index supporting a constraint. We can use this not only for
unique/primary-key constraints, but also foreign-key constraints, which
depend on the unique index that constrains the referenced columns.
tgconstrindid is just copied from the constraint's conindid field, or is
zero for triggers not associated with constraints.
This is mainly intended as infrastructure for upcoming patches, but it has
some virtue in itself, since it exposes a relationship that you formerly
had to grovel in pg_depend to determine. I simplified one information_schema
view accordingly. (There is a pg_dump query that could also use conindid,
but I left it alone because it wasn't clear it'd get any faster.)
The original syntax made it difficult to add options without making them
into reserved words. This change parenthesizes the options to avoid that
problem, and makes provision for an explicit (and perhaps non-Boolean)
value for each option. The original syntax is still supported, but only
for the two original options ANALYZE and VERBOSE.
As a test case, add a COSTS option that can suppress the planner cost
estimates. This may be useful for including EXPLAIN output in the regression
tests, which are otherwise unable to cope with cross-platform variations in
cost estimates.
Robert Haas
This is believed to not change the output at all, with one known exception:
"Subquery Scan foo" becomes "Subquery Scan on foo". (We can fix that if
anyone complains, but it would be a wart, because the old code was clearly
inconsistent.) The main intention is to remove duplicate coding and
provide a cleaner base for subsequent EXPLAIN patching.
Robert Haas
It appears that, for no particularly good reason, pg_listener.h deviates from
the usual convention for declaring attribute number constants. Normally, it's
#define Anum_{catalog-name}_{column-name} {attribute-number}
pg_listener.h, however substitutes a different string that is similar, but not
the same as, the column name. This change fixes that.
Author: Robert Haas <robertmhaas@gmail.com>
This alters various incidental uses of C++ key words to use other similar
identifiers, so that a C++ compiler won't choke outright. You still
(probably) need extern "C" { }; around the inclusion of backend headers.
based on a patch by Kurt Harriman <harriman@acm.org>
Also add a script cpluspluscheck to check for C++ compatibility in the
future. As of right now, this passes without error for me.
distinction between the external API (parser.h) and declarations that only
need to be visible within the raw parser code (gramparse.h, which now is only
included by parser.c, gram.y, scan.l, and keywords.c). This is in preparation
for the upcoming change to a reentrant lexer, which will require referencing
YYSTYPE in the declarations of base_yylex and filtered_base_yylex, hence
gram.h will have to be included by gramparse.h. We don't want any more files
than absolutely necessary to depend on gram.h, so some cleanup is called for.
types in CREATE TRIGGER. While at it, clean up the amazingly tedious and
inextensible way that the trigger event type list was handled. Per report
from Greg Sabino Mullane.
behavior in cases where we don't know the heap tuple count accurately; in
particular partial vacuum, but this also makes the API a bit more useful
for ANALYZE. This patch adds "estimated_count" flags to both structs so
that an approximate count can be flagged as such, and adjusts the logic
so that approximate counts are not used for updating pg_class.reltuples.
This fixes my previous complaint that VACUUM was putting ridiculous values
into pg_class.reltuples for indexes. The actual impact of that bug is
limited, because the planner only pays attention to reltuples for an index
if the index is partial; which probably explains why beta testers hadn't
noticed a degradation in plan quality from it. But it needs to be fixed.
The whole thing is a bit messy and should be redesigned in future, because
reltuples now has the potential to drift quite far away from reality when
a long period elapses with no non-partial vacuums. But this is as good as
it's going to get for 8.4.
there's no analyzable attributes or indexes. We also used to report 0 live
and dead tuples for such tables, which messed with autovacuum threshold
calculations.
This fixes bug #4812 reported by George Su. Backpatch back to 8.1.
errors when tables are concurrently dropped. To do this we must take lock
on each relation before we check its privileges. The old code was trying
to do that the other way around, which is a bit pointless when there are lots
of other commands that lock relations before checking privileges. I did keep
it checking each relation's privilege before locking the next relation, which
is a detail that ALTER TABLE isn't too picky about.
ability to lock relations as they scan pg_inherits, and to ignore any
relations that have disappeared by the time we get lock on them. This
makes uses of these functions safe against concurrent DROP operations
on child tables: we will effectively ignore any just-dropped child,
rather than possibly throwing an error as in recent bug report from
Thomas Johansson (and similar past complaints). The behavior should
not change otherwise, since the code was acquiring those same locks
anyway, just a little bit later.
An exception is LockTableCommand(), which is still behaving unsafely;
but that seems to require some more discussion before we change it.
find_inheritance_children() and find_all_inheritors(). I got annoyed that
these are buried inside the planner but mostly used elsewhere. So, create
a new file catalog/pg_inherits.c and put them there, along with a couple
of other functions that search pg_inherits.
The code that modifies pg_inherits is (still) in tablecmds.c --- it's
kind of entangled with unrelated code that modifies pg_depend and other
stuff, so pulling it out seemed like a bigger change than I wanted to make
right now. But this file provides a natural home for it if anyone ever
gets around to that.
This commit just moves code around; it doesn't change anything, except
I succumbed to the temptation to make a couple of trivial optimizations
in typeInheritsFrom().
a toast table to be built, even if the sum-of-column-widths calculation
indicates one isn't needed. This is needed by pg_migrator because if the
old table has a toast table, we have to migrate over the toast table since
it might contain some live data, even though subsequent column drops could
mean that no recently-added rows could require toasting.
must be used for the new database, except when copying from template0.
This is the same rule that we now enforce for locale settings, and it has
the same motivation: databases other than template0 might contain data that
would be invalid according to a different setting. This represents another
step in a continuing process of locking down ways in which encoding violations
could occur inside the backend. Per discussion of a few days ago.
In passing, fix pre-existing breakage of mbregress.sh, and fix up a couple
of ereport() calls in dbcommands.c that failed to specify sqlstate codes.
ANALYZE's total sample. The original coding is at risk of overflow for
statistics targets exceeding about 2675; this was not a problem before
8.4 but it is now. Per bug #4793 from Dennis Noordsij.
are individually labeled, rather than just grouped under an "InitPlan"
or "SubPlan" heading. This in turn makes it possible for decompilation of
a subplan reference to usefully identify which subplan it's referencing.
I also made InitPlans identify which parameter symbol(s) they compute,
so that references to those parameters elsewhere in the plan tree can
be connected to the initplan that will be executed. Per a gripe from
Robert Haas about EXPLAIN output of a WITH query being inadequate,
plus some longstanding pet peeves of my own.
of adding optional namespace and action fields to DefElem. Having three
node types that do essentially the same thing bloats the code and leads
to errors of confusion, such as in yesterday's bug report from Khee Chin.
when we are waiting for old snapshots to go away during a concurrent index
build. In particular, this rule lets us avoid waiting for
idle-in-transaction sessions.
This logic could be improved further if we had some way to wake up when
the session we are currently waiting for goes idle-in-transaction. However
that would be a significantly more complex/invasive patch, so it'll have to
wait for some other day.
Simon Riggs, with some improvements by Tom.
temp relations; this is no more expensive than before, now that we have
pg_class.relistemp. Insert tests into bufmgr.c to prevent attempting
to fetch pages from nonlocal temp relations. This provides a low-level
defense against bugs-of-omission allowing temp pages to be loaded into shared
buffers, as in the contrib/pgstattuple problem reported by Stuart Bishop.
While at it, tweak a bunch of places to use new relcache tests (instead of
expensive probes into pg_namespace) to detect local or nonlocal temp tables.
multiple index entries in a holding area before adding them to the main index
structure. This helps because bulk insert is (usually) significantly faster
than retail insert for GIN.
This patch also removes GIN support for amgettuple-style index scans. The
API defined for amgettuple is difficult to support with fastupdate, and
the previously committed partial-match feature didn't really work with
it either. We might eventually figure a way to put back amgettuple
support, but it won't happen for 8.4.
catversion bumped because of change in GIN's pg_am entry, and because
the format of GIN indexes changed on-disk (there's a metapage now,
and possibly a pending list).
Teodor Sigaev
composite, enum and array types, as those are surely not binary-compatible
with anything else because of the embedded OIDs.
Inspired by bug report by Oleg Serov.
function for the specified source and destination encodings. We do that by
calling the function with an empty string. If it can't perform the requested
conversion, it will throw an error.
Backport to 7.4 - 8.3. Per bug report #4680 by Denis Afonin.
wrappers (similar to procedural languages). This way we don't need to retain
the nearly empty libraries, and we are more free in how to implement the
wrapper API in the future.
TABLE: if the command is executed by someone other than the table owner (eg,
a superuser) and the table has a toast table, the toast table's pg_type row
ends up with the wrong typowner, ie, the command issuer not the table owner.
This is quite harmless for most purposes, since no interesting permissions
checks consult the pg_type row. However, it could lead to unexpected failures
if one later tries to drop the role that issued the command (in 8.1 or 8.2),
or strange warnings from pg_dump afterwards (in 8.3 and up, which will allow
the DROP ROLE because we don't create a "redundant" owner dependency for table
rowtypes). Problem identified by Cott Lang.
Back-patch to 8.1. The problem is actually far older --- the CLUSTER variant
can be demonstrated in 7.0 --- but it's mostly cosmetic before 8.1 because we
didn't track ownership dependencies before 8.1. Also, fixing it before 8.1
would require changing the call signature of heap_create_with_catalog(), which
seems to carry a nontrivial risk of breaking add-on modules.
any LISTEN command. This is more important than it used to be because
DISCARD ALL invokes UNLISTEN. Connection-pooled applications making heavy
use of DISCARD ALL were seeing significant contention for pg_listener,
as reported by Matteo Beccati. It seems unlikely that clients using LISTEN
would use pooled connections, so this simple tweak seems sufficient,
especially since the pg_listener implementation is slated to go away soon
anyway.
Back-patch to 8.3, where DISCARD ALL was introduced.
get rid of the OID column. This eliminates the problem discovered by Heikki
back in November that 8.4's suppression of "unnecessary" junk filtering in
INSERT/SELECT could lead to an Assert failure, or storing of oids into a table
that shouldn't have them if Asserts are off. While that particular problem
could have been solved in other ways, it seems likely to be just a forerunner
of things to come if we continue to allow tables to contain rows that disagree
with the pg_class.relhasoids setting. It's better to make this operation slow
than to sacrifice performance or risk bugs in more common code paths.
Also, add ALTER TABLE SET WITH OIDS to rewrite the table to add oids.
This was a bit more controversial, but in view of the very small amount of
extra code needed given the current ALTER TABLE infrastructure, it seems best
to eliminate the asymmetry in features.
has_column_privilege and has_any_column_privilege SQL functions; fix the
information_schema views that are supposed to pay attention to column
privileges; adjust pg_stats to show stats for any column you have select
privilege on; and fix COPY to allow copying a subset of columns if the user
has suitable per-column privileges for all the columns.
To improve efficiency of some of the information_schema views, extend the
has_xxx_privilege functions to allow inquiring about the OR of a set of
privileges in just one call. This is just exposing capability that already
existed in the underlying aclcheck routines.
In passing, make the information_schema views report the owner's own
privileges as being grantable, since Postgres assumes this even when the grant
option bit is not set in the ACL. This is a longstanding oversight.
Also, make the new has_xxx_privilege functions for foreign data objects follow
the same coding conventions used by the older ones.
Stephen Frost and Tom Lane
qualifier, and add support for this in pg_dump.
This allows TOAST tables to have user-defined fillfactor, and will also
enable us to move the autovacuum parameters to reloptions without taking
away the possibility of setting values for TOAST tables.
It's missing in older versions too, but it doesn't seem worth
back-porting. All negative are just harmlessly treated as "no limit", and
tightening the check might even brake an application that relies on it.
pages were marked as clean as well. The idea is to avoid defeating OS
readahead by skipping a page here and there, and also makes it less likely
that we miss an opportunity to advance relfrozenxid, for the sake of only
a few skipped pages.
be used instead of the normal exclusive lock, and make WAL redo functions
responsible for calling RestoreBkpBlocks(). They know better what kind of a
lock they need.
At the moment, this just moves things around with no functional change, but
makes the hot standby patch that's under review cleaner.
CREATE/ALTER/DROP USER MAPPING are now allowed either by the server owner or
by a user with USAGE privileges for his own user name. This is more or less
what the SQL standard wants anyway (plus "implementation-defined")
Hide information_schema.user_mapping_options.option_value, unless the current
user is the one associated with the user mapping, or is the server owner and
the mapping is for PUBLIC, or is a superuser. This is to protect passwords.
