Commit 0b0baf262 et al made this case print "(null)" on the grounds that
that's what happened on platforms that didn't crash. But neither behavior
was actually intentional. What we should print is just an empty string,
for compatibility with the behavior of SHOW and other ways of examining
string GUCs. Those code paths don't distinguish NULL from empty strings,
so we should not here either. Per gripe from Alain Radix.
Like the previous patch, back-patch to 9.2 where -C option was introduced.
Discussion: <CA+YdpwxPUADrmxSD7+Td=uOshMB1KkDN7G7cf+FGmNjjxMhjbw@mail.gmail.com>
The original upper-planner-pathification design (commit 3fc6e2d7f5)
assumed that we could always determine during Path formation whether or not
we would need a Result plan node to perform projection of a targetlist.
That turns out not to work very well, though, because createplan.c still
has some responsibilities for choosing the specific target list associated
with sorting/grouping nodes (in particular it might choose to add resjunk
columns for sorting). We might not ever refactor that --- doing so would
push more work into Path formation, which isn't attractive --- and we
certainly won't do so for 9.6. So, while create_projection_path and
apply_projection_to_path can tell for sure what will happen if the subpath
is projection-capable, they can't tell for sure when it isn't. This is at
least a latent bug in apply_projection_to_path, which might think it can
apply a target to a non-projecting node when the node will end up computing
something different.
Also, I'd tied the creation of a ProjectionPath node to whether or not a
Result is needed, but it turns out that we sometimes need a ProjectionPath
node anyway to avoid modifying a possibly-shared subpath node. Callers had
to use create_projection_path for such cases, and we added code to them
that knew about the potential omission of a Result node and attempted to
adjust the cost estimates for that. That was uncertainly correct and
definitely ugly/unmaintainable.
To fix, have create_projection_path explicitly check whether a Result
is needed and adjust its cost estimate accordingly, though it creates
a ProjectionPath in either case. apply_projection_to_path is now mostly
just an optimized version that can avoid creating an extra Path node when
the input is known to not be shared with any other live path. (There
is one case that create_projection_path doesn't handle, which is pushing
parallel-safe expressions below a Gather node. We could make it do that
by duplicating the GatherPath, but there seems no need as yet.)
create_projection_plan still has to recheck the tlist-match condition,
which means that if the matching situation does get changed by createplan.c
then we'll have made a slightly incorrect cost estimate. But there seems
no help for that in the near term, and I doubt it occurs often enough,
let alone would change planning decisions often enough, to be worth
stressing about.
I added a "dummypp" field to ProjectionPath to track whether
create_projection_path thinks a Result is needed. This is not really
necessary as-committed because create_projection_plan doesn't look at the
flag; but it seems like a good idea to remember what we thought when
forming the cost estimate, if only for debugging purposes.
In passing, get rid of the target_parallel parameter added to
apply_projection_to_path by commit 54f5c5150. I don't think that's a good
idea because it involves callers in what should be an internal decision,
and opens us up to missing optimization opportunities if callers think they
don't need to provide a valid flag, as most don't. For the moment, this
just costs us an extra has_parallel_hazard call when planning a Gather.
If that starts to look expensive, I think a better solution would be to
teach PathTarget to carry/cache knowledge of parallel-safety of its
contents.
The annotation for "ERROR: language "foo" is not trusted" used to say
"HINT: Only superusers can use untrusted languages", which was fairly
poorly thought out. For one thing, it's not a hint about what to do,
but a statement of fact, which makes it errdetail. But also, this
fails to clarify things much, because there's a missing step in the
chain of reasoning. I think it's more useful to say "GRANT and REVOKE
are not allowed on untrusted languages, because only superusers can use
untrusted languages".
It's been like this for a long time, but given the lack of previous
complaints, I don't think this is worth back-patching.
Discussion: <1417.1466289901@sss.pgh.pa.us>
This patch provides a new implementation of the logic added by commit
137805f89 and later removed by 77ba61080. It differs from the original
primarily in expending much less effort per joinrel in large queries,
which it accomplishes by doing most of the matching work once per query not
once per joinrel. Hopefully, it's also less buggy and better commented.
The never-documented enable_fkey_estimates GUC remains gone.
There remains work to be done to make the selectivity estimates account
for nulls in FK referencing columns; but that was true of the original
patch as well. We may be able to address this point later in beta.
In the meantime, any error should be in the direction of overestimating
rather than underestimating joinrel sizes, which seems like the direction
we want to err in.
Tomas Vondra and Tom Lane
Discussion: <31041.1465069446@sss.pgh.pa.us>
The previous code neglected the fact that the scanjoin_target might
carry sortgroupref labelings that we need to absorb. Instead, do
create_projection_path() unconditionally, and tweak the path's cost
estimate after the fact. (I'm now convinced that we ought to refactor
the way we account for sometimes not needing a separate projection step,
but right now is not the time for that sort of cleanup.)
Problem identified by Amit Kapila, patch by me.
When doing partial aggregation, the args list of the upper (combining)
Aggref node is replaced by a Var representing the output of the partial
aggregation steps, which has either the aggregate's transition data type
or a serialized representation of that. However, nodeAgg.c blindly
continued to use the args list as an indication of the user-level argument
types. This broke resolution of polymorphic transition datatypes at
executor startup (though it accidentally failed to fail for the ANYARRAY
case, which is likely the only one anyone had tested). Moreover, the
constructed FuncExpr passed to the finalfunc contained completely wrong
information, which would have led to bogus answers or crashes for any case
where the finalfunc examined that information (which is only likely to be
with polymorphic aggregates using a non-polymorphic transition type).
As an independent bug, apply_partialaggref_adjustment neglected to resolve
a polymorphic transition datatype before assigning it as the output type
of the lower-level Aggref node. This again accidentally failed to fail
for ANYARRAY but would be unlikely to work in other cases.
To fix the first problem, record the user-level argument types in a
separate OID-list field of Aggref, and look to that rather than the args
list when asking what the argument types were. (It turns out to be
convenient to include any "direct" arguments in this list too, although
those are not currently subject to being overwritten.)
Rather than adding yet another resolve_aggregate_transtype() call to fix
the second problem, add an aggtranstype field to Aggref, and store the
resolved transition type OID there when the planner first computes it.
(By doing this in the planner and not the parser, we can allow the
aggregate's transition type to change from time to time, although no DDL
support yet exists for that.) This saves nothing of consequence for
simple non-polymorphic aggregates, but for polymorphic transition types
we save a catalog lookup during executor startup as well as several
planner lookups that are new in 9.6 due to parallel query planning.
In passing, fix an error that was introduced into count_agg_clauses_walker
some time ago: it was applying exprTypmod() to something that wasn't an
expression node at all, but a TargetEntry. exprTypmod silently returned
-1 so that there was not an obvious failure, but this broke the intended
sensitivity of aggregate space consumption estimates to the typmod of
varchar and similar data types. This part needs to be back-patched.
Catversion bump due to change of stored Aggref nodes.
Discussion: <8229.1466109074@sss.pgh.pa.us>
This requires some core changes as well so that we can properly
WAL-log the truncation. Specifically, it changes the format of the
XLOG_SMGR_TRUNCATE WAL record, so bump XLOG_PAGE_MAGIC.
Patch by me, reviewed but not fully endorsed by Andres Freund.
Commit 04ae11f62e removed some broken
code to apply the scan/join target to partial paths, but its theory
that this processing step is totally unnecessary turns out to be wrong.
Put similar code back again, but this time, check for parallel-safety
and avoid in-place modifications to paths that may already have been
used as part of some other path.
(This is not an entirely elegant solution to this problem; it might
be better, for example, to postpone generate_gather_paths for the
topmost scan/join rel until after the scan/join target has been
applied. But this is not the time for such redesign work.)
Amit Kapila and Robert Haas
If you really want to vacuum every single page in the relation,
regardless of apparent visibility status or anything else, you can use
this option. In previous releases, this behavior could be achieved
using VACUUM (FREEZE), but because we can now recognize all-frozen
pages as not needing to be frozen again, that no longer works. There
should be no need for routine use of this option, but maybe bugs or
disaster recovery will necessitate its use.
Patch by me, reviewed by Andres Freund.
The inet/cidr types sometimes failed to reject IPv6 inputs with too many
colon-separated fields, instead translating them to '::/0'. This is the
result of a thinko in the original ISC code that seems to be as yet
unreported elsewhere. Per bug #14198 from Stefan Kaltenbrunner.
Report: <20160616182222.5798.959@wrigleys.postgresql.org>
The fact that no workers were successfully launched in the previous
iteration does not excuse us from setting up properly to try again.
This appears to explain crashes I saw in parallel regression testing
due to error_mqh being NULL when it shouldn't be.
Minor other cosmetic fixes too.
The main point of doing this is to allow the cutoff to be set very small,
even zero, to allow parallel-query behavior to be tested on relatively
small tables such as we typically use in the regression tests. But it
might be of use to users too. The number-of-workers scaling behavior in
create_plain_partial_paths() is pretty ad-hoc and subject to change, so
we won't expose anything about that, but the notion of not considering
parallel query at all for tables below size X seems reasonably stable.
