Robert Haas [Tue, 21 Dec 2010 11:30:32 +0000 (06:30 -0500)]
Work around unfortunate getppid() behavior on BSD-ish systems.
On MacOS X, and apparently also on other BSD-derived systems, attaching
a debugger causes getppid() to return the pid of the debugging process
rather than the actual parent PID. As a result, debugging the
autovacuum launcher, startup process, or WAL sender on such systems
causes it to exit, because the previous coding of PostmasterIsAlive()
detects postmaster death by testing whether getppid() == PostmasterPid.
Work around that behavior by checking the return value of getppid()
more carefully. If it's PostmasterPid, the postmaster must be alive;
if it's 1, assume the postmaster is dead. If it's any other value,
assume we've been debugged and fall through to the less-reliable
kill() test.
Robert Haas [Mon, 20 Dec 2010 17:59:33 +0000 (12:59 -0500)]
Allow transactions that don't write WAL to commit asynchronously.
This case can arise if a transaction has written data, but only to
temporary tables. Loss of the commit record in case of a crash won't
matter, because the temporary tables will be lost anyway.
Magnus Hagander [Sun, 19 Dec 2010 20:31:23 +0000 (21:31 +0100)]
Remove thread dumping constant that requires newer Platform SDK
Since we're not multithreaded it only provides marginally useful
information, and it does require a newer version of the Platform SDK
than we target. We may want to reconsider this in the future along
with a fix for MinGW.
Tom Lane [Sun, 19 Dec 2010 20:30:44 +0000 (15:30 -0500)]
Fix up handling of simple-form CASE with constant test expression.
eval_const_expressions() can replace CaseTestExprs with constants when
the surrounding CASE's test expression is a constant. This confuses
ruleutils.c's heuristic for deparsing simple-form CASEs, leading to
Assert failures or "unexpected CASE WHEN clause" errors. I had put in
a hack solution for that years ago (see commit 514ce7a331c5bea8e55b106d624e55732a002295 of 2006-10-01), but bug #5794
from Peter Speck shows that that solution failed to cover all cases.
Fortunately, there's a much better way, which came to me upon reflecting
that Peter's "CASE TRUE WHEN" seemed pretty redundant: we can "simplify"
the simple-form CASE to the general form of CASE, by simply omitting the
constant test expression from the rebuilt CASE construct. This is
intuitively valid because there is no need for the executor to evaluate
the test expression at runtime; it will never be referenced, because any
CaseTestExprs that would have referenced it are now replaced by constants.
This won't save a whole lot of cycles, since evaluating a Const is pretty
cheap, but a cycle saved is a cycle earned. In any case it beats kluging
ruleutils.c still further. So this patch improves const-simplification
and reverts the previous change in ruleutils.c.
Back-patch to all supported branches. The bug exists in 8.1 too, but it's
out of warranty.
After parsing a parenthesized subexpression, we must pop all pending
ANDs and NOTs off the stack, just like the case for a simple operand.
Per bug #5793.
Also fix clones of this routine in contrib/intarray and contrib/ltree,
where input of types query_int and ltxtquery had the same problem.
Magnus Hagander [Sun, 19 Dec 2010 15:45:28 +0000 (16:45 +0100)]
Support for collecting crash dumps on Windows
Add support for collecting "minidump" style crash dumps on
Windows, by setting up an exception handling filter. Crash
dumps will be generated in PGDATA/crashdumps if the directory
is created (the existance of the directory is used as on/off
switch for the generation of the dumps).
Robert Haas [Fri, 17 Dec 2010 13:30:57 +0000 (08:30 -0500)]
Reset 'ps' display just once when resolving VXID conflicts.
This prevents the word "waiting" from briefly disappearing from the ps
status line when ResolveRecoveryConflictWithVirtualXIDs begins a new
iteration of the outer loop.
Along the way, remove some useless pgstat_report_waiting() calls;
the startup process doesn't appear in pg_stat_activity.