Also, fix a bug in information_schema._pg_foreign_servers, which hid servers
using wrappers where the current user did not have privileges on the wrapper.
The correct behavior is to hide servers where the current user has no
privileges on the server.
that are set up for execution with ExecPrepareExpr rather than going through
the full planner process. By introducing an explicit notion of "expression
planning", this patch also lays a bit of groundwork for maybe someday
allowing sub-selects in standalone expressions.
If the table was smaller than REL_TRUNCATE_FRACTION (= 16) pages, we always
tried to acquire AccessExclusiveLock on it even if there was no empty pages
at the end.
Report by Simon Riggs. Back-patch all the way to 7.4.
practically free given prior 8.4 changes in plancache and portal management,
and it makes it a lot easier for ExecutorStart/Run/End hooks to get at the
query text. Extracted from Itagaki Takahiro's pg_stat_statements patch,
with minor editorialization.
so that user-defined window functions are possible. For the moment you'll
have to write them in C, for lack of any interface to the WindowObject API
in the available PLs, but it's better than no support at all.
There was some debate about the best syntax for this. I ended up choosing
the "it's an attribute" position --- the other approach will inevitably be
more work, and the likely market for user-defined window functions is
probably too small to justify it.
This doesn't do any remote or external things yet, but it gives modules
like plproxy and dblink a standardized and future-proof system for
managing their connection information.
Martin Pihlak and Peter Eisentraut
skipped. We could update relpages anyway, but it seems better to only
update it together with reltuples, because we use the reltuples/relpages
ratio in the planner. Also don't update n_live_tuples in pgstat.
ANALYZE in VACUUM ANALYZE now needs to update pg_class, if the
VACUUM-phase didn't do so. Added some boolean-passing to let analyze_rel
know if it should update pg_class or not.
I also moved the relcache invalidation (to update rd_targblock) from
vac_update_relstats to where RelationTruncate is called, because
vac_update_relstats is not called for partial vacuums anymore. It's more
obvious to send the invalidation close to the truncation that requires it.
Per report by Ned T. Crigler.
when they are invoked by the parser. We had been setting up a snapshot at
plan time but really it needs to be done earlier, before parse analysis.
Per report from Dmitry Koterov.
Also fix two related problems discovered while poking at this one:
exec_bind_message called datatype input functions without establishing a
snapshot, and SET CONSTRAINTS IMMEDIATE could call trigger functions without
establishing a snapshot.
Backpatch to 8.2. The underlying problem goes much further back, but it is
masked in 8.1 and before because we didn't attempt to invoke domain check
constraints within datatype input. It would only be exposed if a C-language
datatype input function used the snapshot; which evidently none do, or we'd
have heard complaints sooner. Since this code has changed a lot over time,
a back-patch is hardly risk-free, and so I'm disinclined to patch further
than absolutely necessary.
replication patch needs a signal, but we've already used SIGUSR1 and
SIGUSR2 in normal backends. This patch allows reusing SIGUSR1 for that,
and for other purposes too if the need arises.
the visibility map patch that because autovacuum always sets
VacuumStmt->freeze_min_age, visibility map was never used for autovacuum,
only for manually launched vacuums. This patch introduces a new scan_all
field to VacuumStmt, indicating explicitly whether the visibility map
should be used, or the whole relation should be scanned, to advance
relfrozenxid. Anti-wraparound vacuums still need to scan all pages.
heap page, where a set bit indicates that all tuples on the page are
visible to all transactions, and the page therefore doesn't need
vacuuming. It is stored in a new relation fork.
Lazy vacuum uses the visibility map to skip pages that don't need
vacuuming. Vacuum is also responsible for setting the bits in the map.
In the future, this can hopefully be used to implement index-only-scans,
but we can't currently guarantee that the visibility map is always 100%
up-to-date.
In addition to the visibility map, there's a new PD_ALL_VISIBLE flag on
each heap page, also indicating that all tuples on the page are visible to
all transactions. It's important that this flag is kept up-to-date. It
is also used to skip visibility tests in sequential scans, which gives a
small performance gain on seqscans.
toasted values, since those could get dropped once the cursor's transaction
is over. Per bug #4553 from Andrew Gierth.
Back-patch as far as 8.1. The bug actually exists back to 7.4 when holdable
cursors were introduced, but this patch won't work before 8.1 without
significant adjustments. Given the lack of field complaints, it doesn't seem
worth the work (and risk of introducing new bugs) to try to make a patch for
the older branches.
that a Portal is a useful and sufficient additional argument for
CreateDestReceiver --- it just isn't, in most cases. Instead formalize
the approach of passing any needed parameters to the receiver separately.
One unexpected benefit of this change is that we can declare typedef Portal
in a less surprising location.
This patch is just code rearrangement and doesn't change any functionality.
I'll tackle the HOLD-cursor-vs-toast problem in a follow-on patch.
the basic representational details (typlen, typalign, typbyval, typstorage)
to be copied from an existing type rather than listed explicitly in the
CREATE TYPE command. The immediate reason for this is to provide a simple
solution for add-on modules that want to define types represented as int8,
float4, or float8: as of 8.4 the appropriate PASSEDBYVALUE setting is
platform-specific and so it's hard for a SQL script to know what to do.
This patch fixes the contrib/isn breakage reported by Rushabh Lathia.
truncations in FSM code, call FreeSpaceMapTruncateRel from smgr_redo. To
make that cleaner from modularity point of view, move the WAL-logging one
level up to RelationTruncate, and move RelationTruncate and all the
related WAL-logging to new src/backend/catalog/storage.c file. Introduce
new RelationCreateStorage and RelationDropStorage functions that are used
instead of calling smgrcreate/smgrscheduleunlink directly. Move the
pending rel deletion stuff from smgrcreate/smgrscheduleunlink to the new
functions. This leaves smgr.c as a thin wrapper around md.c; all the
transactional stuff is now in storage.c.
This will make it easier to add new forks with similar truncation logic,
like the visibility map.
* Refactor explain.c slightly to export a convenient-to-use subroutine
for printing EXPLAIN results.
* Provide hooks for plugins to get control at ExecutorStart and ExecutorEnd
as well as ExecutorRun.
* Add some minimal support for tracking the total runtime of ExecutorRun.
This code won't actually do anything unless a plugin prods it to.
* Change the API of the DefineCustomXXXVariable functions to allow nonzero
"flags" to be specified for a custom GUC variable. While at it, also make
the "bootstrap" default value for custom GUCs be explicitly specified as a
parameter to these functions. This is to eliminate confusion over where the
default comes from, as has been expressed in the past by some users of the
custom-variable facility.
* Refactor GUC code a bit to ensure that a custom variable gets initialized to
something valid (like its default value) even if the placeholder value was
invalid.
if the user is superuser. This makes available to extension modules the same
sort of trick being practiced by array_agg(). The reason for the superuser
restriction is that you could crash the system by connecting up an
incompatible pair of internal-using functions as an aggregate. It shouldn't
interfere with any legitimate use, since you'd have to be superuser to create
the internal-using transition and final functions anyway.
returns VOID. This is the last of the easy fixes I recommended in
11870.1218838360@sss.pgh.pa.us --- the others got done awhile ago but
I forgot about this one.
heap_form_tuple. Since this removes the last remaining caller of
heap_addheader, remove it.
Extracted from the column privileges patch from Stephen Frost, with further
code cleanups by me.
("there might be triggers") rather than an exact count. This is necessary
catalog infrastructure for the upcoming patch to reduce the strength of
locking needed for trigger addition/removal. Split out and committed
separately for ease of reviewing/testing.
In passing, also get rid of the unused pg_class columns relukeys, relfkeys,
and relrefs, which haven't been maintained in many years and now have no
chance of ever being maintained (because of wishing to avoid locking).
Simon Riggs
(but not locked, as that would risk deadlocks). Also, make it work in a small
ring of buffers to avoid having bulk inserts trash the whole buffer arena.
Robert Haas, after an idea of Simon Riggs'.
and heap_deformtuple in favor of the newer functions heap_form_tuple et al
(which do the same things but use bool control flags instead of arbitrary
char values). Eliminate the former duplicate coding of these functions,
reducing the deprecated functions to mere wrappers around the newer ones.
We can't get rid of them entirely because add-on modules probably still
contain many instances of the old coding style.
Kris Jurka
functions into one ReadBufferExtended function, that takes the strategy
and mode as argument. There's three modes, RBM_NORMAL which is the default
used by plain ReadBuffer(), RBM_ZERO, which replaces ZeroOrReadBuffer, and
a new mode RBM_ZERO_ON_ERROR, which allows callers to read corrupt pages
without throwing an error. The FSM needs the new mode to recover from
corrupt pages, which could happend if we crash after extending an FSM file,
and the new page is "torn".
Add fork number to some error messages in bufmgr.c, that still lacked it.
backwards scan could actually happen. In particular, pass a flag to
materialize-mode SRFs that tells them whether they need to require random
access. In passing, also suppress unneeded backward-scan overhead for a
Portal's holdStore tuplestore. Per my proposal about reducing I/O costs for
tuplestores.
recent proposal. In typical cases, we now need 12 bytes per insert or delete
event and 16 bytes per update event; previously we needed 40 bytes per
event on 32-bit hardware and 80 bytes per event on 64-bit hardware. Even
in the worst case usage pattern with a large number of distinct triggers being
fired in one query, usage is at most 32 bytes per event. It seems to be a
bit faster than the old code as well, due to reduction of palloc overhead.
This commit doesn't address the TODO item of allowing the event list to spill
to disk; rather it's trying to stave off the need for that. However, it
probably makes that task a bit easier by reducing the data structure's
dependency on pointers. It would now be practical to dump an event list to
disk by "chunks" instead of individual events.
pseudo-type record[] to represent arrays of possibly-anonymous composite
types. Since composite datums carry their own type identification, no
extra knowledge is needed at the array level.
The main reason for doing this right now is that it is necessary to support
the general case of detection of cycles in recursive queries: if you need to
compare more than one column to detect a cycle, you need to compare a ROW()
to an array built from ROW()s, at least if you want to do it as the spec
suggests. Add some documentation and regression tests concerning the cycle
detection issue.
to process any pending unlinks for the source database.
Before, if you dropped a relation in the template database just before
CREATE DATABASE, and a checkpoint happened during copydir(), the checkpoint
might delete a file that we're just about to copy, causing lstat() in
copydir() to fail with ENOENT.
Backpatch to 8.3, where the pending unlinks were introduced.
Per report by Matthew Wakeling and analysis by Tom Lane.
use the old relfilenode in the new tablespace. There might be another relation
in the new tablespace with the same relfilenode, so we must generate a fresh
relfilenode in the new tablespace.
The 8.3 patch to let deleted relation files linger as zero-length files until
the next checkpoint made this more obvious: moving a relation from one table
space another, and then back again, caused a collision with the lingering
file.
Back-patch to 8.1. The issue is present in 8.0 as well, but it doesn't seem
worth fixing there, because we didn't have protection from OID collisions
after OID wraparound before 8.1.
Report by Guillaume Lelarge.
get_name_for_var_field didn't have enough context to interpret a reference to
a CTE query's output. Fixing this requires separate hacks for the regular
deparse case (pg_get_ruledef) and for the EXPLAIN case, since the available
context information is quite different. It's pretty nearly parallel to the
existing code for SUBQUERY RTEs, though. Also, add code to make sure we
qualify a relation name that matches a CTE name; else the CTE will mistakenly
capture the reference when reloading the rule.
In passing, fix a pre-existing problem with get_name_for_var_field not working
on variables in targetlists of SubqueryScan plan nodes. Although latent all
along, this wasn't a problem until we made EXPLAIN VERBOSE try to print
targetlists. To do this, refactor the deparse_context_for_plan API so that
the special case for SubqueryScan is all on ruleutils.c's side.
There are some unimplemented aspects: recursive queries must use UNION ALL
(should allow UNION too), and we don't have SEARCH or CYCLE clauses.
These might or might not get done for 8.4, but even without them it's a
pretty useful feature.
There are also a couple of small loose ends and definitional quibbles,
which I'll send a memo about to pgsql-hackers shortly. But let's land
the patch now so we can get on with other development.
Yoshiyuki Asaba, with lots of help from Tatsuo Ishii and Tom Lane
free space information is stored in a dedicated FSM relation fork, with each
relation (except for hash indexes; they don't use FSM).
This eliminates the max_fsm_relations and max_fsm_pages GUC options; remove any
trace of them from the backend, initdb, and documentation.