Amit Kapila, per a suggestion from me
Discussion: <17170.1465830165@sss.pgh.pa.us>
Emit "(null)" instead, which was the behavior all along on platforms
that don't crash, eg OS X. Per report from Jehan-Guillaume de Rorthais.
Back-patch to 9.2 where -C option was introduced.
Michael Paquier
Report: <20160615204036.2d35d86a@firost>
Commit a892234f83 added a new bit per
page to the visibility map fork indicating whether the page is
all-frozen, but incorrectly assumed that if lazy_scan_heap chose to
freeze a tuple then that tuple would not need to later be frozen
again. This turns out to be false, because xmin and xmax (and
conceivably xvac, if dealing with tuples from very old releases) could
be frozen at separate times.
Thanks to Andres Freund for help in uncovering and tracking down this
issue.
We disable statement_timeout and lock_timeout during dump and restore, to
prevent any global settings that might exist from breaking routine backups.
Commit c6dda1f48 should have added idle_in_transaction_session_timeout to
that list, but failed to.
Another place where these timeouts get turned off is autovacuum. While
I doubt an idle timeout could fire there, it seems better to be safe than
sorry.
pg_dump issue noted by Bernd Helmle, the other one found by grepping.
Report: <352F9B77DB5D3082578D17BB@eje.land.credativ.lan>
pg_type_aclmask reported the wrong type's OID when complaining that
it could not find a type's typelem. It also failed to provide a
suitable errcode when the initially given OID doesn't exist (which
is a user-facing error, since that OID can be user-specified).
pg_foreign_data_wrapper_aclmask and pg_foreign_server_aclmask likewise
lacked errcode specifications. Trivial cosmetic adjustments too.
The wrong-type-OID problem was reported by Petru-Florin Mihancea in
bug #14186; the other issues noted by me while reading the code.
These errors all seem to be aboriginal in the respective routines, so
back-patch as necessary.
Report: <20160613163159.5798.52928@wrigleys.postgresql.org>
The struct definition for PathTarget specifies that a NULL sortgrouprefs
pointer means no sortgroupref labels. While it's likely that there
should always be at least one labeled column in the places that were
unconditionally fetching through the pointer, it seems wiser to adhere to
the data structure specification and test first. Add a macro to make this
convenient. Per experimentation with running the regression tests with a
very small parallelization threshold --- the crash I observed may well
represent a bug elsewhere, but still this coding was not very robust.
Report: <20756.1465834072@sss.pgh.pa.us>
While beneficial, both for throughput and average/worst case latency, in
a significant number of workloads, there are other workloads in which
backend_flush_after can cause significant performance regressions in
comparison to < 9.6 releases. The regression is most likely when the hot
data set is bigger than shared buffers, but significantly smaller than
the operating system's page cache.
I personally think that the benefit of enabling backend flush control is
considerably bigger than the potential downsides, but a fair argument
can be made that not regressing is more important than improving
performance/latency. As the latter is the consensus, change the default
to 0.
The other settings introduced in 428b1d6b2 do not have the same
potential for regressions, so leave them enabled.
Benchmarks leading up to changing the default have been performed by
Mithun Cy, Ashutosh Sharma and Robert Haas.
Discussion: CAD__OuhPmc6XH=wYRm_+Q657yQE88DakN4=Ybh2oveFasHkoeA@mail.gmail.com
Commit b12fd41c6 added a "reltarget_has_non_vars" field to RelOptInfo,
but failed to maintain it accurately. Since its only purpose was to skip
calls to has_parallel_hazard() in the simple case where a rel's targetlist
is all Vars, and that call is really pretty cheap in that case anyway, it
seems like this is just a case of premature optimization. Let's drop the
flag and do the calls unconditionally until it's proven that we need more
smarts here.
As noted by Andres Freund, we'd accumulated quite a few similar functions
in clauses.c that examine all functions in an expression tree to see if
they satisfy some boolean test. Reduce the duplication by inventing a
function check_functions_in_node() that applies a simple callback function
to each SQL function OID appearing in a given expression node. This also
fixes some arguable oversights; for example, contain_mutable_functions()
did not check aggregate or window functions for mutability. I doubt that
that represents a live bug at the moment, because we don't really consider
mutability for aggregates; but it might someday be one.
I chose to put check_functions_in_node() in nodeFuncs.c because it seemed
like other modules might wish to use it in future. That in turn forced
moving set_opfuncid() et al into nodeFuncs.c, as the alternative was for
nodeFuncs.c to depend on optimizer/setrefs.c which didn't seem very clean.
In passing, teach contain_leaked_vars_walker() about a few more expression
node types it can safely look through, and improve the rather messy and
undercommented code in has_parallel_hazard_walker().
Discussion: <20160527185853.ziol2os2zskahl7v@alap3.anarazel.de>
Since indexes are created without valid LSNs, an index created
while a snapshot older than old_snapshot_threshold existed could
cause queries to return incorrect results when those old snapshots
were used, if any relevant rows had been subject to early pruning
before the index was built. Prevent usage of a newly created index
until all such snapshots are released, for relations where this can
happen.
Questions about the interaction of "snapshot too old" with index
creation were initially raised by Andres Freund.
Reviewed by Robert Haas.
Transmit the leader's temp-namespace state to workers. This is important
because without it, the workers do not really have the same search path
as the leader. For example, there is no good reason (and no extant code
either) to prevent a worker from executing a temp function that the
leader created previously; but as things stood it would fail to find the
temp function, and then either fail or execute the wrong function entirely.
We still prohibit a worker from creating a temp namespace on its own.
In effect, a worker can only see the session's temp namespace if the leader
had created it before starting the worker, which seems like the right
semantics.
Also, transmit the leader's BackendId to workers, and arrange for workers
to use that when determining the physical file path of a temp relation
belonging to their session. While the original intent was to prevent such
accesses entirely, there were a number of holes in that, notably in places
like dbsize.c which assume they can safely access temp rels of other
sessions anyway. We might as well get this right, as a small down payment
on someday allowing workers to access the leader's temp tables. (With
this change, directly using "MyBackendId" as a relation or buffer backend
ID is deprecated; you should use BackendIdForTempRelations() instead.
I left a couple of such uses alone though, as they're not going to be
reachable in parallel workers until we do something about localbuf.c.)
Move the thou-shalt-not-access-thy-leader's-temp-tables prohibition down
into localbuf.c, which is where it actually matters, instead of having it
in relation_open(). This amounts to recognizing that access to temp
tables' catalog entries is perfectly safe in a worker, it's only the data
in local buffers that is problematic.
Having done all that, we can get rid of the test in has_parallel_hazard()
that says that use of a temp table's rowtype is unsafe in parallel workers.
That test was unduly expensive, and if we really did need such a
prohibition, that was not even close to being a bulletproof guard for it.
(For example, any user-defined function executed in a parallel worker
might have attempted such access.)
Such paths are unsafe. To make it cheaper to detect when this case
applies, track whether a relation's default PathTarget contains any
non-Vars. In most cases, the answer will be no, which enables us to
determine cheaply that the target list for a proposed path is
parallel-safe. However, subquery pull-up can create cases that
require us to inspect the target list more carefully.
Amit Kapila, reviewed by me.
Fix a couple of overlooked uses of "degree" terminology. Make the parallel
worker count selection logic in create_plain_partial_paths more robust (in
particular, it failed with max_parallel_workers_per_gather set to zero).
This terminology provoked widespread complaints. So, instead, rename
the GUC max_parallel_degree to max_parallel_workers_per_gather
(leaving room for a possible future GUC max_parallel_workers that acts
as a system-wide limit), and rename the parallel_degree reloption to
parallel_workers. Rename structure members to match.
These changes create a dump/restore hazard for users of PostgreSQL
9.6beta1 who have set the reloption (or applied the GUC using ALTER
USER or ALTER DATABASE).
This commit reverts 137805f89 as well as the associated commits 015e88942,
5306df283, and 68d704edb. We found multiple bugs in this feature, and
there was concern about possible planner slowdown (though to be fair,
exhibiting a very large slowdown proved difficult). The way forward
requires a considerable rewrite, which may or may not be possible to
accomplish in time for beta2. In my judgment reviewing the rewrite will
be easier to accomplish starting from a clean slate, so let's temporarily
revert what's there now. This also leaves us in a safe state if it turns
out to be necessary to postpone the rewrite to the next development cycle.
Discussion: <20160429102531.GA13701@huehner.biz>
If we ANALYZE only selected columns of a table, we should not postpone
auto-analyze because of that; other columns may well still need stats
updates. As committed, the counter is left alone if a column list is
given, whether or not it includes all analyzable columns of the table.
Per complaint from Tomasz Ostrowski.
It's been like this a long time, so back-patch to all supported branches.
Report: <ef99c1bd-ff60-5f32-2733-c7b504eb960c@ato.waw.pl>
If a Gather node has read as many tuples as it needs (for example, due
to Limit) it may detach the queue connecting it to the worker before
reading all of the worker's tuples. Rather than let the worker
continue to generate and send all of the results, have it stop after
sending the next tuple.
More could be done here to stop the worker even quicker, but this is
about as well as we can hope to do for 9.6.
This is in response to a problem report from Andreas Seltenreich.
Commit 44339b892a should be actually be
sufficient to fix that example even without this change, but it seems
better to do this, too, since we might otherwise waste quite a large
amount of effort in one or more workers.