Tom Lane [Thu, 16 Dec 2010 21:22:05 +0000 (16:22 -0500)]
Remove optreset from src/port/ implementations of getopt and getopt_long.
We don't actually need optreset, because we can easily fix the code to
ensure that it's cleanly restartable after having completed a scan over the
argv array; which is the only case we need to restart in. Getting rid of
it avoids a class of interactions with the system libraries and allows
reversion of my change of yesterday in postmaster.c and postgres.c.
Back-patch to 8.4. Before that the getopt code was a bit different anyway.
Bruce Momjian [Thu, 16 Dec 2010 15:13:43 +0000 (10:13 -0500)]
Fix crash caused by NULL lookup when reporting IP address of failed
libpq connection, per report from Magnus. This happens only on GIT
master and only on Win32 because that is the platform where "" maps to
an IP address (localhost).
Tom Lane [Thu, 16 Dec 2010 04:50:41 +0000 (23:50 -0500)]
Fix up getopt() reset management so it works on recent mingw.
The mingw people don't appear to care about compatibility with non-GNU
versions of getopt, so force use of our own copy of getopt on Windows.
Also, ensure that we make use of optreset when using our own copy.
Per report from Andrew Dunstan. Back-patch to all versions supported
on Windows.
Tom Lane [Thu, 16 Dec 2010 02:14:24 +0000 (21:14 -0500)]
Fix contrib/seg's GiST picksplit method.
This patch replaces Guttman's generalized split method with a simple
sort-by-center-points algorithm. Since the data is only one-dimensional
we don't really need the slow and none-too-stable Guttman method.
This is in part a bug fix, since seg has the same size_alpha versus
size_beta typo that was recently fixed in contrib/cube. It seems
prudent to apply this rather aggressive fix only in HEAD, though.
Back branches will just get the typo fix.
Itagaki Takahiro [Wed, 15 Dec 2010 21:56:28 +0000 (06:56 +0900)]
Add pg_read_binary_file() and whole-file-at-once versions of pg_read_file().
One of the usages of the binary version is to read files in a different
encoding from the server encoding.
Robert Haas [Tue, 14 Dec 2010 03:37:55 +0000 (22:37 -0500)]
Improved tab completion for views with triggers.
Allow INSERT INTO, UPDATE, and DELETE FROM to be completed with
either the name of a table (as before) or the name of a view with
an appropriate INSTEAD OF rule.
Along the way, allow CREATE TRIGGER to be completed with INSTEAD OF,
as well as BEFORE and AFTER.
Robert Haas [Mon, 13 Dec 2010 17:34:26 +0000 (12:34 -0500)]
Generalize concept of temporary relations to "relation persistence".
This commit replaces pg_class.relistemp with pg_class.relpersistence;
and also modifies the RangeVar node type to carry relpersistence rather
than istemp. It also removes removes rd_istemp from RelationData and
instead performs the correct computation based on relpersistence.
For clarity, we add three new macros: RelationNeedsWAL(),
RelationUsesLocalBuffers(), and RelationUsesTempNamespace(), so that we
can clarify the purpose of each check that previous depended on
rd_istemp.
This is intended as infrastructure for the upcoming unlogged tables
patch, as well as for future possible work on global temporary tables.
Tom Lane [Sun, 12 Dec 2010 20:09:53 +0000 (15:09 -0500)]
Reset all database-level stats in pgstat_recv_resetcounter().
We were failing to zero out some pg_stat_database counters that have
been added since the initial pgstats coding. This is a bug, but not
back-patching the fix since changing this behavior in a minor release
seems a cure worse than the disease.
This code appears to have been copied from ancient versions of the
corresponding backend routines, and not patched to incorporate subsequent
fixes (see my commits of 2008-03-31 and 2010-01-14 respectively).
It might be a good idea to change it to just *call* those routines,
but for now I'll just transpose these fixes over.