Rewrite contrib/pg_freespacemap to match the new FSM implementation. Also
introduce a new variant of the get_raw_page(regclass, int4, int4) function in
contrib/pageinspect that let's you to return pages from any relation fork, and
a new fsm_page_contents() function to inspect the new FSM pages.
element types. Since the backend doesn't actually pay attention to the array
type's delimiter, this has no functional effect, but it seems better for the
catalog entries to be consistent. Per gripe from Greg Mullane and subsequent
discussion.
ctype are now more like encoding, stored in new datcollate and datctype
columns in pg_database.
This is a stripped-down version of Radek Strnad's patch, with further
changes by me.
GetOldestXmin() instead of RecentGlobalXmin; this is safer because we do not
depend on the latter being correctly set elsewhere, and while it is more
expensive, this code path is not performance-critical. This is a real
risk for autovacuum, because it can execute whole cycles without doing
a single vacuum, which would mean that RecentGlobalXmin would stay at its
initialization value, FirstNormalTransactionId, causing a bogus value to be
inserted in pg_database. This bug could explain some recent reports of
failure to truncate pg_clog.
At the same time, change the initialization of RecentGlobalXmin to
InvalidTransactionId, and ensure that it's set to something else whenever
it's going to be used. Using it as FirstNormalTransactionId in HOT page
pruning could incur in data loss. InitPostgres takes care of setting it
to a valid value, but the extra checks are there to prevent "special"
backends from behaving in unusual ways.
Per Tom Lane's detailed problem dissection in 29544.1221061979@sss.pgh.pa.us
SELECT foo.*) so that it cannot be confused with a quoted identifier "*".
Instead create a separate node type A_Star to represent this notation.
Per pgsql-hackers discussion of 2007-Sep-27.
most node types used in expression trees (both before and after parse
analysis). This allows us to place an error cursor in many situations
where we formerly could not, because the information wasn't available
beyond the very first level of parse analysis. There's a fair amount
of work still to be done to persuade individual ereport() calls to actually
include an error location, but this gets the initdb-forcing part of the
work out of the way; and the situation is already markedly better than
before for complaints about unimplementable implicit casts, such as
CASE and UNION constructs with incompatible alternative data types.
Per my proposal of a few days ago.
into nodes/nodeFuncs, so as to reduce wanton cross-subsystem #includes inside
the backend. There's probably more that should be done along this line,
but this is a start anyway.
debug_print_plan to appear at LOG message level, not DEBUG1 as historically.
Make debug_pretty_print default to on. Also, cause plans generated via
EXPLAIN to be subject to debug_print_plan. This is all to make
debug_print_plan a reasonably comfortable substitute for the former behavior
of EXPLAIN VERBOSE.
and anti joins. To do this, pass the SpecialJoinInfo struct for the current
join as an additional optional argument to operator join selectivity
estimation functions. This allows the estimator to tell not only what kind
of join is being formed, but which variable is on which side of the join;
a requirement long recognized but not dealt with till now. This also leaves
the door open for future improvements in the estimators, such as accounting
for the null-insertion effects of lower outer joins. I didn't do anything
about that in the current patch but the information is in principle deducible
from what's passed.
The patch also clarifies the definition of join selectivity for semi/anti
joins: it's the fraction of the left input that has (at least one) match
in the right input. This allows getting rid of some very fuzzy thinking
that I had committed in the original 7.4-era IN-optimization patch.
There's probably room to estimate this better than the present patch does,
but at least we know what to estimate.
Since I had to touch CREATE OPERATOR anyway to allow a variant signature
for join estimator functions, I took the opportunity to add a couple of
additional checks that were missing, per my recent message to -hackers:
* Check that estimator functions return float8;
* Require execute permission at the time of CREATE OPERATOR on the
operator's function as well as the estimator functions;
* Require ownership of any pre-existing operator that's modified by
the command.
I also moved the lookup of the functions out of OperatorCreate() and
into operatorcmds.c, since that seemed more consistent with most of
the other catalog object creation processes, eg CREATE TYPE.
the old JOIN_IN code, but antijoins are new functionality.) Teach the planner
to convert appropriate EXISTS and NOT EXISTS subqueries into semi and anti
joins respectively. Also, LEFT JOINs with suitable upper-level IS NULL
filters are recognized as being anti joins. Unify the InClauseInfo and
OuterJoinInfo infrastructure into "SpecialJoinInfo". With that change,
it becomes possible to associate a SpecialJoinInfo with every join attempt,
which permits some cleanup of join selectivity estimation. That needs to be
taken much further than this patch does, but the next step is to change the
API for oprjoin selectivity functions, which seems like material for a
separate patch. So for the moment the output size estimates for semi and
especially anti joins are quite bogus.
main tables.
This requires vacuum() to accept processing a toast table standalone, so
there's a user-visible change in that it's now possible (for a superuser) to
execute "VACUUM pg_toast.pg_toast_XXX".
of multiple forks, and each fork can be created and grown separately.
The bulk of this patch is about changing the smgr API to include an extra
ForkNumber argument in every smgr function. Also, smgrscheduleunlink and
smgrdounlink no longer implicitly call smgrclose, because other forks might
still exist after unlinking one. The callers of those functions have been
modified to call smgrclose instead.
This patch in itself doesn't have any user-visible effect, but provides the
infrastructure needed for upcoming patches. The additional forks envisioned
are a rewritten FSM implementation that doesn't rely on a fixed-size shared
memory block, and a visibility map to allow skipping portions of a table in
VACUUM that have no dead tuples.
This completes my project of improving usage of hashing for duplicate
elimination (aggregate functions with DISTINCT remain undone, but that's
for some other day).
As with the previous patches, this means we can INTERSECT/EXCEPT on datatypes
that can hash but not sort, and it means that INTERSECT/EXCEPT without ORDER
BY are no longer certain to produce sorted output.
or target database is being accessed by other users, it tells you whether
the "other users" are live sessions or uncommitted prepared transactions.
(Indeed, it tells you exactly how many of each, but that's mostly just
because it was easy to do so.) This should help forestall the gotcha of
not realizing that a prepared transaction is what's blocking the command.
Per discussion.
as per my recent proposal:
1. Fold SortClause and GroupClause into a single node type SortGroupClause.
We were already relying on them to be struct-equivalent, so using two node
tags wasn't accomplishing much except to get in the way of comparing items
with equal().
2. Add an "eqop" field to SortGroupClause to carry the associated equality
operator. This is cheap for the parser to get at the same time it's looking
up the sort operator, and storing it eliminates the need for repeated
not-so-cheap lookups during planning. In future this will also let us
represent GROUP/DISTINCT operations on datatypes that have hash opclasses
but no btree opclasses (ie, they have equality but no natural sort order).
The previous representation simply didn't work for that, since its only
indicator of comparison semantics was a sort operator.
3. Add a hasDistinctOn boolean to struct Query to explicitly record whether
the distinctClause came from DISTINCT or DISTINCT ON. This allows removing
some complicated and not 100% bulletproof code that attempted to figure
that out from the distinctClause alone.
This patch doesn't in itself create any new capability, but it's necessary
infrastructure for future attempts to use hash-based grouping for DISTINCT
and UNION/INTERSECT/EXCEPT.
or domains). This was already effectively required because you had to own
the I/O functions, and the I/O functions pretty much have to be written in
C since we don't let PL functions take or return cstring. But given the
possible security consequences of a malicious type definition, it seems
prudent to enforce superuser requirement directly. Per recent discussion.
only type categories in which the previous coding made *every* type
preferred; so there is no change in effective behavior, because the function
resolution rules only do something different when faced with a choice
between preferred and non-preferred types in the same category. It just
seems safer and less surprising to have CREATE TYPE default to non-preferred
status ...
with system catalog lookups, as was foreseen to be necessary almost since
their creation. Instead put the information into two new pg_type columns,
typcategory and typispreferred. Add support for setting these when
creating a user-defined base type.
The category column is just a "char" (i.e. a poor man's enum), allowing
a crude form of user extensibility of the category list: just use an
otherwise-unused character. This seems sufficient for foreseen uses,
but we could upgrade to having an actual category catalog someday, if
there proves to be a huge demand for custom type categories.
In this patch I have attempted to hew exactly to the behavior of the
previous hardwired logic, except for introducing new type categories for
arrays, composites, and enums. In particular the default preferred state
for user-defined types remains TRUE. That seems worth revisiting, but it
should be done as a separate patch from introducing the infrastructure.
Likewise, any adjustment of the standard set of categories should be done
separately.
a portal are never NULL, but reliably provide the source text of the query.
It turns out that there was only one place that was really taking a short-cut,
which was the 'EXECUTE' utility statement. That doesn't seem like a
sufficiently critical performance hotspot to justify not offering a guarantee
of validity of the portal source text. Fix it to copy the source text over
from the cached plan. Add Asserts in the places that set up cached plans and
portals to reject null source strings, and simplify a bunch of places that
formerly needed to guard against nulls.
There may be a few places that cons up statements for execution without
having any source text at all; I found one such in ConvertTriggerToFK().
It seems sufficient to inject a phony source string in such a case,
for instance
ProcessUtility((Node *) atstmt,
"(generated ALTER TABLE ADD FOREIGN KEY command)",
NULL, false, None_Receiver, NULL);
We should take a second look at the usage of debug_query_string,
particularly the recently added current_query() SQL function.
ITAGAKI Takahiro and Tom Lane
need to deconstruct proargmodes for each pg_proc entry inspected by
FuncnameGetCandidates(). Fixes function lookup performance regression
caused by yesterday's variadic-functions patch.
In passing, make pg_proc.probin be NULL, rather than a dummy value '-',
in cases where it is not actually used for the particular type of function.
This should buy back some of the space cost of the extra column.
so long as all the trailing arguments are of the same (non-array) type.
The function receives them as a single array argument (which is why they
have to all be the same type).
It might be useful to extend this facility to aggregates, but this patch
doesn't do that.
This patch imposes a noticeable slowdown on function lookup --- a follow-on
patch will fix that by adding a redundant column to pg_proc.
Pavel Stehule
SizeOfPageHeaderData instead of sizeof(PageHeaderData) in places where that
makes the code clearer, and avoid casting between Page and PageHeader where
possible. Zdenek Kotala, with some additional cleanup by Heikki Linnakangas.
I did not apply the parts of the proposed patch that would have resulted in
slightly changing the on-disk format of hash indexes; it seems to me that's
not a win as long as there's any chance of having in-place upgrade for 8.4.
Document return type of cast functions.
Also change documentation to prefer the term "binary coercible" in its
present sense instead of the previous term "binary compatible".
of different types than the underlying column. The capability isn't yet
used for anything, but will be required by upcoming patch to analyze
tsvector columns.
Jan Urbanski
corresponding struct definitions. This allows other headers to avoid including
certain highly-loaded headers such as rel.h and relscan.h, instead using just
relcache.h, heapam.h or genam.h, which are more lightweight and thus cause less
unnecessary dependencies.
is necessary to avoid deadlock against ordinary queries, but we'd broken it
with recent changes that made the DROP machinery lock the index before
arriving at index_drop. Per intermittent buildfarm failures.
grammar allows ALTER TABLE/INDEX/SEQUENCE/VIEW interchangeably for all
subforms of those commands, and then we sort out what's really legal
at execution time. This allows the ALTER SEQUENCE/VIEW reference pages
to fully document all the ALTER forms available for sequences and views
respectively, and eliminates a longstanding cause of confusion for users.
The net effect is that the following forms are allowed that weren't before:
ALTER SEQUENCE OWNER TO
ALTER VIEW ALTER COLUMN SET/DROP DEFAULT
ALTER VIEW OWNER TO
ALTER VIEW SET SCHEMA
(There's no actual functionality gain here, but formerly you had to say
ALTER TABLE instead.)
Interestingly, the grammar tables actually get smaller, probably because
there are fewer special cases to keep track of.
I did not disallow using ALTER TABLE for these operations. Perhaps we
should, but there's a backwards-compatibility issue if we do; in fact
it would break existing pg_dump scripts. I did however tighten up
ALTER SEQUENCE and ALTER VIEW to reject non-sequences and non-views
in the new cases as well as a couple of cases where they didn't before.
The patch doesn't change pg_dump to use the new syntaxes, either.
objects are specified, we drop them all in a single performMultipleDeletions
call. This makes the RESTRICT/CASCADE checks more relaxed: it's not counted
as a cascade if one of the later objects has a dependency on an earlier one.
NOTICE messages about such cases go away, too.
In passing, fix the permissions check for DROP CONVERSION, which for some
reason was never made role-aware, and omitted the namespace-owner exemption
too.