Discussion: CAA4eK1KOKGqmz9bGu+Z42qhRwMbm4R5rfnqsLCNqFs9j14jzEA@mail.gmail.com
Amit Kapila
Prior to this patch, it was occasionally possible, after shm_mq_sendv
had previously returned SHM_MQ_DETACHED, for a later shm_mq_sendv
operation to fail an assertion instead of just again returning
SHM_MQ_ATTACHED. From the shm_mq code's point of view, it was
expecting to be called again with the same arguments, since the
previous operation had only partially completed. However, a caller
who isn't using non-blocking mode won't be prepared to repeat the call
with the same arguments, and this code shouldn't expect that they
will. Repair in such a way that we'll be OK whether the next call
uses the same arguments or not.
Found by Andreas Seltenreich. Analysis and sketch of fix by Amit
Kapila. Patch by me, reviewed by Amit Kapila.
Fix still another bug in commit 35fcb1b3d: it failed to fully initialize
the SortSupport states it introduced to allow the executor to re-check
ORDER BY expressions containing distance operators. That led to a null
pointer dereference if the sortsupport code tried to use ssup_cxt. The
problem only manifests in narrow cases, explaining the lack of previous
field reports. It requires a GiST-indexable distance operator that lacks
SortSupport and is on a pass-by-ref data type, which among core+contrib
seems to be only btree_gist's interval opclass; and it requires the scan
to be done as an IndexScan not an IndexOnlyScan, which explains how
btree_gist's regression test didn't catch it. Per bug #14134 from
Jihyun Yu.
Peter Geoghegan
Report: <20160511154904.2603.43889@wrigleys.postgresql.org>
It'd be good for "(x AND y) AND z" to produce a three-child AND node
whether or not operator_precedence_warning is on, but that failed to
happen when it's on because makeAndExpr() didn't look through the added
AEXPR_PAREN node. This has no effect on generated plans because prepqual.c
would flatten the AND nest anyway; but it does affect the number of parens
printed in ruleutils.c, for example. I'd already fixed some similar
hazards in parse_expr.c in commit abb164655, but didn't think to search
gram.y for problems of this ilk. Per gripe from Jean-Pierre Pelletier.
Report: <fa0535ec6d6428cfec40c7e8a6d11156@mail.gmail.com>
This attempts to buy back some of whatever performance we lost from fixing
bug #14174 by inlining the initial checks in MakeExpandedObjectReadOnly()
into the callers. We can do that in a macro without creating multiple-
evaluation hazards, so it's pretty much free notationally; and the amount
of code added to callers should be minimal as well. (Testing a value can't
take many more instructions than passing it to a subroutine.)
Might as well inline DatumIsReadWriteExpandedObject() while we're at it.
This is an ABI break for callers, so it doesn't seem safe to put into 9.5,
but I see no reason not to do it in HEAD.
Further thought about bug #14174 motivated me to try the case of a
R/W datum being returned from a VALUES list, and sure enough it was
broken. Fix that.
Also add a regression test case exercising the same scenario for
FunctionScan. That's not broken right now, because the function's
result will get shoved into a tuplestore between generation and use;
but it could easily become broken whenever we get around to optimizing
FunctionScan better.
There don't seem to be any other places where we put the result of
expression evaluation into a virtual tuple slot that could then be
the source for Vars of further expression evaluation, so I think
this is the end of this bug.
If a plan node output expression returns an "expanded" datum, and that
output column is referenced in more than one place in upper-level plan
nodes, we need to ensure that what is returned is a read-only reference
not a read/write reference. Otherwise one of the referencing sites could
scribble on or even delete the expanded datum before we have evaluated the
others. Commit 1dc5ebc907, which introduced this feature, supposed
that it'd be sufficient to make SubqueryScan nodes force their output
columns to read-only state. The folly of that was revealed by bug #14174
from Andrew Gierth, and really should have been immediately obvious
considering that the planner will happily optimize SubqueryScan nodes
out of the plan without any regard for this issue.
The safest fix seems to be to make ExecProject() force its results into
read-only state; that will cover every case where a plan node returns
expression results. Actually we can delegate this to ExecTargetList()
since we can recursively assume that plain Vars will not reference
read-write datums. That should keep the extra overhead down to something
minimal. We no longer need ExecMakeSlotContentsReadOnly(), which was
introduced only in support of the idea that just a few plan node types
would need to do this.
In the future it would be nice to have the planner account for this problem
and inject force-to-read-only expression evaluation nodes into only the
places where there's a risk of multiple evaluation. That's not a suitable
solution for 9.5 or even 9.6 at this point, though.
Report: <20160603124628.9932.41279@wrigleys.postgresql.org>
The partial paths that get modified may already have been used as
part of a GatherPath which appears in the path list, so modifying
them is not a good idea at this stage - especially because this
code has no check that the PathTarget is in fact parallel-safe.
When partial aggregation is being performed, this is actually
harmless because we'll end up replacing the pathtargets here with
the correct ones within create_grouping_paths(). But if we've got
a query tree containing only scan/join operations then this can
result in incorrectly pushing down parallel-restricted target
list entries. If those are, for example, references to subqueries,
that can crash the server; but it's wrong in any event.
Amit Kapila
The "snapshot too old" condition was not being recognized when
using a copied snapshot, since the original timestamp and lsn were
not being passed along. Noticed when testing the combination of
"snapshot too old" with parallel query execution.
Mostly these are just comments but there are a few in documentation
and a handful in code and tests. Hopefully this doesn't cause too much
unnecessary pain for backpatching. I relented from some of the most
common like "thru" for that reason. The rest don't seem numerous
enough to cause problems.
Thanks to Kevin Lyda's tool https://pypi.python.org/pypi/misspellings
Per post-commit review comments from Andres Freund, improve variable
names, comments, and in one place, slightly improve the code structure.
Masahiko Sawada
Use MAXALIGN size/alignment to guarantee that later uses of memory are
aligned correctly. E.g. epoll_event might need 8 byte alignment on some
platforms, but earlier allocations like WaitEventSet and WaitEvent might
not sized to guarantee that when purely using sizeof().
Found by myself while testing on an Sun Ultra 5 (Sparc IIi) with some
editorializing by Andres Freund.
In passing fix a couple typos in the area
Commit 2ed5b87f96 introduced a bug in
mark/restore, in an attempt to optimize repeated restores to the
same page. This caused an assertion failure during a merge join
which fed directly from an index scan, although the impact would
not be limited to that case. Revert the bad chunk of code from
that commit.
While investigating this bug it was discovered that a particular
"paranoia" set of the mark position field would not prevent bad
behavior; it would just make it harder to diagnose. Change that
into an assertion, which will draw attention to any future problem
in that area more directly.
Backpatch to 9.5, where the bug was introduced.
Bug #14169 reported by Shinta Koyanagi.
Preliminary analysis by Tom Lane identified which commit caused
the bug.
The original intent in the stats collector was that we should not write out
stats data oftener than every PGSTAT_STAT_INTERVAL msec. Backends will not
make requests at all if they see the existing data is newer than that, and
the stats collector is supposed to disregard requests having a cutoff_time
older than its most recently written data, so that close-together requests
don't result in multiple writes. But the latter part of that got broken
in commit 187492b6c2, so that if two backends concurrently decide
the existing stats are too old, the collector would write the data twice.
(In principle the collector's logic would still merge requests as long as
the second one arrives before we've actually written data ... but since
the message collection loop would write data immediately after processing
a single inquiry message, that never happened in practice, and in any case
the window in which it might work would be much shorter than
PGSTAT_STAT_INTERVAL.)
To fix, improve pgstat_recv_inquiry so that it checks whether the cutoff
time is too old, and doesn't add a request to the queue if so. This means
that we do not need DBWriteRequest.request_time, because the decision is
taken before making a queue entry. And that means that we don't really
need the DBWriteRequest data structure at all; an OID list of database
OIDs will serve and allow removal of some rather verbose and crufty code.
In passing, improve the comments in this area, which have been rather
neglected. Also change backend_read_statsfile so that it's not silently
relying on MyDatabaseId to have some particular value in the autovacuum
launcher process. It accidentally worked as desired because MyDatabaseId
is zero in that process; but that does not seem like a dependency we want,
especially with no documentation about it.
Although this patch is mine, it turns out I'd rediscovered a known bug,
for which Tomas Vondra had already submitted a patch that's functionally
equivalent to the non-cosmetic aspects of this patch. Thanks to Tomas
for reviewing this version.
Back-patch to 9.3 where the bug was introduced.
Prior-Discussion: <1718942738eb65c8407fcd864883f4c8@fuzzy.cz>
Patch: <4625.1464202586@sss.pgh.pa.us>
BRIN was relying on the ability to remove a tuple from an index page,
then putting another tuple in the same line pointer. But PageAddItem
refuses to add a tuple beyond the first free item past the last used
item, and in particular, it rejects an attempt to add an item to an
empty page anywhere other than the first line pointer. PageAddItem
issues a WARNING and indicates to the caller that it failed, which in
turn causes the BRIN calling code to issue a PANIC, so the whole
sequence looks like this:
WARNING: specified item offset is too large
PANIC: failed to add BRIN tuple
To fix, create a new function PageAddItemExtended which is like
PageAddItem except that the two boolean arguments become a flags bitmap;
the "overwrite" and "is_heap" boolean flags in PageAddItem become
PAI_OVERWITE and PAI_IS_HEAP flags in the new function, and a new flag
PAI_ALLOW_FAR_OFFSET enables the behavior required by BRIN.