Tom Lane [Sat, 11 Dec 2010 00:42:44 +0000 (19:42 -0500)]
Move a couple of initdb's subroutines into src/port/.
mkdir_p and check_data_dir will be useful in CREATE TABLESPACE, since we
have agreed that that command should handle subdirectory creation just like
initdb creates the PGDATA directory. Push them into src/port/ so that they
are available to both initdb and the backend. Rename to pg_mkdir_p and
pg_check_dir, just to be on the safe side. Add FreeBSD's copyright notice
to pgmkdirp.c, since that's where the code came from originally (this
really should have been in initdb.c). Very marginal code/comment cleanup.
Tom Lane [Fri, 10 Dec 2010 22:35:33 +0000 (17:35 -0500)]
Use symbolic names not octal constants for file permission flags.
Purely cosmetic patch to make our coding standards more consistent ---
we were doing symbolic some places and octal other places. This patch
fixes all C-coded uses of mkdir, chmod, and umask. There might be some
other calls I missed. Inconsistency noted while researching tablespace
directory permissions issue.
Tom Lane [Fri, 10 Dec 2010 16:33:38 +0000 (11:33 -0500)]
Fix efficiency problems in tuplestore_trim().
The original coding in tuplestore_trim() was only meant to work efficiently
in cases where each trim call deleted most of the tuples in the store.
Which, in fact, was the pattern of the original usage with a Material node
supporting mark/restore operations underneath a MergeJoin. However,
WindowAgg now uses tuplestores and it has considerably less friendly
trimming behavior. In particular it can attempt to trim one tuple at a
time off a large tuplestore. tuplestore_trim() had O(N^2) runtime in this
situation because of repeatedly shifting its tuple pointer array. Fix by
avoiding shifting the array until a reasonably large number of tuples have
been deleted. This can waste some pointer space, but we do still reclaim
the tuples themselves, so the percentage wastage should be pretty small.
Per Jie Li's report of slow percent_rank() evaluation. cume_dist() and
ntile() would certainly be affected as well, along with any other window
function that has a moving frame start and requires reading substantially
ahead of the current row.
Back-patch to 8.4, where window functions were introduced. There's no
need to tweak it before that.
Tom Lane [Thu, 9 Dec 2010 18:03:11 +0000 (13:03 -0500)]
Eliminate O(N^2) behavior in parallel restore with many blobs.
With hundreds of thousands of TOC entries, the repeated searches in
reduce_dependencies() become the dominant cost. Get rid of that searching
by constructing reverse-dependency lists, which we can do in O(N) time
during the fix_dependencies() preprocessing. I chose to store the reverse
dependencies as DumpId arrays for consistency with the forward-dependency
representation, and keep the previously-transient tocsByDumpId[] array
around to locate actual TOC entry structs quickly from dump IDs.
While this fixes the slow case reported by Vlad Arkhipov, there is still
a potential for O(N^2) behavior with sufficiently many tables:
fix_dependencies itself, as well as mark_create_done and
inhibit_data_for_failed_table, are doing repeated searches to deal with
table-to-table-data dependencies. Possibly this work could be extended
to deal with that, although the latter two functions are also used in
non-parallel restore where we currently don't run fix_dependencies.
Another TODO is that we fail to parallelize restore of multiple blobs
at all. This appears to require changes in the archive format to fix.
Back-patch to 9.0 where the problem was reported. 8.4 has potential issues
as well; but since it doesn't create a separate TOC entry for each blob,
it's at much less risk of having enough TOC entries to cause real problems.
Simon Riggs [Thu, 9 Dec 2010 09:41:47 +0000 (09:41 +0000)]
Reduce spurious Hot Standby conflicts from never-visible records.
Hot Standby conflicts only with tuples that were visible at
some point. So ignore tuples from aborted transactions or for
tuples updated/deleted during the inserting transaction when
generating the conflict transaction ids.