Alex Hunsaker, with further fiddling by me.
forks. XLogOpenRelation() and the associated light-weight relation cache in
xlogutils.c is gone, and XLogReadBuffer() now takes a RelFileNode as argument,
instead of Relation.
For functions that still need a Relation struct during WAL replay, there's a
new function called CreateFakeRelcacheEntry() that returns a fake entry like
XLogOpenRelation() used to.
more logical that way, and also it reduces the amount of unnecessary includes
in bufpage.h, which is widely used.
Zdenek Kotala.
My previous patch to bufpage.h should also have credited him as author, but I
forgot (sorry about that).
patches that dealt with object ownership. It wasn't updating pg_shdepend
nor adjusting the aggregate's ACL. In 8.2 and up, fix this permanently
by making it use AlterFunctionOwner_oid. In 8.1, the function code wasn't
factored that way, so just copy and paste.
sequence to be reset to its original starting value. This requires adding the
original start value to the set of parameters (columns) of a sequence object,
which is a user-visible change with potential compatibility implications;
it also forces initdb.
Also add hopefully-SQL-compatible RESTART/CONTINUE IDENTITY options to
TRUNCATE TABLE. RESTART IDENTITY executes ALTER SEQUENCE RESTART for all
sequences "owned by" any of the truncated relations. CONTINUE IDENTITY is
a no-op option.
Zoltan Boszormenyi
functions.
Note that because this patch changes FmgrInfo, any external C functions
you might be testing with 8.4 will need to be recompiled.
Patch by Martin Pihlak, some editorialization by me (principally, removing
tracking of getrusage() numbers)
file portability/instr_time.h, and add a couple more macros to eliminate
some abstraction leakage we formerly had. Also update psql to use this
header instead of its own copy of nearly the same code.
This commit in itself is just code cleanup and shouldn't change anything.
It lays some groundwork for the upcoming function-stats patch, though.
There are two ways to track a snapshot: there's the "registered" list, which
is used for arbitrary long-lived snapshots; and there's the "active stack",
which is used for the snapshot that is considered "active" at any time.
This also allows users of snapshots to stop worrying about snapshot memory
allocation and freeing, and about using PG_TRY blocks around ActiveSnapshot
assignment. This is all done automatically now.
As a consequence, this allows us to reset MyProc->xmin when there are no
more snapshots registered in the current backend, reducing the impact that
long-running transactions have on VACUUM.
unnecessary #include lines in it. Also, move some tuple routine prototypes and
macros to htup.h, which allows removal of heapam.h inclusion from some .c
files.
For this to work, a new header file access/sysattr.h needed to be created,
initially containing attribute numbers of system columns, for pg_dump usage.
While at it, make contrib ltree, intarray and hstore header files more
consistent with our header style.
as those for inherited columns; that is, it's no longer allowed for a child
table to not have a check constraint matching one that exists on a parent.
This satisfies the principle of least surprise (rows selected from the parent
will always appear to meet its check constraints) and eliminates some
longstanding bogosity in pg_dump, which formerly had to guess about whether
check constraints were really inherited or not.
The implementation involves adding conislocal and coninhcount columns to
pg_constraint (paralleling attislocal and attinhcount in pg_attribute)
and refactoring various ALTER TABLE actions to be more like those for
columns.
Alex Hunsaker, Nikhil Sontakke, Tom Lane
checked to see if it's been initialized to all non-nulls. The implicit NOT
NULL constraint was not being checked during the ALTER (in fact, not even if
there was an explicit NOT NULL too), because ATExecAddColumn neglected to
set the flag needed to make the test happen. This has been broken since
the capability was first added, in 8.0.
Brendan Jurd, per a report from Kaloyan Iliev.
where Datum is 8 bytes wide. Since this will break old-style C functions
(those still using version 0 calling convention) that have arguments or
results of these types, provide a configure option to disable it and retain
the old pass-by-reference behavior. Likewise, provide a configure option
to disable the recently-committed float4 pass-by-value change.
Zoltan Boszormenyi, plus configurability stuff by me.
uses of the long-deprecated float32 in contrib/seg; the definitions themselves
are still there, but no longer used. fmgr/README updated to match.
I added a CREATE FUNCTION to account for existing seg_center() code in seg.c
too, and some tests for it and the neighbor functions. At the same time,
remove checks for NULL which are not needed (because the functions are declared
STRICT).
I had to do some adjustments to contrib's btree_gist too. The choices for
representation there are not ideal for changing the underlying types :-(
Original patch by Zoltan Boszormenyi, with some adjustments by me.
file; the idea is that we should clean up as much as we can, even if there's
some problem removing one file. Make the error messages a bit less misleading,
too. In passing, const-ify function arguments.
place to prevent reusing relation OIDs before next checkpoint, and DROP
DATABASE. First, if a database was dropped, bgwriter would still try to unlink
the files that the rmtree() call by the DROP DATABASE command has already
deleted, or is just about to delete. Second, if a database is dropped, and
another database is created with the same OID, bgwriter would in the worst
case delete a relation in the new database that happened to get the same OID
as a dropped relation in the old database.
To fix these race conditions:
- make rmtree() ignore ENOENT errors. This fixes the 1st race condition.
- make ForgetDatabaseFsyncRequests forget unlink requests as well.
- force checkpoint on in dropdb on all platforms
Since ForgetDatabaseFsyncRequests() is asynchronous, the 2nd change isn't
enough on its own to fix the problem of dropping and creating a database with
same OID, but forcing a checkpoint on DROP DATABASE makes it sufficient.
Per Tom Lane's bug report and proposal. Backpatch to 8.3.
of each plan node, instead of its former behavior of dumping the internal
representation of the plan tree. The latter display is still available for
those who really want it (see debug_print_plan), but uses for it are certainly
few and and far between. Per discussion.
This patch also removes the explain_pretty_print GUC, which is obsoleted
by the change.
of each plan node. For the moment this is debug support only and is
not enabled unless EXPLAIN_PRINT_TLISTS is defined at build time.
Later I'll see about the idea of letting EXPLAIN VERBOSE do it.
corrupted. (Neither is very important if SIGTERM is used to shut down the
whole database cluster together, but there's a problem if someone tries to
SIGTERM individual backends.) To do this, introduce new infrastructure
macros PG_ENSURE_ERROR_CLEANUP/PG_END_ENSURE_ERROR_CLEANUP that take care
of transiently pushing an on_shmem_exit cleanup hook. Also use this method
for createdb cleanup --- that wasn't a shared-memory-corruption problem,
but SIGTERM abort of createdb could leave orphaned files lying around.
Backpatch as far as 8.2. The shmem corruption cases don't exist in 8.1,
and the createdb usage doesn't seem important enough to risk backpatching
further.
"consistent" functions, and remove pg_amop.opreqcheck, as per recent
discussion. The main immediate benefit of this is that we no longer need
8.3's ugly hack of requiring @@@ rather than @@ to test weight-using tsquery
searches on GIN indexes. In future it should be possible to optimize some
other queries better than is done now, by detecting at runtime whether the
index match is exact or not.
Tom Lane, after an idea of Heikki's, and with some help from Teodor.
instead of plan time. Extend the amgettuple API so that the index AM returns
a boolean indicating whether the indexquals need to be rechecked, and make
that rechecking happen in nodeIndexscan.c (currently the only place where
it's expected to be needed; other callers of index_getnext are just erroring
out for now). For the moment, GIN and GIST have stub logic that just always
sets the recheck flag to TRUE --- I'm hoping to get Teodor to handle pushing
that control down to the opclass consistent() functions. The planner no
longer pays any attention to amopreqcheck, and that catalog column will go
away in due course.
classed all as "dead"; also get it to count DEAD item pointers as dead rows,
instead of ignoring them as before. Also improve matters so that tuples
previously inserted or deleted by our own transaction are handled nicely:
the stats collector's live-tuple and dead-tuple counts will end up correct
after our transaction ends, regardless of whether we end in commit or abort.
While there's more work that could be done to improve the counting of in-doubt
tuples in both VACUUM and ANALYZE, this commit is enough to alleviate some
known bad behaviors in 8.3; and the other stuff that's been discussed seems
like research projects anyway.
Pavan Deolasee and Tom Lane
responsible for copying the query string into the new Portal. Such copying
is unnecessary in the common code path through exec_simple_query, and in
this case it can be enormously expensive because the string might contain
a large number of individual commands; we were copying the entire, long
string for each command, resulting in O(N^2) behavior for N commands.
(This is the cause of bug #4079.) A second problem with it is that
PortalDefineQuery really can't risk error, because if it elog's before
having set up the Portal, we will leak the plancache refcount that the
caller is trying to hand off to the portal. So go back to the design in
which the caller is responsible for making sure everything is copied into
the portal if necessary.
while EState still contains pointers to those relations. Exposed by the
CLOBBER_CACHE_ALWAYS tests that buildfarm member jaguar is running (I knew
those cycles would pay off...)
inclusions in src/include/catalog/*.h files. The main idea here is to push
function declarations for src/backend/catalog/*.c files into separate headers,
rather than sticking them into the corresponding catalog definition file as
has been done in the past. This commit only carries out that idea fully for
pg_proc, pg_type and pg_conversion, but that's enough for the moment ---
if pg_list.h ever becomes unsafe for frontend code to include, we'll need
to work a bit more.
Zdenek Kotala
snapmgmt.c file for the former. The header files have also been reorganized
in three parts: the most basic snapshot definitions are now in a new file
snapshot.h, and the also new snapmgmt.h keeps the definitions for snapmgmt.c.
tqual.h has been reduced to the bare minimum.
This patch is just a first step towards managing live snapshots within a
transaction; there is no functionality change.
Per my proposal to pgsql-patches on 20080318191940.GB27458@alvh.no-ip.org and
subsequent discussion.
strings. This patch introduces four support functions cstring_to_text,
cstring_to_text_with_len, text_to_cstring, and text_to_cstring_buffer, and
two macros CStringGetTextDatum and TextDatumGetCString. A number of
existing macros that provided variants on these themes were removed.
Most of the places that need to make such conversions now require just one
function or macro call, in place of the multiple notational layers that used
to be needed. There are no longer any direct calls of textout or textin,
and we got most of the places that were using handmade conversions via
memcpy (there may be a few still lurking, though).
This commit doesn't make any serious effort to eliminate transient memory
leaks caused by detoasting toasted text objects before they reach
text_to_cstring. We changed PG_GETARG_TEXT_P to PG_GETARG_TEXT_PP in a few
places where it was easy, but much more could be done.
Brendan Jurd and Tom Lane
identical to tuplestore_puttuple(), except it operates on arrays of
Datums + nulls rather than a fully-formed HeapTuple. In several places
that use the tuplestore API, this means we can avoid creating a
HeapTuple altogether, saving a copy.
friends. Avoid double translation of some messages, ensure other messages
are exposed for translation (and make them follow the style guidelines),
avoid unsafe passing of an unpredictable message text as a format string.
pg_listener modifications commanded by LISTEN and UNLISTEN until the end
of the current transaction. This allows us to hold the ExclusiveLock on
pg_listener until after commit, with no greater risk of deadlock than there
was before. Aside from fixing the race condition, this gets rid of a
truly ugly kludge that was there before, namely having to ignore
HeapTupleBeingUpdated failures during NOTIFY. There is a small potential
incompatibility, which is that if a transaction issues LISTEN or UNLISTEN
and then looks into pg_listener before committing, it won't see any resulting
row insertion or deletion, where before it would have. It seems unlikely
that anyone would be depending on that, though.
This patch also disallows LISTEN and UNLISTEN inside a prepared transaction.
That case had some pretty undesirable properties already, such as possibly
allowing pg_listener entries to be made for PIDs no longer present, so
disallowing it seems like a better idea than trying to maintain the behavior.
calculating a page's initial free space was fine, and should not have been
"improved" by letting PageGetHeapFreeSpace do it. VACUUM FULL is going to
reclaim LP_DEAD line pointers later, so there is no need for a guard
against the page being too full of line pointers, and having one risks
rejecting pages that are perfectly good move destinations.
This also exposed a second bug, which is that the empty_end_pages logic
assumed that any page with no live tuples would get entered into the
fraged_pages list automatically (by virtue of having more free space than
the threshold in the do_frag calculation). This assumption certainly
seems risky when a low fillfactor has been chosen, and even without
tunable fillfactor I think it could conceivably fail on a page with many
unused line pointers. So fix the code to force do_frag true when notup
is true, and patch this part of the fix all the way back.