PageAddItem() retains its original signature, for compatibility with
third-party modules (other callers in core code are not modified,
either).
Also, in the belt-and-suspenders spirit, I added a new sanity check in
brinGetTupleForHeapBlock to raise an error if an TID found in the revmap
is not marked as live by the page header. This causes it to react with
"ERROR: corrupted BRIN index" to the bug at hand, rather than a hard
crash.
Backpatch to 9.5.
Bug reported by Andreas Seltenreich as detected by his handy sqlsmith
fuzzer.
Discussion: https://www.postgresql.org/message-id/87mvni77jh.fsf@elite.ansel.ydns.eu
The IF EXISTS option was documented, and implemented in the grammar, but
it didn't actually work for lack of support in does_not_exist_skipping().
Per bug #14160.
Report and patch by Kouhei Sutou
Report: <20160527070433.19424.81712@wrigleys.postgresql.org>
If both timeout indicators are set when we arrive at ProcessInterrupts,
we've historically just reported "lock timeout". However, some buildfarm
members have been observed to fail isolationtester's timeouts test by
reporting "lock timeout" when the statement timeout was expected to fire
first. The cause seems to be that the process is allowed to sleep longer
than expected (probably due to heavy machine load) so that the lock
timeout happens before we reach the point of reporting the error, and
then this arbitrary tiebreak rule does the wrong thing. We can improve
matters by comparing the scheduled timeout times to decide which error
to report.
I had originally proposed greatly reducing the 1-second window between
the two timeouts in the test cases. On reflection that is a bad idea,
at least for the case where the lock timeout is expected to fire first,
because that would assume that it takes negligible time to get from
statement start to the beginning of the lock wait. Thus, this patch
doesn't completely remove the risk of test failures on slow machines.
Empirically, however, the case this handles is the one we are seeing
in the buildfarm. The explanation may be that the other case requires
the scheduler to take the CPU away from a busy process, whereas the
case fixed here only requires the scheduler to not give the CPU back
right away to a process that has been woken from a multi-second sleep
(and, perhaps, has been swapped out meanwhile).
Back-patch to 9.3 where the isolationtester timeouts test was added.
Discussion: <8693.1464314819@sss.pgh.pa.us>
As part of upper planner pathification (commit 3fc6e2d7f5) I redid
createplan.c's approach to the physical-tlist optimization, in which scan
nodes are allowed to return exactly the underlying table's columns so as
to save doing a projection step at runtime. The logic was intentionally
more aggressive than before about applying the optimization, which is
generally a good thing, but Andres Freund found a case in which it got
too aggressive. Namely, if any column is referenced more than once in
the parent plan node's sorting or grouping column list, we can't optimize
because then that column would need to have more than one ressortgroupref
label, and we only have space for one.
Add logic to detect this situation in use_physical_tlist(), and also add
some error checking in apply_pathtarget_labeling_to_tlist(), which this
example proves was being overly cavalier about whether what it was doing
made any sense.
The added test case exposes the problem only because we do not eliminate
duplicate grouping keys. That might be something to fix someday, but it
doesn't seem like appropriate post-beta work.
Report: <20160526021235.w4nq7k3gnheg7vit@alap3.anarazel.de>
Dating back to commit f10b63923, our grammar has allowed "USING" to
optionally appear before an opclass name in CREATE INDEX (and, lately,
some related places such as ON CONFLICT specifications). Nikolay Shaplov
noticed that this syntax existed but wasn't documented, and proposed
documenting it. But what seems like a better idea is to remove the
production, thereby making the code match the docs not vice versa.
This isn't our usual modus operandi for such cases, but there are a
couple of good reasons to proceed this way:
* So far as I can find, this syntax has never been documented anywhere.
It isn't relied on by any of our own code or test cases, and there seems
little reason to suppose that it's been used in the wild either.
* Documenting it would mean that there would be two separate uses of
USING in the CREATE INDEX syntax, the other being "USING access_method".
That can lead to nothing but confusion.
So, let's just remove it. On the off chance that somebody somewhere
is using it, this isn't something to back-patch, but we can fix it
in HEAD.
Discussion: <1593237.l7oKHRpxSe@nataraj-amd64>
Ever since we split the statistics collector's reports into per-database
files (commit 187492b6c2), backends have been seeing stale statistics
for shared catalogs. This is because the inquiry message only prompts the
collector to write the per-database file for the requesting backend's own
database. Stats for shared catalogs are in a separate file for "DB 0",
which didn't get updated.
In normal operation this was partially masked by the fact that the
autovacuum launcher would send an inquiry message at least once per
autovacuum_naptime that asked for "DB 0"; so the shared-catalog stats would
never be more than a minute out of date. However the problem becomes very
obvious with autovacuum disabled, as reported by Peter Eisentraut.
To fix, redefine the semantics of inquiry messages so that both the
specified DB and DB 0 will be dumped. (This might seem a bit inefficient,
but we have no good way to know whether a backend's transaction will look
at shared-catalog stats, so we have to read both groups of stats whenever
we request stats. Sending two inquiry messages would definitely not be
better.)
Back-patch to 9.3 where the bug was introduced.
Report: <56AD41AC.1030509@gmx.net>
Because vac_update_datfrozenxid() updates datfrozenxid and datminmxid
in-place, it's unsafe to assume that successive reads of those values will
give consistent results. Fetch each one just once to ensure sane behavior
in the minimum calculation. Noted while reviewing Alexander Korotkov's
patch in the same area.
Discussion: <8564.1464116473@sss.pgh.pa.us>
vac_truncate_clog() uses its own transaction ID as the comparison point in
a sanity check that no database's datfrozenxid has already wrapped around
"into the future". That was probably fine when written, but in a lazy
vacuum we won't have assigned an XID, so calling GetCurrentTransactionId()
causes an XID to be assigned when otherwise one would not be. Most of the
time that's not a big problem ... but if we are hard up against the
wraparound limit, consuming XIDs during antiwraparound vacuums is a very
bad thing.
Instead, use ReadNewTransactionId(), which not only avoids this problem
but is in itself a better comparison point to test whether wraparound
has already occurred.
Report and patch by Alexander Korotkov. Back-patch to all versions.
Report: <CAPpHfdspOkmiQsxh-UZw2chM6dRMwXAJGEmmbmqYR=yvM7-s6A@mail.gmail.com>
Commit 1aba62ec moved the range check of that option form guc.c into
bufmgr.c, but introduced a bug by changing a >= 0.0 to > 0.0, which made
the value 0 no longer accepted. Put it back.
Reported by Jeff Janes, diagnosed by Tom Lane
Commit 65c5fcd353 broke this by removing a
header include directive that is conditionally required. Add that back
to nbtree.c, with annotation to keep pgrminclude from re-breaking it.
Peter Geoghegan
Report: <CAM3SWZTNjHFYW_UG8bu0BnogqQ2HfsTgkzXLueuUhfTcYbu5HA@mail.gmail.com>
Needed for cases in which INSERT ... ON CONFLICT appears inside a
recursive CTE item. Per bug #14153 from Thomas Alton.
Patch by Peter Geoghegan, slightly adjusted by me
Report: <20160521232802.22598.13537@wrigleys.postgresql.org>
If RAW_EXPRESSION_COVERAGE_TEST is defined, do a no-op tree walk over
every basic DML statement submitted to parse analysis. If we'd had this
in place earlier, bug #14153 would have been caught by buildfarm testing.
The difficulty is that raw_expression_tree_walker() is only used in
limited cases involving CTEs (particularly recursive ones), so it's
very easy for an oversight in it to not be noticed during testing of a
seemingly-unrelated feature.
The type of error we can expect to catch with this is complete omission
of a node type from raw_expression_tree_walker(), and perhaps also
recursion into a field that doesn't contain a node tree, though that
would be an unlikely mistake. It won't catch failure to add new fields
that need to be recursed into, unfortunately.
I'll go enable this on one or two of my own buildfarm animals once
bug #14153 is dealt with.
Discussion: <27861.1464040417@sss.pgh.pa.us>
do_text_output_multiline() would fail (typically with a null pointer
dereference crash) if its input string did not end with a newline. Such
cases do not arise in our current sources; but it certainly could happen
in future, or in extension code's usage of the function, so we should fix
it. To fix, replace "eol += len" with "eol = text + len".
While at it, make two cosmetic improvements: mark the input string const,
and rename the argument from "text" to "txt" to dodge pgindent strangeness
(since "text" is a typedef name).
Even though this problem is only latent at present, it seems like a good
idea to back-patch the fix, since it's a very simple/safe patch and it's
not out of the realm of possibility that we might in future back-patch
something that expects sane behavior from do_text_output_multiline().
Per report from Hao Lee.
Report: <CAGoxFiFPAGyPAJLcFxTB5cGhTW2yOVBDYeqDugYwV4dEd1L_Ag@mail.gmail.com>
Page image should be MAXALIGN'ed because existing code could directly align
pointers in page instead of align offset from beginning of page.