Following detailed analysis and test case by Noah Misch.
Original report covered btree delete records, correctly observed
by Heikki Linnakangas that this applies to other cases also.
Fix covers all sources of cleanup records via common code.
Tom Lane [Thu, 9 Dec 2010 01:01:09 +0000 (20:01 -0500)]
Force default wal_sync_method to be fdatasync on Linux.
Recent versions of the Linux system header files cause xlogdefs.h to
believe that open_datasync should be the default sync method, whereas
formerly fdatasync was the default on Linux. open_datasync is a bad
choice, first because it doesn't actually outperform fdatasync (in fact
the reverse), and second because we try to use O_DIRECT with it, causing
failures on certain filesystems (e.g., ext4 with data=journal option).
This part of the patch is largely per a proposal from Marti Raudsepp.
More extensive changes are likely to follow in HEAD, but this is as much
change as we want to back-patch.
Also clean up confusing code and incorrect documentation surrounding the
fsync_writethrough option. Those changes shouldn't result in any actual
behavioral change, but I chose to back-patch them anyway to keep the
branches looking similar in this area.
In 9.0 and HEAD, also do some copy-editing on the WAL Reliability
documentation section.
Back-patch to all supported branches, since any of them might get used
on modern Linux versions.
Simon Riggs [Wed, 8 Dec 2010 18:48:03 +0000 (18:48 +0000)]
Optimize commit_siblings in two ways to improve group commit.
First, avoid scanning the whole ProcArray once we know there
are at least commit_siblings active; second, skip the check
altogether if commit_siblings = 0.
Fix bugs in the hot standby known-assigned-xids tracking logic. If there's
an old transaction running in the master, and a lot of transactions have
started and finished since, and a WAL-record is written in the gap between
the creating the running-xacts snapshot and WAL-logging it, recovery will fail
with "too many KnownAssignedXids" error. This bug was reported by
Joachim Wieland on Nov 19th.
In the same scenario, when fewer transactions have started so that all the
xids fit in KnownAssignedXids despite the first bug, a more serious bug
arises. We incorrectly initialize the clog code with the oldest still running
transaction, and when we see the WAL record belonging to a transaction with
an XID larger than one that committed already before the checkpoint we're
recovering from, we zero the clog page containing the already committed
transaction, leading to data loss.
In hindsight, trying to track xids in the known-assigned-xids array before
seeing the running-xacts record was too complicated. To fix that, hold
XidGenLock while the running-xacts snapshot is taken and WAL-logged. That
ensures that no transaction can begin or end in that gap, so that in recvoery
we know that the snapshot contains all transactions running at that point in
WAL.
Tom Lane [Tue, 7 Dec 2010 03:55:43 +0000 (22:55 -0500)]
Add a stack overflow check to copyObject().
There are some code paths, such as SPI_execute(), where we invoke
copyObject() on raw parse trees before doing parse analysis on them. Since
the bison grammar is capable of building heavily nested parsetrees while
itself using only minimal stack depth, this means that copyObject() can be
the front-line function that hits stack overflow before anything else does.
Accordingly, it had better have a check_stack_depth() call. I did a bit of
performance testing and found that this slows down copyObject() by only a
few percent, so the hit ought to be negligible in the context of complete
processing of a query.
Per off-list report from Toshihide Katayama. Back-patch to all supported
branches.
Andrew Dunstan [Mon, 6 Dec 2010 20:31:55 +0000 (15:31 -0500)]
Allow the low level COPY routines to read arbitrary numbers of fields.
This doesn't involve any user-visible change in behavior, but will be
useful when the COPY routines are exposed to allow their use by Foreign
Data Wrapper routines, which will be able to use these routines to read
irregular CSV files, for example.
Avoid eating quite so much memory for large inheritance trees, by
reclaiming the space used by temporary copies of the original parsetree and
range table, as well as the workspace needed during planning. The cost is
needing to copy the finished plan trees out of the child memory context.