Per report from Tomas Szepe.
statement be a list of bare C strings, rather than String nodes, which is
what they need to be for copyfuncs/equalfuncs to work. Fortunately these
node types never go out to disk (if they did, we'd likely have noticed the
problem sooner), so we can just fix it without creating a need for initdb.
This bug has been there since 8.0, but 8.3 exposes it in a more common
code path (Parse messages) than prior releases did. Per bug #3940 from
Vladimir Kokovic.
tablespace permissions failures when copying an index that is in the
database's default tablespace. A side-effect of the change is that explicitly
specifying the default tablespace no longer triggers a permissions check;
this is not how it was done in pre-8.3 releases but is argued to be more
consistent. Per bug #3921 from Andrew Gilligan. (Note: I argued in the
subsequent discussion that maybe LIKE shouldn't copy index tablespaces
at all, but since no one indicated agreement with that idea, I've refrained
from doing it.)
operations when the current transaction has any open references to the
target relation or index (implying it has an active query using the relation).
The need for this was previously recognized in connection with ALTER TABLE,
but anything that summarily eliminates tuples or moves them around would
confuse an active scan.
While this patch does not in itself fix bug #3883 (the deadlock would happen
before the new check fires), it will discourage people from attempting the
sequence of operations that creates a deadlock risk, so it's at least a
partial response to that problem.
In passing, add a previously-missing check to REINDEX to prevent trying to
reindex another backend's temp table. This isn't a security problem since
only a superuser would get past the schema permission checks, but if we are
testing for this in other utility commands then surely REINDEX should too.
constraint, the constraint is renamed as well. This avoids inconsistent
situations that could confuse pg_dump (not to mention humans). We might at
some point provide ALTER TABLE RENAME CONSTRAINT as a more general solution,
but there seems no reason not to allow doing it this way too. Per bug #3854
and related discussions.
and CLUSTER) execute as the table owner rather than the calling user, using
the same privilege-switching mechanism already used for SECURITY DEFINER
functions. The purpose of this change is to ensure that user-defined
functions used in index definitions cannot acquire the privileges of a
superuser account that is performing routine maintenance. While a function
used in an index is supposed to be IMMUTABLE and thus not able to do anything
very interesting, there are several easy ways around that restriction; and
even if we could plug them all, there would remain a risk of reading sensitive
information and broadcasting it through a covert channel such as CPU usage.
To prevent bypassing this security measure, execution of SET SESSION
AUTHORIZATION and SET ROLE is now forbidden within a SECURITY DEFINER context.
Thanks to Itagaki Takahiro for reporting this vulnerability.
Security: CVE-2007-6600
in the current backend for the target table. These operations move tuples
around and would thus invalidate the TIDs stored in the trigger event records.
(We need not worry about events in other backends, since acquiring exclusive
lock should be enough to ensure there aren't any.) It might be sufficient
to forbid only the table-rewriting variants of ALTER TABLE, but in the absence
of any compelling use-case, let's just be safe and simple. Per follow-on
investigation of bug #3847, though this is not actually the same problem
reported therein.
Possibly this should be back-patched, but since the case has never been
reported from the field, I didn't bother.
were reporting ERROR for interactive assignments and LOG for other cases,
some were saying nothing for non-interactive cases, and a few did yet other
things. Make them use a new function GUC_complaint_elevel() to establish
a reasonably uniform policy about how to report. There are still a few
edge cases such as assign_search_path(), but it's much better than before.
Per gripe from Devrim Gunduz and subsequent discussion.
As noted by Alvaro, it'd be better to fold these custom messages into the
standard "invalid parameter value" complaint from guc.c, perhaps as the DETAIL
field. However that will require more redesign than seems prudent for 8.3.
This is a relatively safe, low-impact change that we can afford to risk now.
COPY. We need a restriction here because when the delimiter occurs as a
data character, it is emitted with a backslash, and that will only work
as desired if CopyReadAttributesText() will interpret the backslash sequence
as representing the second character literally. This is currently untrue
for 'b', 'f', 'n', 'r', 't', 'v', 'x', and octal digits. For future-proofing
and simplicity of explanation, it seems best to disallow a-z and 0-9.
We must also disallow dot, since "\." by itself would look like copy EOF.
Note: "\N" is by default the null print string, so N would also cause a
problem, but that is already tested for.
CopyAttributeOutText(), so that control characters are converted to the
C-style escape sequences even if they happen to be equal to the column
delimiter (as is true by default for tab, for example). Oversight in my
previous patch to restore pre-8.3 behavior of COPY OUT escaping. Per report
from Tomas Szepe.
namely that \r, \n, \t, \b, \f, \v are dumped as those two-character
representations rather than a backslash and the literal control character.
I had made it do the other to save some code, but this was ill-advised,
because dump files in which these characters appear literally are prone to
newline mangling. Fortunately, doing it the old way should only cost a few
more lines of code, and not slow down the copy loop materially.
Per bug #3795 from Lou Duchez.
constraint status of copied indexes (bug #3774), as well as various other
small bugs such as failure to pstrdup when needed. Allow INCLUDING INDEXES
indexes to be merged with identical declared indexes (perhaps not real useful,
but the code is there and having it not apply to LIKE indexes seems pretty
unorthogonal). Avoid useless work in generateClonedIndexStmt(). Undo some
poorly chosen API changes, and put a couple of routines in modules that seem
to be better places for them.
but no database changes have been made since the last CommandCounterIncrement.
This should result in a significant improvement in the number of "commands"
that can typically be performed within a transaction before hitting the 2^32
CommandId size limit. In particular this buys back (and more) the possible
adverse consequences of my previous patch to fix plan caching behavior.
The implementation requires tracking whether the current CommandCounter
value has been "used" to mark any tuples. CommandCounter values stored into
snapshots are presumed not to be used for this purpose. This requires some
small executor changes, since the executor used to conflate the curcid of
the snapshot it was using with the command ID to mark output tuples with.
Separating these concepts allows some small simplifications in executor APIs.
Something for the TODO list: look into having CommandCounterIncrement not do
AcceptInvalidationMessages. It seems fairly bogus to be doing it there,
but exactly where to do it instead isn't clear, and I'm disinclined to mess
with asynchronous behavior during late beta.
checkpoint. This guards against an unlikely data-loss scenario in which
we re-use the relfilenode, then crash, then replay the deletion and
recreation of the file. Even then we'd be OK if all insertions into the
new relation had been WAL-logged ... but that's not guaranteed given all
the no-WAL-logging optimizations that have recently been added.
Patch by Heikki Linnakangas, per a discussion last month.
even in code paths where we don't pay any subsequent attention to the typmod
value. This seems needed in view of the fact that 8.3's generalized typmod
support will accept a lot of bogus syntax, such as "timestamp(foo)" or
"record(int, 42)" --- if we allow such things to pass without comment,
users will get confused. Per a recent example from Greg Stark.
To implement this in a way that's not very vulnerable to future
bugs-of-omission, refactor the API of parse_type.c's TypeName lookup routines
so that typmod validation is folded into the base lookup operation. Callers
can still choose not to receive the encoded typmod, but we'll check the
decoration anyway if it's present.
foreign keys, one more time. Insist on matching up all three triggers before
we create a constraint; this will avoid creation of duplicate constraints
in scenarios where a broken FK constraint was repaired by re-adding the
constraint without removing the old partial trigger set. Basically, this will
work nicely in all cases where the FK was actually functioning correctly in
the database that was dumped. It will fail to restore an FK in just one case
where we theoretically could restore it: where we find the referenced table's
triggers and not the referencing table's trigger. However, in such a scenario
it's likely that the user doesn't even realize he still has an FK at all
(since the more-likely-to-fail cases aren't enforced), and we'd probably not
accomplish much except to cause the reload to fail because the data doesn't
meet the FK constraint. Also make the NOTICE logging still more verbose, by
adding detail about which of the triggers were found. This seems about all
we can do without solving the problem of getting the user's attention at
session end.
commands into proper foreign-key constraints. Believe the constraint name
given in the trigger arguments in preference to the trigger name --- to judge
from Olivier Prenant's example, pg_dump must at some time have used the
autogenerated trigger name there, though AFAICT no current release branch tip
does. Improve the emitted NOTICEs to provide more detail (PK table's name and
column names). Handle the case where pg_dump forgot to provide the FROM table
(a bug that never did get fixed in 7.0.x apparently). This commit doesn't
do anything about the question of what to do with incomplete trigger groups.
trigger definitions into regular foreign key constraints. This seems
necessary given that some people evidently never did get around to
running adddepend on their schemas, and without some sort of hack the
old definitions will no longer work. Per report from Olivier Prenant
and subsequent investigation.
Instead put in a test to drop a NULL default at the last moment before
storing the catalog entry. This changes the behavior in a couple of ways:
* Specifying DEFAULT NULL when creating an inheritance child table will
successfully suppress inheritance of any default expression from the
parent's column, where formerly it failed to do so.
* Specifying DEFAULT NULL for a column of a domain type will correctly
override any default belonging to the domain; likewise for a sub-domain.
The latter change happens because by the time the clause is checked,
it won't be a simple null Const but a CoerceToDomain expression.
Personally I think this should be back-patched, but there doesn't seem to
be consensus for that on pgsql-hackers, so refraining.
the sequence. Also, make setval() with is_called = false not affect the
currval state, either. Per report from Kris Jurka that an implicit
ALTER SEQUENCE OWNED BY unexpectedly caused currval() to become valid.
Since this isn't 100% backwards compatible, it will go into HEAD only;
I'll put a more limited patch into 8.2.
in corner cases such as re-fetching a just-deleted row. We may be able to
relax this someday, but let's find out how many people really care before
we invest a lot of work in it. Per report from Heikki and subsequent
discussion.
While in the neighborhood, make the combination of INSENSITIVE and FOR UPDATE
throw an error, since they are semantically incompatible. (Up to now we've
accepted but just ignored the INSENSITIVE option of DECLARE CURSOR.)
having several of them. Add two more flags: whether the process is
executing an ANALYZE, and whether a vacuum is for Xid wraparound (which
is obviously only set by autovacuum).
Sneakily move the worker's recently-acquired PostAuthDelay to a more useful
place.
renumbering of encoding IDs done between 8.2 and 8.3 turns out to break 8.2
initdb and psql if they are run with an 8.3beta1 libpq.so. For the moment
we can rearrange the order of enum pg_enc to keep the same number for
everything except PG_JOHAB, which isn't a problem since there are no direct
references to it in the 8.2 programs anyway. (This does force initdb
unfortunately.)
Going forward, we want to fix things so that encoding IDs can be changed
without an ABI break, and this commit includes the changes needed to allow
libpq's encoding IDs to be treated as fully independent of the backend's.
The main issue is that libpq clients should not include pg_wchar.h or
otherwise assume they know the specific values of libpq's encoding IDs,
since they might encounter version skew between pg_wchar.h and the libpq.so
they are using. To fix, have libpq officially export functions needed for
encoding name<=>ID conversion and validity checking; it was doing this
anyway unofficially.
It's still the case that we can't renumber backend encoding IDs until the
next bump in libpq's major version number, since doing so will break the
8.2-era client programs. However the code is now prepared to avoid this
type of problem in future.
Note that initdb is no longer a libpq client: we just pull in the two
source files we need directly. The patch also fixes a few places that
were being sloppy about checking for an unrecognized encoding name.
on pg_global even to superusers, and replace it with checks in various
other places to complain about invalid uses of pg_global. This ends
up being a bit more code but it allows a more specific error message
to be given, and it un-breaks pg_tablespace_size() on pg_global.
Per discussion.
CREATE INDEX CONCURRENTLY). Such an index might not have entries for every
heap row and thus clustering with it would result in silent data loss.
The scenario requires a pretty foolish DBA, but still ...
ALTER TABLE on a composite type or ALTER TYPE on a table's rowtype.
We already rejected these cases, but the error messages were a bit
random and didn't always provide a HINT to use the other command type.
databases with encodings that are incompatible with the server's LC_CTYPE
locale, when we can determine that (which we can on most modern platforms,
I believe). C/POSIX locale is compatible with all encodings, of course,
so there is still some usefulness to CREATE DATABASE's ENCODING option,
but this will insulate us against all sorts of recurring complaints
caused by mismatched settings.
I moved initdb's existing LC_CTYPE-to-encoding mapping knowledge into
a new src/port/ file so it could be shared by CREATE DATABASE.
table, by allocating just enough for a hardcoded number of dead tuples per
page. The current estimate is 200 dead tuples per page.
Per reports from Jeff Amiel, Erik Jones and Marko Kreen, and subsequent
discussion.