Found during play with indexes as extenstion, Alexander Korotkov and me
Some comments mentioned XLogReplayBuffer, but there's no such function:
that was an interim name for a function that got renamed to
XLogReadBufferForRedo, before commit 2c03216d83 was pushed.
While it could be argued that rejecting system column mentions in the
ON CONFLICT list is an unsupported feature, falling over altogether
just because the table has a unique index on OID is indubitably a bug.
As far as I can tell, fixing infer_arbiter_indexes() is sufficient to
make ON CONFLICT (oid) actually work, though making a regression test
for that case is problematic because of the impossibility of setting
the OID counter to a known value.
Minor cosmetic cleanups along with the bug fix.
subquery_planner() failed to apply expression preprocessing to the
arbiterElems and arbiterWhere fields of an OnConflictExpr. No doubt the
theory was that this wasn't necessary because we don't actually try to
execute those expressions; but that's wrong, because it results in failure
to match to index expressions or index predicates that are changed at all
by preprocessing. Per bug #14132 from Reynold Smith.
Also add pullup_replace_vars processing for onConflictWhere. Perhaps
it's impossible to have a subquery reference there, but I'm not exactly
convinced; and even if true today it's a failure waiting to happen.
Also add some comments to other places where one or another field of
OnConflictExpr is intentionally ignored, with explanation as to why it's
okay to do so.
Also, catalog/dependency.c failed to record any dependency on the named
constraint in ON CONFLICT ON CONSTRAINT, allowing such a constraint to
be dropped while rules exist that depend on it, and allowing pg_dump to
dump such a rule before the constraint it refers to. The normal execution
path managed to error out reasonably for a dangling constraint reference,
but ruleutils.c dumped core; so in addition to fixing the omission, add
a protective check in ruleutils.c, since we can't retroactively add a
dependency in existing databases.
Back-patch to 9.5 where this code was introduced.
Report: <20160510190350.2608.48667@wrigleys.postgresql.org>
The table-skipping logic in autovacuum would fail to consider that
multiple workers could be processing the same shared catalog in
different databases. This normally wouldn't be a problem: firstly
because autovacuum workers not for wraparound would simply ignore tables
in which they cannot acquire lock, and secondly because most of the time
these tables are small enough that even if multiple for-wraparound
workers are stuck in the same catalog, they would be over pretty
quickly. But in cases where the catalogs are severely bloated it could
become a problem.
Backpatch all the way back, because the problem has been there since the
beginning.
Reported by Ondřej Světlík
Discussion: https://www.postgresql.org/message-id/572B63B1.3030603%40flexibee.euhttps://www.postgresql.org/message-id/572A1072.5080308%40flexibee.eu
This patch essentially reverts commit 4c6780fd17, in favor of a much
simpler solution for the case where the new cluster would choose to create
a TOAST table but the old cluster doesn't have one: just don't create a
TOAST table.
The existing code failed in at least two different ways if the situation
arose: (1) ALTER TABLE RESET didn't grab an exclusive lock, so that the
lock sanity check in create_toast_table failed; (2) pg_upgrade did not
provide a pg_type OID for the new toast table, so that the crosscheck in
TypeCreate failed. While both these problems were introduced by later
patches, they show that the hack being used to cause TOAST table creation
is overwhelmingly fragile (and untested). I also note that before the
TypeCreate crosscheck was added, the code would have resulted in assigning
an indeterminate pg_type OID to the toast table, possibly causing a later
OID conflict in that catalog; so that it didn't really work even when
committed.
If we simply don't create a TOAST table, there will only be a problem if
the code tries to store a tuple that's wider than a page, and field
compression isn't sufficient to get it under a page. Given that the TOAST
creation threshold is intended to be about a quarter of a page, it's very
hard to believe that cross-version differences in the do-we-need-a-toast-
table heuristic could result in an observable problem. So let's just
follow the old version's conclusion about whether a TOAST table is needed.
(If we ever do change needs_toast_table() so much that this conclusion
doesn't apply, we can devise a solution at that time, and hopefully do
it in a less klugy way than 4c6780fd17 did.)
Back-patch to 9.3, like the previous patch.
Discussion: <8110.1462291671@sss.pgh.pa.us>
Limit maintenance of time to xid mapping to once per minute. At
least in the tested case this brings performance within 5% of when
the feature is off, compared to several times slower without this
patch.
While there, fix comments and whitespace.
Ants Aasma, with cosmetic adjustments suggested by Andres Freund
Reviewed by Kevin Grittner and Andres Freund
Discussion is still underway as to whether to revert the entire patch
that added this function, but that discussion may not conclude before
beta1. So, in the meantime, let's do at least this much.
David Rowley
This new limit affects both the max_parallel_degree GUC and the
parallel_degree reloption. There may some day be a use case for using
more than 1024 CPUs for a single query, but that's surely not the case
right now. Not only do not very many people have that many CPUs, but
the code hasn't been tested at that kind of scale and is very unlikely
to perform well, or even work at all, without a lot more work. The
issue addressed by commit 06bd458cb8 is
probably just one problem of many.
The idea of a more reasonable limit here was suggested by Tom Lane;
the value of 1024 was suggested by Amit Kapila.
That way, if the result overflows size_t, you'll get an error instead
of undefined behavior, which seems like a plus. This also has the
effect of casting the number of workers from int to Size, which is
better because it's harder to overflow int than size_t.
Dilip Kumar reported this issue and provided a patch upon which this
patch is based, but his version did use mul_size.
Default roles really should be like regular roles, for the most part.
This removes a number of checks that were trying to make default roles
extra special by not allowing them to be used as regular roles.
We still prevent users from creating roles in the "pg_" namespace or
from altering roles which exist in that namespace via ALTER ROLE, as
we can't preserve such changes, but otherwise the roles are very much
like regular roles.
Based on discussion with Robert and Tom.
to_timestamp() handles the TH/th format codes by advancing over two input
characters, whatever those are. It failed to notice whether there were
two characters available to be skipped, making it possible to advance
the pointer past the end of the input string and keep on parsing.
A similar risk existed in the handling of "Y,YYY" format: it would advance
over three characters after the "," whether or not three characters were
available.
In principle this might be exploitable to disclose contents of server
memory. But the security team concluded that it would be very hard to use
that way, because the parsing loop would stop upon hitting any zero byte,
and TH/th format codes can't be consecutive --- they have to follow some
other format code, which would have to match whatever data is there.
So it seems impractical to examine memory very much beyond the end of the
input string via this bug; and the input string will always be in local
memory not in disk buffers, making it unlikely that anything very
interesting is close to it in a predictable way. So this doesn't quite
rise to the level of needing a CVE.
Thanks to Wolf Roediger for reporting this bug.
Hash indexes are not WAL-logged, and so do not maintain the LSN of
index pages. Since the "snapshot too old" feature counts on
detecting error conditions using the LSN of a table and all indexes
on it, this makes it impossible to safely do early vacuuming on any
table with a hash index, so add this to the tests for whether the
xid used to vacuum a table can be adjusted based on
old_snapshot_threshold.
While at it, add a paragraph to the docs for old_snapshot_threshold
which specifically mentions this and other aspects of the feature
which may otherwise surprise users.
Problem reported and patch reviewed by Amit Kapila
The similarity of the original names to SQL keywords seems like a bad
idea. Rename them before we're stuck with 'em forever.
In passing, minor code and docs cleanup.
Discussion: <4875.1462210058@sss.pgh.pa.us>
Commit 7d9a4737c2 greatly improved the
accuracy of the numeric transcendental functions, however it failed to
consider the case where the result from pow() is close to the overflow
threshold, for example 0.12 ^ -2345.6. For such inputs, where the
result has more than 2000 digits before the decimal point, the decimal
result weight estimate was being clamped to 2000, leading to a loss of
precision in the final calculation.
Fix this by replacing the clamping code with an overflow test that
aborts the calculation early if the final result is sure to overflow,
based on the overflow limit in exp_var(). This provides the same
protection against integer overflow in the subsequent result scale
computation as the original clamping code, but it also ensures that
precision is never lost and saves compute cycles in cases that are
sure to overflow.
The new early overflow test works with the initial low-precision
result (expected to be accurate to around 8 significant digits) and
includes a small fuzz factor to ensure that it doesn't kick in for
values that would not overflow exp_var(), so the overall overflow
threshold of pow() is unchanged and consistent for all inputs with
non-integer exponents.
Author: Dean Rasheed
Reviewed-by: Tom Lane
Discussion: http://www.postgresql.org/message-id/CAEZATCUj3U-cQj0jjoia=qgs0SjE3auroxh8swvNKvZWUqegrg@mail.gmail.com
See-also: http://www.postgresql.org/message-id/CAEZATCV7w+8iB=07dJ8Q0zihXQT1semcQuTeK+4_rogC_zq5Hw@mail.gmail.com
This reverts commits f07d18b6e9, 82c83b3372, 3a3b309041, and
24c5f1a103.
This feature has shown enough immaturity that it was deemed better to
rip it out before rushing some more fixes at the last minute. There are
discussions on larger changes in this area for the next release.