Although this looks like it ought to slow things down, my testing shows
it actually is faster, apparently because fewer interactions with malloc()
are needed and/or we can do the work within a more readily cacheable amount
of memory. That result might be platform-dependent, but I'll take it.
Per a gripe from John Papandriopoulos, in which it was pointed out that the
memory consumption actually grew as O(N^2) for sufficiently many child
tables, since we were creating N copies of the N-element range table.
Tom Lane [Sat, 4 Dec 2010 18:47:08 +0000 (13:47 -0500)]
Fix two small bugs in new gistget.c logic.
1. Complain, rather than silently doing nothing, if an "invalid" tuple
is found on a leaf page. Per off-list discussion with Heikki.
2. Fix oversight in code that removes a GISTSearchItem from the search
queue: we have to reset lastHeap if this was the last heap item in the
parent GISTSearchTreeItem. Otherwise subsequent additions will do the
wrong thing. This was probably masked in early testing because in typical
cases the parent item would now be completely empty and would be deleted on
next call. You'd need a queued non-leaf page at exactly the same distance
as a heap tuple to expose the bug.
Tom Lane [Sat, 4 Dec 2010 01:52:18 +0000 (20:52 -0500)]
KNNGIST, otherwise known as order-by-operator support for GIST.
This commit represents a rather heavily editorialized version of
Teodor's builtin_knngist_itself-0.8.2 and builtin_knngist_proc-0.8.1
patches. I redid the opclass API to add a separate Distance method
instead of turning the Consistent method into an illogical mess,
fixed some bit-rot in the rbtree interfaces, and generally worked over
the code style and comments.
There's still no non-code documentation to speak of, but I'll work on
that separately. Some contrib-module changes are also yet to come
(right now, point <-> point is the only KNN-ified operator).
Tom Lane [Fri, 3 Dec 2010 01:50:48 +0000 (20:50 -0500)]
Create core infrastructure for KNNGIST.
This is a heavily revised version of builtin_knngist_core-0.9. The
ordering operators are no longer mixed in with actual quals, which would
have confused not only humans but significant parts of the planner.
Instead, ordering operators are carried separately throughout planning and
execution.
Since the API for ambeginscan and amrescan functions had to be changed
anyway, this commit takes the opportunity to rationalize that a bit.
RelationGetIndexScan no longer forces a premature index_rescan call;
instead, callers of index_beginscan must call index_rescan too. Aside from
making the AM-side initialization logic a bit less peculiar, this has the
advantage that we do not make a useless extra am_rescan call when there are
runtime key values. AMs formerly could not assume that the key values
passed to amrescan were actually valid; now they can.
Refactor the pg_dump zlib code from pg_backup_custom.c to a separate file,
to make it easier to reuse that code. There is no user-visible changes.
This is in preparation for the patch to add a new archive format, a directory,
to perform a custom-like dump but with each table being dumped to a separate
file (that in turn is a prerequisite for parallel pg_dump). This also makes it
easier to add new compression methods in the future, and makes the
pg_backup_custom.c code easier to read, when the compression-related code is
factored out.
Joachim Wieland, with heavy editorialization by me.
Tom Lane [Wed, 1 Dec 2010 05:53:18 +0000 (00:53 -0500)]
Prevent inlining a SQL function with multiple OUT parameters.
There were corner cases in which the planner would attempt to inline such
a function, which would result in a failure at runtime due to loss of
information about exactly what the result record type is. Fix by disabling
inlining when the function's recorded result type is RECORD. There might
be some sub-cases where inlining could still be allowed, but this is a
simple and backpatchable fix, so leave refinements for another day.
Per bug #5777 from Nate Carson.
Back-patch to all supported branches. 8.1 happens to avoid a core-dump
here, but it still does the wrong thing.