CVS: ----------------------------------------------------------------------
CVS: Enter Log. Lines beginning with `CVS:' are removed automatically
CVS:
CVS: Committing in .
CVS:
CVS: Modified Files:
CVS: commands/vacuumlazy.c
CVS: ----------------------------------------------------------------------
* stats_start_collector goes away; we always start the collector process,
unless prevented by a problem with setting up the stats UDP socket.
* stats_reset_on_server_start goes away; it seems useless in view of the
availability of pg_stat_reset().
* stats_block_level and stats_row_level are merged into a single variable
"track_counts", which controls all reports sent to the collector process.
* stats_command_string is renamed to track_activities.
* log_autovacuum is renamed to log_autovacuum_min_duration to better reflect
its meaning.
The log_autovacuum change is not a compatibility issue since it didn't exist
before 8.3 anyway. The other changes need to be release-noted.
columns, and the new version can be stored on the same heap page, we no longer
generate extra index entries for the new version. Instead, index searches
follow the HOT-chain links to ensure they find the correct tuple version.
In addition, this patch introduces the ability to "prune" dead tuples on a
per-page basis, without having to do a complete VACUUM pass to recover space.
VACUUM is still needed to clean up dead index entries, however.
Pavan Deolasee, with help from a bunch of other people.
no-longer-needed pages at the end of a table. We thought we could throw away
pages containing HEAPTUPLE_DEAD tuples; but this is not so, because such
tuples very likely have index entries pointing at them, and we wouldn't have
removed the index entries. The problem only emerges in a somewhat unlikely
race condition: the dead tuples have to have been inserted by a transaction
that later aborted, and this has to have happened between VACUUM's initial
scan of the page and then rechecking it for empty in count_nondeletable_pages.
But that timespan will include an index-cleaning pass, so it's not all that
hard to hit. This seems to explain a couple of previously unsolved bug
reports.
than two independent bits (one of which was never used in heap pages anyway,
or at least hadn't been in a very long time). This gives us flexibility to
add the HOT notions of redirected and dead item pointers without requiring
anything so klugy as magic values of lp_off and lp_len. The state values
are chosen so that for the states currently in use (pre-HOT) there is no
change in the physical representation.
an exclusive lock on the table at this point, which we want to release as soon
as possible. This is called in the phase of lazy vacuum where we truncate the
empty pages at the end of the table.
An alternative solution would be to lower the vacuum delay settings before
starting the truncating phase, but this doesn't work very well in autovacuum
due to the autobalancing code (which can cause other processes to change our
cost delay settings). This case could be considered in the balancing code, but
it is simpler this way.
ReadNewTransactionId from GetSnapshotData --- with a "latestCompletedXid"
variable that is updated during transaction commit or abort. Since
latestCompletedXid is written only in places that had to lock ProcArrayLock
exclusively anyway, and is read only in places that had to lock ProcArrayLock
shared anyway, it adds no new locking requirements to the system despite being
cluster-wide. Moreover, removing ReadNewTransactionId from snapshot
acquisition eliminates the need to take both XidGenLock and ProcArrayLock at
the same time. Since XidGenLock is sometimes held across I/O this can be a
significant win. Some preliminary benchmarking suggested that this patch has
no effect on average throughput but can significantly improve the worst-case
transaction times seen in pgbench. Concept by Florian Pflug, implementation
by Tom Lane.
no need for serialization against snapshot-taking because the xact doesn't
affect anyone else's snapshot anyway. Per discussion. Also, move various
info about the interlocking of transactions and snapshots out of code comments
and into a hopefully-more-cohesive discussion in access/transam/README.
Also, remove a couple of now-obsolete comments about having to force some WAL
to be written to persuade RecordTransactionCommit to do its thing.
null::char(3) to a simple Const node. (It already worked for non-null values,
but not when we skipped evaluation of a strict coercion function.) This
prevents loss of typmod knowledge in situations such as exhibited in bug
#3598. Unfortunately there seems no good way to fix that bug in 8.1 and 8.2,
because they simply don't carry a typmod for a plain Const node.
In passing I made all the other callers of makeNullConst supply "real" typmod
values too, though I think it probably doesn't matter anywhere else.
rows will normally never obtain an XID at all. We already did things this way
for subtransactions, but this patch extends the concept to top-level
transactions. In applications where there are lots of short read-only
transactions, this should improve performance noticeably; not so much from
removal of the actual XID-assignments, as from reduction of overhead that's
driven by the rate of XID consumption. We add a concept of a "virtual
transaction ID" so that active transactions can be uniquely identified even
if they don't have a regular XID. This is a much lighter-weight concept:
uniqueness of VXIDs is only guaranteed over the short term, and no on-disk
record is made about them.
Florian Pflug, with some editorialization by Tom.
(Actually, it works as a plain statement too, but I didn't document that
because it seems a bit useless.) Unify VariableResetStmt with
VariableSetStmt, and clean up some ancient cruft in the representation of
same.
There are still some loose ends: I didn't do anything about the SET FROM
CURRENT idea yet, and it's not real clear whether we are happy with the
interaction of SET LOCAL with function-local settings. The documentation
is a bit spartan, too.
namespace isn't necessarily first in the search path (there could be implicit
schemas ahead of it). Examples are
test=# set search_path TO s1;
test=# create view pg_timezone_names as select * from pg_timezone_names();
ERROR: "pg_timezone_names" is already a view
test=# create table pg_class (f1 int primary key);
ERROR: permission denied: "pg_class" is a system catalog
You'd expect these commands to create the requested objects in s1, since
names beginning with pg_ aren't supposed to be reserved anymore. What is
happening is that we create the requested base table and then execute
additional commands (here, CREATE RULE or CREATE INDEX), and that code is
passed the same RangeVar that was in the original command. Since that
RangeVar has schemaname = NULL, the secondary commands think they should do a
path search, and that means they find system catalogs that are implicitly in
front of s1 in the search path.
This is perilously close to being a security hole: if the secondary command
failed to apply a permission check then it'd be possible for unprivileged
users to make schema modifications to system catalogs. But as far as I can
find, there is no code path in which a check doesn't occur. Which makes it
just a weird corner-case bug for people who are silly enough to want to
name their tables the same as a system catalog.
The relevant code has changed quite a bit since 8.2, which means this patch
wouldn't work as-is in the back branches. Since it's a corner case no one
has reported from the field, I'm not going to bother trying to back-patch.
relcache entry after having heap_close'd it. This could lead to misbehavior
if a relcache flush wiped out the cache entry meanwhile. In 8.2 there is a
very real risk of CREATE INDEX CONCURRENTLY using the wrong relid for locking
and waiting purposes. I think the bug is only cosmetic in 8.0 and 8.1,
because their transgression is limited to using RelationGetRelationName(rel)
in an ereport message immediately after heap_close, and there's no way (except
with special debugging options) for a cache flush to occur in that interval.
Not quite sure that it's cosmetic in 7.4, but seems best to patch anyway.
Found by trying to run the regression tests with CLOBBER_CACHE_ALWAYS enabled.
Maybe we should try to do that on a regular basis --- it's awfully slow,
but perhaps some fast buildfarm machine could do it once in awhile.
This prevents needing to do complex and poorly-defined updates of the
mapping table if the new parser has different token types than the old.
Per discussion.
init options of the template as top-level options in the syntax. This also
makes ALTER a bit easier to use, since options can be replaced individually.
I also made these statements verify that the tmplinit method will accept
the new settings before they get stored; in the original coding you didn't
find out about mistakes until the dictionary got invoked.
Under the hood, init methods now get options as a List of DefElem instead
of a raw text string --- that lets tsearch use existing options-pushing code
instead of duplicating functionality.
'with map' parameter; as things now stand there's really not much point
in specifying a config-to-copy if you don't copy its map. Also, use
COPY instead of TEMPLATE as the key word for a config-to-copy, so as
to avoid confusion with text search templates. Per discussion; the
just-committed reference page for the command already describes it
this way.
Oleg Bartunov and Teodor Sigaev, but I did a lot of editorializing,
so anything that's broken is probably my fault.
Documentation is nonexistent as yet, but let's land the patch so we can
get some portability testing done.
are not one of the query's defined result relations, but nonetheless have
triggers fired against them while the query is active. This was formerly
impossible but can now occur because of my recent patch to fix the firing
order for RI triggers. Caching a ResultRelInfo avoids duplicating work by
repeatedly opening and closing the same relation, and also allows EXPLAIN
ANALYZE to "see" and report on these extra triggers. Use the same mechanism
to cache open relations when firing deferred triggers at transaction shutdown;
this replaces the former one-element-cache strategy used in that case, and
should improve performance a bit when there are deferred triggers on a number
of relations.
row within one query: we were firing check triggers before all the updates
were done, leading to bogus failures. Fix by making the triggers queued by
an RI update go at the end of the outer query's trigger event list, thereby
effectively making the processing "breadth-first". This was indeed how it
worked pre-8.0, so the bug does not occur in the 7.x branches.
Per report from Pavel Stehule.
First, we cannot assume that XLogAsyncCommitFlush guarantees hint bits will be
settable, because clog.c's inexact LSN bookkeeping results in windows where a
previously flushed transaction is considered unhintable because it shares an
LSN slot with a later unflushed transaction. But repair_frag requires
XMIN_COMMITTED to be correct so that it can distinguish tuples moved by the
current vacuum. Since not being able to set the bit is an uncommon corner
case, the most practical way of dealing with it seems to be to abandon
shrinking (ie, don't invoke repair_frag) when we find a non-dead tuple whose
XMIN_COMMITTED bit couldn't be set.
Second, it is possible for the same reason that a RECENTLY_DEAD tuple does not
get its XMAX_COMMITTED bit set during scan_heap. But by the time repair_frag
examines the tuple it might be possible to set the bit. We therefore must
take buffer content lock when calling HeapTupleSatisfiesVacuum a second time,
else we can get an Assert failure in SetBufferCommitInfoNeedsSave. This
latter bug is latent in existing releases, but I think it cannot actually
occur without async commit, since the first HeapTupleSatisfiesVacuum call
should always have set the bit. So I'm not going to back-patch it.
In passing, reduce the existing "cannot shrink relation" messages from NOTICE
to LOG level. The new message must be no higher than LOG if we don't want
unpredictable regression test failures, and consistency seems like a good
idea. Also arrange that only one such message is reported per VACUUM FULL;
in typical scenarios you could get spammed with many such messages, which
seems a bit useless.
displayed in the postmaster log. This avoids Windows-specific problems with
localized time zone names that are in the wrong encoding, and generally seems
like a good idea to forestall other potential platform-dependent issues.
To preserve the existing behavior that all backends will log in the same time
zone, create a new GUC variable log_timezone that can only be changed on a
system-wide basis, and reference log-related calculations to that zone instead
of the TimeZone variable.
This fixes the issue reported by Hiroshi Saito that timestamps printed by
xlog.c startup could be improperly localized on Windows. We still need a
simpler patch for that problem in the back branches, however.
before reporting a transaction committed. Data consistency is still
guaranteed (unlike setting fsync = off), but a crash may lose the effects
of the last few transactions. Patch by Simon, some editorialization by Tom.
referencing table does not change the tuple's FK column(s), we don't bother
to check the PK table since the constraint was presumably already valid.
However, the check is still necessary if the tuple was inserted by our own
transaction, since in that case the INSERT trigger will conclude it need not
make the check (since its version of the tuple has been deleted). We got this
right for simple cases, but not when the insert and update are in different
subtransactions of the current top-level transaction; in such cases the FK
check would never be made at all. (Hence, problem dates back to 8.0 when
subtransactions were added --- it's actually the subtransaction version of a
bug fixed in 7.3.5.) Fix, and add regression test cases. Report and fix by
Affan Salman.
Sequences and views could previously be renamed using ALTER TABLE, but
this was a repeated source of confusion for users. Update the docs,
and psql tab completion. Patch from David Fetter; various minor fixes
by myself.
that are fired at end-of-statement (as is the normal case for foreign keys,
for example). In this situation the per-subxact deferred trigger context
is always empty when subtransaction exit is reached; so we could free it,
but were not doing so, leading to an intratransaction leak of 8K or more
per subtransaction. Per off-list example from Viatcheslav Kalinin
subsequent to bug #3418 (his original bug report omitted a foreign key
constraint needed to cause this leak).
Back-patch to 8.2; prior versions were not using per-subxact contexts
for deferred triggers, so did not have this leak.
over a fairly long period of time, rather than being spat out in a burst.