Unfortunately the segment size checks from 72a98a6395 had the negative
side-effect of breaking a corner case in mdsync(): When processing a
fsync request for a truncated away segment mdsync() could fail with
"could not fsync file" (if previous segment < RELSEG_SIZE) because
_mdfd_getseg() now wouldn't return the relevant segment anymore.
The cleanest fix seems to be to allow the caller of _mdfd_getseg() to
specify whether checks for RELSEG_SIZE are performed. To allow doing so,
change the ExtensionBehavior enum into a bitmask. Besides allowing for
the addition of EXTENSION_DONT_CHECK_SIZE, this makes for a nicer
implementation of EXTENSION_REALLY_RETURN_NULL.
Besides mdsync() the only callsite that should change behaviour due to
this is mdprefetch() which now doesn't create segments anymore, even in
recovery. Given the uses of mdprefetch() that seems better.
Reported-By: Thom Brown
Discussion: CAA-aLv72QazLvPdKZYpVn4a_Eh+i4_cxuB03k+iCuZM_xjc+6Q@mail.gmail.com
Conversion functions were previously marked as parallel-unsafe, since
that is the default, but in fact they are safe. Parallel-safe
functions defined in pg_proc.h and redefined in system_views.sql were
ending up as parallel-unsafe because the redeclarations were not
marked PARALLEL SAFE. While editing system_views.sql, mark ts_debug()
parallel safe also.
Andreas Karlsson
These adjustments adjust code and comments in minor ways to prevent
pgindent from mangling them. Among other things, I tried to avoid
situations where pgindent would emit "a +b" instead of "a + b", and I
tried to avoid having it break up inline comments across multiple
lines.
Back in 3b02ea4f07 I added some comments in various places to explain
how logical decoding and other things worked. Not all of the changes
were welcome, because they were misleading or wrong. This changes them
a little bit to make them more accurate.
Some other comments are also changed to be more accurate. Also, fix a
bunch of typos.
Author: Álvaro Herrera, Craig Ringer
Andres Freund reviewed some parts of this.
Commit 7117685461 made pg_start_backup
parallel-restricted rather than parallel-safe, because it now relies
on backend-private state that won't be synchronized with the parallel
worker. However, it didn't update pg_proc.h. Separately, Andreas
Karlsson observed that system_views.sql neglected to reiterate the
parallel-safety markings whe redefining various functions, including
this one; so add a PARALLEL RESTRICTED declaration there to match
the new value in pg_proc.h.
Now that Paths have their own rows field, print that rather than
the parent relation's rowcount.
Show the relid sets associated with Paths using table names rather
than numbers; since this code is able to print simple Var references
using table names, it seems a bit silly that print_relids can't.
Print the cheapest_parameterized_paths list for a RelOptInfo, and
include information about a parameterized path's required_outer rels.
Noted while trying to use this feature to debug Alexander Kirkouski's
recent bug report.
We mustn't run generate_gather_paths() during add_paths_to_joinrel(),
because that function can be invoked multiple times for the same target
joinrel. Not only is it wasteful to build GatherPaths repeatedly, but
a later add_partial_path() could delete the partial path that a previously
created GatherPath depends on. Instead establish the convention that we
do generate_gather_paths() for a rel only just before set_cheapest().
The code was accidentally not broken for baserels, because as of today there
never is more than one partial path for a baserel. But that assumption
obviously has a pretty short half-life, so move the generate_gather_paths()
calls for those cases as well.
Also add some generic comments explaining how and why this all works.
Per fuzz testing by Andreas Seltenreich.
Report: <871t5pgwdt.fsf@credativ.de>
If we're not going to reject such setups entirely, throwing a WARNING in
check_synchronous_standby_names() is unhelpful, because it will cause the
warning to be logged again every time the postmaster receives SIGHUP.
Per discussion, just remove the warning.
In passing, improve the documentation for synchronous_commit, which had not
gotten the word that now there can be more than one synchronous standby.
Given a three-or-more-way equivalence class, such as X.Y = Y.Y = Z.Z,
it was possible for the planner to omit one of the quals needed to
enforce that all members of the equivalence class are actually equal.
This only happened in the case of a parameterized join node for two
of the relations, that is a plan tree like
Nested Loop
-> Scan X
-> Nested Loop
-> Scan Y
-> Scan Z
Filter: Z.Z = X.X
The eclass machinery normally expects to apply X.X = Y.Y when those
two relations are joined, but in this shape of plan tree they aren't
joined until the top node --- and, if the lower nested loop is marked
as parameterized by X, the top node will assume that the relevant eclass
condition(s) got pushed down into the lower node. On the other hand,
the scan of Z assumes that it's only responsible for constraining Z.Z
to match any one of the other eclass members. So one or another of
the required quals sometimes fell between the cracks, depending on
whether consideration of the eclass in get_joinrel_parampathinfo()
for the lower nested loop chanced to generate X.X = Y.Y or X.X = Z.Z
as the appropriate constraint there. If it generated the latter,
it'd erroneously suppose that the Z scan would take care of matters.
To fix, force X.X = Y.Y to be generated and applied at that join node
when this case occurs.
This is *extremely* hard to hit in practice, because various planner
behaviors conspire to mask the problem; starting with the fact that the
planner doesn't really like to generate a parameterized plan of the
above shape. (It might have been impossible to hit it before we
tweaked things to allow this plan shape for star-schema cases.) Many
thanks to Alexander Kirkouski for submitting a reproducible test case.
The bug can be demonstrated in all branches back to 9.2 where parameterized
paths were introduced, so back-patch that far.
Without a few entries beyond old_snapshot_threshold, the lookup
would often fail, resulting in the more aggressive pruning or
vacuum being skipped often enough to matter. This was very clearly
shown by a python test script posted by Ants Aasma, and was likely
a factor in an earlier but somewhat less clear-cut test case posted
by Jeff Janes.
This patch makes no change to the logic, per se -- it just makes
the array of mapping entries big enough to make lookup misses based
on timing much less likely. An occasional miss is still possible
if a thread stalls for more than 10 minutes, but that does not
create any problem with correctness of behavior. Besides, if
things are so busy that a thread is stalling for more than 10
minutes, it is probably OK to skip the more aggressive cleanup at
that particular point in time.
Adjust the way we detect the locale. As a result the minumum Windows
version supported by VS2015 and later is Windows Vista. Add some tweaks
to remove new compiler warnings. Remove documentation references to the
now obsolete msysGit.
Michael Paquier, somewhat edited by me, reviewed by Christian Ullrich.
Backpatch to 9.5
Since 5a991ef8 we're explicitly asking for feedback from the receiving
side when shutting down walsender, if there's not yet replicated
data.
Unfortunately we didn't remember (i.e. set waiting_for_ping_response to
true) having asked for feedback, leading to scenarios in which replies
were requested at a high frequency.
I can't reproduce this problem on my laptop, I think that's because the
problem requires a significant TCP window to manifest due to the
!pq_is_send_pending() condition. But since this clearly is a bug, let's
fix it. There's quite possibly more wrong than just this though.
While fiddling with WalSndDone(), I rewrote a hard to understand comment
about looking at the flush vs. the write position.
Reported-By: Nick Cleaton, Magnus Hagander
Author: Nick Cleaton
Discussion: CAFgz3kus=rC_avEgBV=+hRK5HYJ8vXskJRh8yEAbahJGTzF2VQ@mail.gmail.comCABUevExsjROqDcD0A2rnJ6HK6FuKGyewJr3PL12pw85BHFGS2Q@mail.gmail.com
Backpatch: 9.4, were 5a991ef8 introduced the use of feedback messages
during shutdown.
Previously, ginInsertCleanup could exit early if it detects that someone else
is cleaning up the pending list, without waiting for that someone else to
finish the job. But in this case vacuum could miss tuples to be deleted.
Cleanup process now locks metapage with a help of heavyweight
LockPage(ExclusiveLock), and it guarantees that there is no another cleanup
process at the same time. Lock is taken differently depending on caller of
cleanup process: any vacuums and gin_clean_pending_list() will be blocked
until lock becomes available, ordinary insert uses conditional lock to
prevent indefinite waiting on lock.
Insert into pending list doesn't use this lock, so insertion isn't blocked.
Also, patch adds stopping of cleanup process when at-start-cleanup-tail is
reached in order to prevent infinite cleanup in case of massive insertion. But
it will stop only for automatic maintenance tasks like autovacuum.
Patch introduces choice of limit of memory to use: autovacuum_work_mem,
maintenance_work_mem or work_mem depending on call path.
Patch for previous releases should be reworked due to changes between 9.6 and
previous ones in this area.
Discover and diagnostics by Jeff Janes and Tomas Vondra
Patch by me with some ideas of Jeff Janes
Commit 989be0810d added a flex/bison lexer/parser to interpret
synchronous_standby_names. It was done in a pretty crufty way, though,
making assorted end-use sites responsible for calling the parser at the
right times. That was not only vulnerable to errors of omission, but made
it possible that lexer/parser errors occur at very undesirable times,
and created memory leakages even if there was no error.
Instead, perform the parsing once during check_synchronous_standby_names
and let guc.c manage the resulting data. To do that, we have to flatten
the parsed representation into a single hunk of malloc'd memory, but that
is not very hard.