Tom Lane [Mon, 29 Nov 2010 17:29:42 +0000 (12:29 -0500)]
Simplify and speed up mapping of index opfamilies to pathkeys.
Formerly we looked up the operators associated with each index (caching
them in relcache) and then the planner looked up the btree opfamily
containing such operators in order to build the btree-centric pathkey
representation that describes the index's sort order. This is quite
pointless for btree indexes: we might as well just use the index's opfamily
information directly. That saves syscache lookup cycles during planning,
and furthermore allows us to eliminate the relcache's caching of operators
altogether, which may help in reducing backend startup time.
I added code to plancat.c to perform the same type of double lookup
on-the-fly if it's ever faced with a non-btree amcanorder index AM.
If such a thing actually becomes interesting for production, we should
replace that logic with some more-direct method for identifying the
corresponding btree opfamily; but it's not worth spending effort on now.
There is considerably more to do pursuant to my recent proposal to get rid
of sort-operator-based representations of sort orderings, but this patch
grabs some of the low-hanging fruit. I'll look at the remainder of that
work after the current commitfest.
Simon Riggs [Mon, 29 Nov 2010 01:08:02 +0000 (01:08 +0000)]
Move call to GetTopTransactionId() earlier in LockAcquire(),
removing an infrequently occurring race condition in Hot Standby.
An xid must be assigned before a lock appears in shared memory,
rather than immediately after, else GetRunningTransactionLocks()
may see InvalidTransactionId, causing assertion failures during
lock processing on standby.
Bug report and diagnosis by Fujii Masao, fix by me.
Tom Lane [Sat, 27 Nov 2010 06:30:34 +0000 (01:30 -0500)]
Rewrite PQping to be more like what we agreed to last week.
Basically, we want to distinguish all cases where the connection was
not made from those where it was. A convenient proxy for this is to
see if we got a message with a SQLSTATE code back from the postmaster.
This presumes that the postmaster will always send us a SQLSTATE in
a failure message, which is true for 7.4 and later postmasters in
every case except fork failure. (We could possibly complicate the
postmaster code to do something about that, but it seems not worth
the trouble, especially since pg_ctl's response for that case should
be to keep waiting anyway.)
If we did get a SQLSTATE from the postmaster, there are basically only
two cases, as per last week's discussion: ERRCODE_CANNOT_CONNECT_NOW
and everything else. Any other error code implies that the postmaster
is in principle willing to accept connections, it just didn't like or
couldn't handle this particular request. We want to make a special
case for ERRCODE_CANNOT_CONNECT_NOW so that "pg_ctl start -w" knows
it should keep waiting.
In passing, pick names for the enum constants that are a tad less
likely to present collision hazards in future.
Tom Lane [Fri, 26 Nov 2010 22:59:10 +0000 (17:59 -0500)]
Fix portability issues in new src/port/inet_net_ntop.c file.
1. Don't #include postgres.h in a frontend build.
2. Don't assume that the backend's symbol PGSQL_AF_INET6 has anything to do
with the constant that will be used by system library functions (because,
in point of fact, it usually doesn't). Fortunately, PGSQL_AF_INET is equal
to AF_INET, so we can just cater for both sets of values in one case
construct without fear of conflict.
Robert Haas [Fri, 26 Nov 2010 22:27:23 +0000 (17:27 -0500)]
Add more ALTER <object> .. SET SCHEMA commands.
This adds support for changing the schema of a conversion, operator,
operator class, operator family, text search configuration, text search
dictionary, text search parser, or text search template.
Dimitri Fontaine, with assorted corrections and other kibitzing.
Tom Lane [Fri, 26 Nov 2010 20:20:51 +0000 (15:20 -0500)]
Fix significant memory leak in contrib/xml2 functions.
Most of the functions that execute XPath queries leaked the data structures
created by libxml2. This memory would not be recovered until end of
session, so it mounts up pretty quickly in any serious use of the feature.
Per report from Pavel Stehule, though this isn't his patch.