This happens only for background checkpoints carried out by the bgwriter;
other cases, such as a shutdown checkpoint, are still done at full speed.
Remove the "all buffers" scan in the bgwriter, and associated stats
infrastructure, since this seems no longer very useful when the checkpoint
itself is properly throttled.
Original patch by Itagaki Takahiro, reworked by Heikki Linnakangas,
and some minor API editorialization by me.
(which now deals only in optimizable statements), and put that code
into a new file parser/parse_utilcmd.c. This helps clarify and enforce
the design rule that utility statements shouldn't be processed during
the regular parse analysis phase; all interpretation of their meaning
should happen after they are given to ProcessUtility to execute.
(We need this because we don't retain any locks for a utility statement
that's in a plan cache, nor have any way to detect that it's stale.)
We are also able to simplify the API for parse_analyze() and related
routines, because they will now always return exactly one Query structure.
In passing, fix bug #3403 concerning trying to add a serial column to
an existing temp table (this is largely Heikki's work, but we needed
all that restructuring to make it safe).
profiling that CopyAttributeOutText was taking an unreasonable fraction of
the backend run time (like 66%!) on the following trivial test case:
$ time psql -c "copy (select repeat('xyzzy',50) from generate_series(1,10000000)) to stdout" regression >/dev/null
The time is all being spent on scanning the string for characters to be
escaped, which most of the time there aren't any of. Some tweaking to take
as many tests as possible out of the inner loop reduced the runtime of this
example by more than 10%. In a real-world case it wouldn't be as useful
a speedup, but it still seems worth adding a few lines here.
an array of strings rather than an array of integers, and allow any simple
constant or identifier to be used in typmods; for example
create table foo (f1 widget(42,'23skidoo',point));
Of course the typmodin function has still got to pack this info into a
non-negative int32 for storage, but it's still a useful improvement in
flexibility, especially considering that you can do nearly anything if you
are willing to keep the info in a side table. We can get away with this
change since we have not yet released a version providing user-definable
typmods. Per discussion.
for each temp file, rather than once per sort or hashjoin; this allows
spreading the data of a large sort or join across multiple tablespaces.
(I remain dubious that this will make any difference in practice, but certain
people insisted.) Arrange to cache the results of parsing the GUC variable
instead of recomputing from scratch on every demand, and push usage of the
cache down to the bottommost fd.c level.
tablespace(s) in which to store temp tables and temporary files. This is a
list to allow spreading the load across multiple tablespaces (a random list
element is chosen each time a temp object is to be created). Temp files are
not stored in per-database pgsql_tmp/ directories anymore, but per-tablespace
directories.
Jaime Casanova and Albert Cervera, with review by Bernd Helmle and Tom Lane.
will exit before failing because of conflicting DB usage. Per discussion,
this seems a good idea to help mask the fact that backend exit takes nonzero
time. Remove a couple of thereby-obsoleted sleeps in contrib and PL
regression test sequences.
buffers, rather than blowing out the whole shared-buffer arena. Aside from
avoiding cache spoliation, this fixes the problem that VACUUM formerly tended
to cause a WAL flush for every page it modified, because we had it hacked to
use only a single buffer. Those flushes will now occur only once per
ring-ful. The exact ring size, and the threshold for seqscans to switch into
the ring usage pattern, remain under debate; but the infrastructure seems
done. The key bit of infrastructure is a new optional BufferAccessStrategy
object that can be passed to ReadBuffer operations; this replaces the former
StrategyHintVacuum API.
This patch also changes the buffer usage-count methodology a bit: we now
advance usage_count when first pinning a buffer, rather than when last
unpinning it. To preserve the behavior that a buffer's lifetime starts to
decrease when it's released, the clock sweep code is modified to not decrement
usage_count of pinned buffers.
Work not done in this commit: teach GiST and GIN indexes to use the vacuum
BufferAccessStrategy for vacuum-driven fetches.
Original patch by Simon, reworked by Heikki and again by Tom.
and/or create plans for hypothetical situations; in particular, investigate
plans that would be generated using hypothetical indexes. This is a
heavily-rewritten version of the hooks proposed by Gurjeet Singh for his
Index Advisor project. In this formulation, the index advisor can be
entirely a loadable module instead of requiring a significant part to be
in the core backend, and plans can be generated for hypothetical indexes
without requiring the creation and rolling-back of system catalog entries.
The index advisor patch as-submitted is not compatible with these hooks,
but it needs significant work anyway due to other 8.2-to-8.3 planner
changes. With these hooks in the core backend, development of the advisor
can proceed as a pgfoundry project.
FreezeXid introduced in a recent commit, so there isn't any data loss in this
approach.
Doing it causes ALTER TABLE (or rather, the forms of it that cause a full table
rewrite) to be affected as well. In this case, the frozen point is RecentXmin,
because after the rewrite all the tuples are relabeled with the rewriting
transaction's Xid.
TOAST tables are fixed automatically as well, as fallout of the way they were
already being handled in the respective code paths.
With this patch, there is no longer need to VACUUM tables for Xid wraparound
purposes that have been cleaned up via TRUNCATE or CLUSTER.
to avoid losing useful Xid information in not-so-old tuples. This makes
CLUSTER behave the same as VACUUM as far a tuple-freezing behavior goes
(though CLUSTER does not yet advance the table's relfrozenxid).
While at it, move the actual freezing operation in rewriteheap.c to a more
appropriate place, and document it thoroughly. This part of the patch from
Tom Lane.
avoid a later needless VACUUM for Xid-wraparound purposes. We can do this
since the table is known to be left empty, so no Xid remains on it.
Per discussion.
there's an indirect dependency on the owner via the parent table. We were
already handling indexes that way, but not toast tables for some reason.
Saves a little catalog space and cuts down the verbosity of checkSharedDependencies
reports.
named foo, would work but the other ordering would not. If a user-specified
type or table name collides with an existing auto-generated array name, just
rename the array type out of the way by prepending more underscores. This
should not create any backward-compatibility issues, since the cases in which
this will happen would have failed outright in prior releases.
Also fix an oversight in the arrays-of-composites patch: ALTER TABLE RENAME
renamed the table's rowtype but not its array type.
needs to check the new constraint against columns of derived domains too.
Also, make it error out if the domain to be modified is used within any
composite-type columns. Eventually we should support that case, but it seems
a bit painful, and not suitable for a back-patch. For the moment just let the
user know we can't do it.
Backpatch to 8.2, which is the only released version that allows nested
domains. Possibly the other part should be back-patched further.
and views (but not system catalogs, nor sequences or toast tables). Get rid
of the hardwired convention that a type's array type is named exactly "_type",
instead using a new column pg_type.typarray to provide the linkage. (It still
will be named "_type", though, except in odd corner cases such as
maximum-length type names.)
Along the way, make tracking of owner and schema dependencies for types more
uniform: a type directly created by the user has these dependencies, while a
table rowtype or auto-generated array type does not have them, but depends on
its parent object instead.
David Fetter, Andrew Dunstan, Tom Lane
true at the very end of its processing, the update is broadcast via a
shared-cache-inval message for the index; without this, existing backends that
already have relcache entries for the index might never see it become valid.
Also, force a relcache inval on the index's parent table at the same time,
so that any cached plans for that table are re-planned; this ensures that
the newly valid index will be used if appropriate. Aside from making
C.I.C. behave more reasonably, this is necessary infrastructure for some
aspects of the HOT patch. Pavan Deolasee, with a little further stuff from
me.
messages to the stats collector. This avoids the problem that enabling
stats_row_level for autovacuum has a significant overhead for short
read-only transactions, as noted by Arjen van der Meijden. We can avoid
an extra gettimeofday call by piggybacking on the one done for WAL-logging
xact commit or abort (although that doesn't help read-only transactions,
since they don't WAL-log anything).
In my proposal for this, I noted that we could change the WAL log entries
for commit/abort to record full TimestampTz precision, instead of only
time_t as at present. That's not done in this patch, but will be committed
separately.
types of unspecified parameters when submitted via extended query protocol.
This worked in 8.2 but I had broken it during plancache changes. DECLARE
CURSOR is now treated almost exactly like a plain SELECT through parse
analysis, rewrite, and planning; only just before sending to the executor
do we divert it away to ProcessUtility. This requires a special-case check
in a number of places, but practically all of them were already special-casing
SELECT INTO, so it's not too ugly. (Maybe it would be a good idea to merge
the two by treating IntoClause as a form of utility statement? Not going to
worry about that now, though.) That approach doesn't work for EXPLAIN,
however, so for that I punted and used a klugy solution of running parse
analysis an extra time if under extended query protocol.
is in progress on the same hashtable. This seems the least invasive way to
fix the recently-recognized problem that a split could cause the scan to
visit entries twice or (with much lower probability) miss them entirely.
The only field-reported problem caused by this is the "failed to re-find
shared lock object" PANIC in COMMIT PREPARED reported by Michel Dorochevsky,
which was caused by multiply visited entries. However, it seems certain
that mdsync() is vulnerable to missing required fsync's due to missed
entries, and I am fearful that RelationCacheInitializePhase2() might be at
risk as well. Because of that and the generalized hazard presented by this
bug, back-patch all the supported branches.
Along the way, fix pg_prepared_statement() and pg_cursor() to not assume
that the hashtables they are examining will stay static between calls.
This is risky regardless of the newly noted dynahash problem, because
hash_seq_search() has never promised to cope with deletion of table entries
other than the just-returned one. There may be no bug here because the only
supported way to call these functions is via ExecMakeTableFunctionResult()
which will cycle them to completion before doing anything very interesting,
but it seems best to get rid of the assumption. This affects 8.2 and HEAD
only, since those functions weren't there earlier.
RESET SESSION, RESET PLANS, and RESET TEMP are now DISCARD ALL,
DISCARD PLANS, and DISCARD TEMP, respectively. This is to avoid
confusion with the pre-existing RESET variants: the DISCARD
commands are not actually similar to RESET. Patch from Marko
Kreen, with some minor editorialization.
processes to be running simultaneously. Also, now autovacuum processes do not
count towards the max_connections limit; they are counted separately from
regular processes, and are limited by the new GUC variable
autovacuum_max_workers.
The launcher now has intelligence to launch workers on each database every
autovacuum_naptime seconds, limited only on the max amount of worker slots
available.
Also, the global worker I/O utilization is limited by the vacuum cost-based
delay feature. Workers are "balanced" so that the total I/O consumption does
not exceed the established limit. This part of the patch was contributed by
ITAGAKI Takahiro.
Per discussion.
a replan. I had originally thought this was not necessary, but the new
SPI facilities create a path whereby queries planned with non-default
options can get into the cache, so it is necessary.
access to the planner's cursor-related planning options, and provide new
FETCH/MOVE routines that allow access to the full power of those commands.
Small refactoring of planner(), pg_plan_query(), and pg_plan_queries()
APIs to make it convenient to pass the planning options down from SPI.
This is the core-code portion of Pavel Stehule's patch for scrollable
cursor support in plpgsql; I'll review and apply the plpgsql changes
separately.
reviewed by Neil Conway. This patch adds the following DDL command
variants: RESET SESSION, RESET TEMP, RESET PLANS, CLOSE ALL, and
DEALLOCATE ALL. RESET SESSION is intended for use by connection
pool software and the like, in order to reset a client session
to something close to its initial state.
Note that while most of these command variants can be executed
inside a transaction block (but are not transaction-aware!),
RESET SESSION cannot. While this is inconsistent, it is intended
to catch programmer mistakes: RESET SESSION in an open transaction
block is probably unintended.
This commit breaks any code that assumes that the mere act of forming a tuple
(without writing it to disk) does not "toast" any fields. While all available
regression tests pass, I'm not totally sure that we've fixed every nook and
cranny, especially in contrib.
Greg Stark with some help from Tom Lane
A DBA is allowed to create a language in his database if it's marked
"tmpldbacreate" in pg_pltemplate. The factory default is that this is set
for all standard trusted languages, but of course a superuser may adjust
the settings. In service of this, add the long-foreseen owner column to
pg_language; renaming, dropping, and altering owner of a PL now follow
normal ownership rules instead of being superuser-only.
Jeremy Drake, with some editorialization by Tom Lane.
search_path that was active when the plan was first made. To do this,
improve namespace.c to support a stack of "override" search path settings
(we must have a stack since nested replan events are entirely possible).
This facility replaces the "special namespace" hack formerly used by
CREATE SCHEMA, and should be able to support per-function search path
settings as well.
doesn't exist. This allows DROP to be used to clean out the pg_tablespace
catalog entry in a situation where a previous DROP attempt failed before
committing but after having removed the directories and symlink.