While at it, work a little harder on making useful error reports for
parsing problems; the previous code felt that "synchronous_standby_names
parser returned 1" was an appropriate user-facing error message. (To
be fair, it did also log a syntax error message, but separately from the
GUC problem report, which is at best confusing.) It had some outright
bugs in the face of invalid input, too.
I (tgl) also concluded that we need to restrict unquoted names in
synchronous_standby_names to be just SQL identifiers. The previous coding
would accept darn near anything, which (1) makes the quoting convention
both nearly-unnecessary and formally ambiguous, (2) makes it very hard to
understand what is a syntax error and what is a creative interpretation of
the input as a standby name, and (3) makes it impossible to further extend
the syntax in future without a compatibility break. I presume that we're
intending future extensions of the syntax, else this parsing infrastructure
is massive overkill, so (3) is an important objection. Since we've taken
a compatibility hit for non-identifier names with this change anyway, we
might as well lock things down now and insist that users use double quotes
for standby names that aren't identifiers.
Kyotaro Horiguchi and Tom Lane
In nodeFuncs.c, pgindent wants to introduce spurious indentation into
the definitions of planstate_tree_walker and planstate_walk_subplans.
Fix that by spreading the definition out across several lines, similar
to what is already done for other walker functions in that file.
In execParallel.c, in the definition of SharedExecutorInstrumentation,
pgindent wants to insert more whitespace between the type name and the
member name. That causes it to mangle comments later on the line. Fix
by moving the comments out of line. Now that we have a bit more room,
add some more details that may be useful to the next person reading
this code.
The way that PartialAggregate and FinalizeAggregate plan nodes were
displaying output columns before was bogus. Now, FinalizeAggregate
produces the same outputs as an Aggregate would have produced, while
PartialAggregate produces each of those outputs prefixed by the word
PARTIAL.
Discussion: 12585.1460737650@sss.pgh.pa.us
Patch by me, reviewed by David Rowley.
Before this commit _mdfd_getseg(), in contrast to mdnblocks(), did not
verify whether all segments leading up to the to-be-opened one, were
RELSEG_SIZE sized. That is e.g. not the case after truncating a
relation, because later segments just get truncated to zero length, not
removed.
Once a "non-existent" segment has been opened in a session, mdnblocks()
will return wrong results, causing errors like "could not read block %u
in file" when accessing blocks. Closing the session, or the later
arrival of relevant invalidation messages, would "fix" the problem.
That, so far, was mostly harmless, because most segment accesses are
only done after an mdnblocks() call. But since 428b1d6b29 we try to
open segments that might have been deleted, to trigger kernel writeback
from a backend's queue of recent writes.
To fix check segment sizes in _mdfd_getseg() when opening previously
unopened segments. In practice this shouldn't imply a lot of additional
lseek() calls, because mdnblocks() will most of the time already have
opened all relevant segments.
This commit also fixes a second problem, namely that _mdfd_getseg(
EXTENSION_RETURN_NULL) extends files during recovery, which is not
desirable for the mdwriteback() case. Add EXTENSION_REALLY_RETURN_NULL,
which does not behave that way, and use it.
Reported-By: Thom Brown
Author: Andres Freund, Abhijit Menon-Sen
Reviewd-By: Robert Haas, Fabien Coehlo
Discussion: CAA-aLv6Dp_ZsV-44QA-2zgkqWKQq=GedBX2dRSrWpxqovXK=Pg@mail.gmail.com
Fixes: 428b1d6b29
So far, when a transaction with pending invalidations, but without an
assigned xid, committed, we simply ignored those invalidation
messages. That's problematic, because those are actually sent for a
reason.
Known symptoms of this include that existing sessions on a hot-standby
replica sometimes fail to notice new concurrently built indexes and
visibility map updates.
The solution is to WAL log such invalidations in transactions without an
xid. We considered to alternatively force-assign an xid, but that'd be
problematic for vacuum, which might be run in systems with few xids.
Important: This adds a new WAL record, but as the patch has to be
back-patched, we can't bump the WAL page magic. This means that standbys
have to be updated before primaries; otherwise
"PANIC: standby_redo: unknown op code 32" errors can be encountered.
XXX:
Reported-By: Васильев Дмитрий, Masahiko Sawada
Discussion:
CAB-SwXY6oH=9twBkXJtgR4UC1NqT-vpYAtxCseME62ADwyK5OA@mail.gmail.comCAD21AoDpZ6Xjg=gFrGPnSn4oTRRcwK1EBrWCq9OqOHuAcMMC=w@mail.gmail.com
The true explanation for Peter Eisentraut's report of inexact asind results
seems to be that (a) he's compiling into x87 instruction set, which uses
wider-than-double float registers, plus (b) the library function asin() on
his platform returns a result that is wider than double and is not rounded
to double width. To fix, we have to force the function's result to be
rounded comparably to what happened to the scaling constant asin_0_5.
Experimentation suggests that storing it into a volatile local variable is
the least ugly way of making that happen. Although only asin() is known to
exhibit an observable inexact result, we'd better do this in all the places
where we're hoping to get an exact result by scaling.
Change max_parallel_degree default from 0 to 2. It is possible that
this is not a good idea, or that we should go with 1 worker rather
than 2, but we won't find out without trying it. Along the way,
reword the documentation for max_parallel_degree a little bit to
hopefully make it more clear.
Discussion: 20160420174631.3qjjhpwsvvx5bau5@alap3.anarazel.de
Commit 65abaab547 tried to prevent the scaling constants used in
the degree-based trig functions from being precomputed at compile time,
because some compilers do that with functions that don't yield results
identical-to-the-last-bit to what you get at runtime. A report from
Peter Eisentraut suggests that some recent compilers are smart enough
to see through that trick, though. Instead, let's put the inputs to
these calculations into non-const global variables, which should be a
more reliable way of convincing the compiler that it can't assume that
they are compile-time constants. (If we really get desperate, we could
mark these variables "volatile", but I do not believe we should have to.)
Several issues:
1) checkpoint_flush_after doc and code disagreed about the default
2) new GUCs were missing from postgresql.conf.sample
3) Outdated source-code comment about bgwriter_flush_after's default
4) Sub-optimal categories assigned to new GUCs
5) Docs suggested backend_flush_after is PGC_SIGHUP, but it's PGC_USERSET.
6) Spell out int as integer in the docs, as done elsewhere
Reported-By: Magnus Hagander, Fujii Masao
Discussion: CAHGQGwETyTG5VYQQ5C_srwxWX7RXvFcD3dKROhvAWWhoSBdmZw@mail.gmail.com
NetBSD has seen fit to invent a libc function named strtoi(), which
conflicts with the long-established static functions of the same name in
datetime.c and ecpg's interval.c. While muttering darkly about intrusions
on application namespace, we'll rename our functions to avoid the conflict.
Back-patch to all supported branches, since this would affect attempts
to build any of them on recent NetBSD.
Thomas Munro
The implementation of that feature involves injecting nodes into the
raw parsetree where explicit parentheses appear. Various places in
parse_expr.c that test to see "is this child node of type Foo" need to
look through such nodes, else we'll get different behavior when
operator_precedence_warning is on than when it is off. Note that we only
need to handle this when testing untransformed child nodes, since the
AEXPR_PAREN nodes will be gone anyway after transformExprRecurse.
Per report from Scott Ribe and additional code-reading. Back-patch
to 9.5 where this feature was added.
Report: <ED37E303-1B0A-4CD8-8E1E-B9C4C2DD9A17@elevated-dev.com>
Given a left join containing a full join in its righthand side, with
the left join's joinclause referencing only one side of the full join
(in a non-strict fashion, so that the full join doesn't get simplified),
the planner could fail with "failed to build any N-way joins" or related
errors. This happened because the full join was seen as overlapping the
left join's RHS, and then recent changes within join_is_legal() caused
that function to conclude that the full join couldn't validly be formed.
Rather than try to rejigger join_is_legal() yet more to allow this,
I think it's better to fix initsplan.c so that the required join order
is explicit in the SpecialJoinInfo data structure. The previous coding
there essentially ignored full joins, relying on the fact that we don't
flatten them in the joinlist data structure to preserve their ordering.
That's sufficient to prevent a wrong plan from being formed, but as this
example shows, it's not sufficient to ensure that the right plan will
be formed. We need to work a bit harder to ensure that the right plan
looks sane according to the SpecialJoinInfos.
Per bug #14105 from Vojtech Rylko. This was apparently induced by
commit 8703059c6 (though now that I've seen it, I wonder whether there
are related cases that could have failed before that); so back-patch
to all active branches. Unfortunately, that patch also went into 9.0,
so this bug is a regression that won't be fixed in that branch.
The coverage was rather lean for cases that bind() or listen() might
return. Add entries for everything that there's a direct equivalent
for in the set of Unix errnos that elog.c has heard of.
When we shoehorned "x op ANY (array)" into the SQL syntax, we created a
fundamental ambiguity as to the proper treatment of a sub-SELECT on the
righthand side: perhaps what's meant is to compare x against each row of
the sub-SELECT's result, or perhaps the sub-SELECT is meant as a scalar
sub-SELECT that delivers a single array value whose members should be
compared against x. The grammar resolves it as the former case whenever
the RHS is a select_with_parens, making the latter case hard to reach ---
but you can get at it, with tricks such as attaching a no-op cast to the
sub-SELECT. Parse analysis would throw away the no-op cast, leaving a
parsetree with an EXPR_SUBLINK SubLink directly under a ScalarArrayOpExpr.
ruleutils.c was not clued in on this fine point, and would naively emit
"x op ANY ((SELECT ...))", which would be parsed as the first alternative,
typically leading to errors like "operator does not exist: text = text[]"
during dump/reload of a view or rule containing such a construct. To fix,
emit a no-op cast when dumping such a parsetree. This might well be
exactly what the user wrote to get the construct accepted in the first
place; and even if she got there with some other dodge, it is a valid
representation of the parsetree.