Per report from William Garrison. Even though his test case depends on an
unrelated bug in PreventTransactionChain, it's certainly possible for this
situation to arise due to other problems, eg a system crash at just the
right time.
rules to be defined with different, per session controllable, behaviors
for replication purposes.
This will allow replication systems like Slony-I and, as has been stated
on pgsql-hackers, other products to control the firing mechanism of
triggers and rewrite rules without modifying the system catalog directly.
The firing mechanisms are controlled by a new superuser-only GUC
variable, session_replication_role, together with a change to
pg_trigger.tgenabled and a new column pg_rewrite.ev_enabled. Both
columns are a single char data type now (tgenabled was a bool before).
The possible values in these attributes are:
'O' - Trigger/Rule fires when session_replication_role is "origin"
(default) or "local". This is the default behavior.
'D' - Trigger/Rule is disabled and fires never
'A' - Trigger/Rule fires always regardless of the setting of
session_replication_role
'R' - Trigger/Rule fires when session_replication_role is "replica"
The GUC variable can only be changed as long as the system does not have
any cached query plans. This will prevent changing the session role and
accidentally executing stored procedures or functions that have plans
cached that expand to the wrong query set due to differences in the rule
firing semantics.
The SQL syntax for changing a triggers/rules firing semantics is
ALTER TABLE <tabname> <when> TRIGGER|RULE <name>;
<when> ::= ENABLE | ENABLE ALWAYS | ENABLE REPLICA | DISABLE
psql's \d command as well as pg_dump are extended in a backward
compatible fashion.
Jan
did not expect that a DEAD tuple could follow a RECENTLY_DEAD tuple in an
update chain, but because the OldestXmin rule for determining deadness is a
simplification of reality, it is possible for this situation to occur
(implying that the RECENTLY_DEAD tuple is in fact dead to all observers,
but this patch does not attempt to exploit that). The code would follow a
chain forward all the way, but then stop before a DEAD tuple when backing
up, meaning that not all of the chain got moved. This could lead to copying
the chain multiple times (resulting in duplicate copies of the live tuple at
its end), or leaving dangling index entries behind (which, aside from
generating warnings from later vacuums, creates a risk of wrong query
results or bogus duplicate-key errors once the heap slot the index entry
points to is repopulated).
The fix is to recheck HeapTupleSatisfiesVacuum while following a chain
forward, and to stop if a DEAD tuple is reached. Each contiguous group
of RECENTLY_DEAD tuples will therefore be copied as a separate chain.
The patch also adds a couple of extra sanity checks to verify correct
behavior.
Per report and test case from Pavan Deolasee.
module and teach PREPARE and protocol-level prepared statements to use it.
In service of this, rearrange utility-statement processing so that parse
analysis does not assume table schemas can't change before execution for
utility statements (necessary because we don't attempt to re-acquire locks
for utility statements when reusing a stored plan). This requires some
refactoring of the ProcessUtility API, but it ends up cleaner anyway,
for instance we can get rid of the QueryContext global.
Still to do: fix up SPI and related code to use the plan cache; I'm tempted to
try to make SQL functions use it too. Also, there are at least some aspects
of system state that we want to ensure remain the same during a replan as in
the original processing; search_path certainly ought to behave that way for
instance, and perhaps there are others.
even if none of the fields in the pg_class row change. This behavior is
necessary to ensure other backends flush rd_targblock values that might
point to truncated-away pages. We got this right pre-8.2 but it was broken
by overoptimistic change to not write out the pg_class row if unchanged.
Per report from Pavan Deolasee.
fixup various places in the tree that were clearing a StringInfo by hand.
Making this function a part of the API simplifies client code slightly,
and avoids needlessly peeking inside the StringInfo interface.
drill down into subplan targetlists to print the referent expression for an
OUTER or INNER var in an upper plan node. Hence, make it do that always, and
banish the old hack of showing "?columnN?" when things got too complicated.
Along the way, fix an EXPLAIN bug I introduced by suppressing subqueries from
execution-time range tables: get_name_for_var_field() assumed it could look at
rte->subquery to find out the real type of a RECORD var. That doesn't work
anymore, but instead we can look at the input plan of the SubqueryScan plan
node.
and quals have varno OUTER, rather than zero, to indicate a reference to
an output of their lefttree subplan. This is consistent with the way
that every other upper-level node type does it, and allows some simplifications
in setrefs.c and EXPLAIN.
useless substructure for its RangeTblEntry nodes. (I chose to keep using the
same struct node type and just zero out the link fields for unneeded info,
rather than making a separate ExecRangeTblEntry type --- it seemed too
fragile to have two different rangetable representations.)
Along the way, put subplans into a list in the toplevel PlannedStmt node,
and have SubPlan nodes refer to them by list index instead of direct pointers.
Vadim wanted to do that years ago, but I never understood what he was on about
until now. It makes things a *whole* lot more robust, because we can stop
worrying about duplicate processing of subplans during expression tree
traversals. That's been a constant source of bugs, and it's finally gone.
There are some consequent simplifications yet to be made, like not using
a separate EState for subplans in the executor, but I'll tackle that later.
storing mostly-redundant Query trees in prepared statements, portals, etc.
To replace Query, a new node type called PlannedStmt is inserted by the
planner at the top of a completed plan tree; this carries just the fields of
Query that are still needed at runtime. The statement lists kept in portals
etc. now consist of intermixed PlannedStmt and bare utility-statement nodes
--- no Query. This incidentally allows us to remove some fields from Query
and Plan nodes that shouldn't have been there in the first place.
Still to do: simplify the execution-time range table; at the moment the
range table passed to the executor still contains Query trees for subqueries.
initdb forced due to change of stored rules.
plan nodes, so that the executor does not need to get these items from
the range table at runtime. This will avoid needing to include these
fields in the compact range table I'm expecting to make the executor use.
an opclass for a generic type such as ANYARRAY. The original coding failed
to check that PK and FK columns were of the same array type. Per discussion
with Tom Dunstan. Also, make the code a shade more readable by not trying
to economize on variables.
continuously, and requests vacuum runs of "autovacuum workers" to postmaster.
The workers do the actual vacuum work. This allows for future improvements,
like allowing multiple autovacuum jobs running in parallel.
For now, the code keeps the original behavior of having a single autovac
process at any time by sleeping until the previous worker has finished.
equality checks it applies, instead of a random dependence on whatever
operators might be named "=". The equality operators will now be selected
from the opfamily of the unique index that the FK constraint depends on to
enforce uniqueness of the referenced columns; therefore they are certain to be
consistent with that index's notion of equality. Among other things this
should fix the problem noted awhile back that pg_dump may fail for foreign-key
constraints on user-defined types when the required operators aren't in the
search path. This also means that the former warning condition about "foreign
key constraint will require costly sequential scans" is gone: if the
comparison condition isn't indexable then we'll reject the constraint
entirely. All per past discussions.
Along the way, make the RI triggers look into pg_constraint for their
information, instead of using pg_trigger.tgargs; and get rid of the always
error-prone fixed-size string buffers in ri_triggers.c in favor of building up
the RI queries in StringInfo buffers.
initdb forced due to columns added to pg_constraint and pg_trigger.
entries for the victim database go away sooner rather than later. We already
did the equivalent thing at the per-relation level, not sure why it's not
been done for whole databases. With this change, pgstat_vacuum_tabstat
should usually not find anything to do; though we still need it as a backstop
in case DROPDB or TABPURGE messages get lost under load.
thought that it didn't have to reposition the underlying tuplestore if the
portal is atEnd. But this is not so, because tuplestores have separate read
and write cursors ... and the read cursor hasn't moved from the start.
This mistake explains bug #2970 from William Zhang.
Note: the coding here is pretty inefficient, but given that no one has noticed
this bug until now, I'd say hardly anyone uses the case where the cursor has
been advanced before being persisted. So maybe it's not worth worrying about.
describe the maximum size of index tuples (which is typically AM-dependent
anyway); and consequently remove the bogus deduction for "special space"
that was built into it.
Adjust TOAST_TUPLE_THRESHOLD and TOAST_MAX_CHUNK_SIZE to avoid wasting two
bytes per toast chunk, and to ensure that the calculation correctly tracks any
future changes in page header size. The computation had been inaccurate in a
way that didn't cause any harm except space wastage, but future changes could
have broken it more drastically.
Fix the calculation of BTMaxItemSize, which was formerly computed as 1 byte
more than it could safely be. This didn't cause any harm in practice because
it's only compared against maxalign'd lengths, but future changes in the size
of page headers or btree special space could have exposed the problem.
initdb forced because of change in TOAST_MAX_CHUNK_SIZE, which alters the
storage of toast tables.
made query plan. Use of ALTER COLUMN TYPE creates a hazard for cached
query plans: they could contain Vars that claim a column has a different
type than it now has. Fix this by checking during plan startup that Vars
at relation scan level match the current relation tuple descriptor. Since
at that point we already have at least AccessShareLock, we can be sure the
column type will not change underneath us later in the query. However,
since a backend's locks do not conflict against itself, there is still a
hole for an attacker to exploit: he could try to execute ALTER COLUMN TYPE
while a query is in progress in the current backend. Seal that hole by
rejecting ALTER TABLE whenever the target relation is already open in
the current backend.
This is a significant security hole: not only can one trivially crash the
backend, but with appropriate misuse of pass-by-reference datatypes it is
possible to read out arbitrary locations in the server process's memory,
which could allow retrieving database content the user should not be able
to see. Our thanks to Jeff Trout for the initial report.
Security: CVE-2007-0556
Standard English uses "may", "can", and "might" in different ways:
may - permission, "You may borrow my rake."
can - ability, "I can lift that log."
might - possibility, "It might rain today."
Unfortunately, in conversational English, their use is often mixed, as
in, "You may use this variable to do X", when in fact, "can" is a better
choice. Similarly, "It may crash" is better stated, "It might crash".
FAMILY; and add FAMILY option to CREATE OPERATOR CLASS to allow adding a
class to a pre-existing family. Per previous discussion. Man, what a
tedious lot of cutting and pasting ...
columns procost and prorows, to allow simple user adjustment of the estimated
cost of a function call, as well as control of the estimated number of rows
returned by a set-returning function. We might eventually wish to extend this
to allow function-specific estimation routines, but there seems to be
consensus that we should try a simple constant estimate first. In particular
this provides a relatively simple way to control the order in which different
WHERE clauses are applied in a plan node, which is a Good Thing in view of the
fact that the recent EquivalenceClass planner rewrite made that much less
predictable than before.
provide just a boolean 'amcanorder', instead of fields that specify the
sort operator strategy numbers. We have decided to require ordering-capable
AMs to use btree-compatible strategy numbers, so the old fields are
overkill (and indeed misleading about what's allowed).
accessing it, like DROP DATABASE. This allows the regression tests to pass
with autovacuum enabled, which open the gates for finally enabling autovacuum
by default.
per-column options for btree indexes. The planner's support for this is still
pretty rudimentary; it does not yet know how to plan mergejoins with
nondefault ordering options. The documentation is pretty rudimentary, too.
I'll work on improving that stuff later.
Note incompatible change from prior behavior: ORDER BY ... USING will now be
rejected if the operator is not a less-than or greater-than member of some
btree opclass. This prevents less-than-sane behavior if an operator that
doesn't actually define a proper sort ordering is selected.
having md.c return a success/failure boolean to smgr.c, which was just going
to elog anyway, let md.c issue the elog messages itself. This allows better
error reporting, particularly in cases such as "short read" or "short write"
which Peter was complaining of. Also, remove the kluge of allowing mdread()
to return zeroes from a read-beyond-EOF: this is now an error condition
except when InRecovery or zero_damaged_pages = true. (Hash indexes used to
require that behavior, but no more.) Also, enforce that mdwrite() is to be
used for rewriting existing blocks while mdextend() is to be used for
extending the relation EOF. This restriction lets us get rid of the old
ad-hoc defense against creating huge files by an accidental reference to
a bogus block number: we'll only create new segments in mdextend() not
mdwrite() or mdread(). (Again, when InRecovery we allow it anyway, since
we need to allow updates of blocks that were later truncated away.)
Also, clean up the original makeshift patch for bug #2737: move the
responsibility for padding relation segments to full length into md.c.
Tom Lane
Andrew Dunstan
Greg Stark
Alvaro Herrera
Robert Haas
Heikki Linnakangas
Peter Eisentraut
Simon Riggs
Bruce Momjian
Itagaki Takahiro
Michael Meskes
Magnus Hagander
Neil Conway
Jan Wieck