Per report from Karl Czajkowski. He mentioned only a case involving
RLS policies, but actually the problem is very old, so back-patch to
all supported branches.
Report: <20160421001832.GB7976@moraine.isi.edu>
Also, avoid reading PGPROC's wait_event field twice, once for the wait
event and again for the wait_event_type, because the value might change
in the middle.
Petr Jelinek and Robert Haas
That commit increased all shared memory allocations to the next higher
multiple of PG_CACHE_LINE_SIZE, but it didn't ensure that allocation
started on a cache line boundary. It also failed to remove a couple
other pieces of now-useless code.
BUFFERALIGN() is perhaps obsolete at this point, and likely should be
removed at some point, too, but that seems like it can be left to a
future cleanup.
Mistakes all pointed out by Andres Freund. The patch is mine, with
a few extra assertions which I adopted from his version of this fix.
Even with old_snapshot_threshold = -1 (which disables the "snapshot
too old" feature), performance regressions were seen at moderate to
high concurrency. For example, a one-socket, four-core system
running 200 connections at saturation could see up to a 2.3%
regression, with larger regressions possible on NUMA machines.
By inlining the early (smaller, faster) tests in the
TestForOldSnapshot() function, the i7 case dropped to a 0.2%
regression, which could easily just be noise, and is clearly an
improvement. Further testing will show whether more is needed.
Commit 36a35c550a turned the interface between ginPlaceToPage and
its subroutines in gindatapage.c and ginentrypage.c into a royal mess:
page-update critical sections were started in one place and finished in
another place not even in the same file, and the very same subroutine
might return having started a critical section or not. Subsequent patches
band-aided over some of the problems with this design by making things
even messier.
One user-visible resulting problem is memory leaks caused by the need for
the subroutines to allocate storage that would survive until ginPlaceToPage
calls XLogInsert (as reported by Julien Rouhaud). This would not typically
be noticeable during retail index updates. It could be visible in a GIN
index build, in the form of memory consumption swelling to several times
the commanded maintenance_work_mem.
Another rather nasty problem is that in the internal-page-splitting code
path, we would clear the child page's GIN_INCOMPLETE_SPLIT flag well before
entering the critical section that it's supposed to be cleared in; a
failure in between would leave the index in a corrupt state. There were
also assorted coding-rule violations with little immediate consequence but
possible long-term hazards, such as beginning an XLogInsert sequence before
entering a critical section, or calling elog(DEBUG) inside a critical
section.
To fix, redefine the API between ginPlaceToPage() and its subroutines
by splitting the subroutines into two parts. The "beginPlaceToPage"
subroutine does what can be done outside a critical section, including
full computation of the result pages into temporary storage when we're
going to split the target page. The "execPlaceToPage" subroutine is called
within a critical section established by ginPlaceToPage(), and it handles
the actual page update in the non-split code path. The critical section,
as well as the XLOG insertion call sequence, are both now always started
and finished in ginPlaceToPage(). Also, make ginPlaceToPage() create and
work in a short-lived memory context to eliminate the leakage problem.
(Since a short-lived memory context had been getting created in the most
common code path in the subroutines, this shouldn't cause any noticeable
performance penalty; we're just moving the overhead up one call level.)
In passing, fix a bunch of comments that had gone unmaintained throughout
all this klugery.
Report: <571276DD.5050303@dalibo.com>
The reverted changes were intended to force a choice of whether any
newly-added BufferGetPage() calls needed to be accompanied by a
test of the snapshot age, to support the "snapshot too old"
feature. Such an accompanying test is needed in about 7% of the
cases, where the page is being used as part of a scan rather than
positioning for other purposes (such as DML or vacuuming). The
additional effort required for back-patching, and the doubt whether
the intended benefit would really be there, have indicated it is
best just to rely on developers to do the right thing based on
comments and existing usage, as we do with many other conventions.
This change should have little or no effect on generated executable
code.
Motivated by the back-patching pain of Tom Lane and Robert Haas
Coverity complained that oldPartitionLock was possibly dereferenced after
having been set to NULL. That actually can't happen, because we'd only use
it if (oldFlags & BM_TAG_VALID) is true. But nonetheless Coverity is
justified in complaining, because at line 1275 we actually overwrite
oldFlags, and then still expect its BM_TAG_VALID bit to be a safe guide to
whether to release the oldPartitionLock. Thus, the code would be incorrect
if someone else had changed the buffer's BM_TAG_VALID flag meanwhile.
That should not happen, since we hold pin on the buffer throughout this
sequence, but it's starting to look like a rather shaky chain of logic.
And there's no need for such assumptions, because we can simply replace
the (oldFlags & BM_TAG_VALID) tests with (oldPartitionLock != NULL),
which has identical results and makes it plain to all comers that we don't
dereference a null pointer. A small side benefit is that the range of
liveness of oldFlags is greatly reduced, possibly allowing the compiler
to save a register.
This is just cleanup, not an actual bug fix, so there seems no need
for a back-patch.
We've had repeated troubles over the years with failures to initialize
spinlocks correctly; see 6b93fcd14 for a recent example. Most of the time,
on most platforms, such oversights can escape notice because all-zeroes is
the expected initial content of an slock_t variable. The only platform
we have where the initialized state of an slock_t isn't zeroes is HPPA,
and that's practically gone in the wild. To make it easier to catch such
errors without needing one of those, adjust the --disable-spinlocks code
so that zero is not a valid value for an slock_t for it.
In passing, remove a bunch of unnecessary #include's from spin.c;
commit daa7527afc removed all the intermodule coupling that
made them necessary.
Although OID acts pretty much like user data, the other system columns do
not, so an index on one would likely misbehave. And it's pretty hard to
see a use-case for one, anyway. Let's just forbid the case rather than
worry about whether it should be supported.
David Rowley
For reasons of sheer brain fade, we (I) was calling systable_endscan()
immediately after systable_getnext() and expecting the tuple returned
by systable_getnext() to still be valid.
That's clearly wrong. Move the systable_endscan() down below the tuple
usage.
Discovered initially by Pavel Stehule and then also by Alvaro.
Add a regression test based on Alvaro's testing.
Careless coding added by commit 07cacba983 could result in a crash
or a bizarre error message if someone tried to select an index on the
OID column as the replica identity index for a table. Back-patch to 9.4
where the feature was introduced.
Discussion: CAKJS1f8TQYgTRDyF1_u9PVCKWRWz+DkieH=U7954HeHVPJKaKg@mail.gmail.com
David Rowley
The previous display was sort of confusing, because it didn't
distinguish between the number of workers that we planned to launch
and the number that actually got launched. This has already confused
several people, so display both numbers and label them clearly.
Julien Rouhaud, reviewed by me.
pg_xlogdump includes bufmgr.h. With a compiler that emits code for
static inline functions even when they're unreferenced, that leads
to unresolved external references in the new static-inline version
of BufferGetPage(). So hide it with #ifndef FRONTEND, as we've done
for similar issues elsewhere. Per buildfarm member pademelon.
The code had a query-lifespan memory leak when encountering GIN entries
that have posting lists (rather than posting trees, ie, there are a
relatively small number of heap tuples containing this index key value).
With a suitable data distribution this could add up to a lot of leakage.
Problem seems to have been introduced by commit 36a35c550, so back-patch
to 9.4.
Julien Rouhaud
My previous attempt at doing so, in 80abbeba23, was not sufficient. While that
fixed the problem for bufmgr.c and lwlock.c , s_lock.c still has non-constant
expressions in the struct initializer, because the file/line/function
information comes from the caller of s_lock().
Give up on using a macro, and use a static inline instead.
Discussion: 4369.1460435533@sss.pgh.pa.us
When re-reading an update involving both an old tuple and a new tuple from
disk, reorderbuffer.c was careless about whether the new tuple is suitably
aligned for direct access --- in general, it isn't. We'd missed seeing
this in the buildfarm because the contrib/test_decoding tests exercise this
code path only a few times, and by chance all of those cases have old
tuples with length a multiple of 4, which is usually enough to make the
access to the new tuple's t_len safe. For some still-not-entirely-clear
reason, however, Debian's sparc build gets a bus error, as reported by
Christoph Berg; perhaps it's assuming 8-byte alignment of the pointer?
The lack of previous field reports is probably because you need all of
these conditions to trigger a crash: an alignment-picky platform (not
Intel), a transaction large enough to spill to disk, an update within
that xact that changes a primary-key field and has an odd-length old tuple,
and of course logical decoding tracing the transaction.
Avoid the alignment assumption by using memcpy instead of fetching t_len
directly, and add a test case that exposes the crash on picky platforms.
Back-patch to 9.4 where the bug was introduced.
Discussion: <20160413094117.GC21485@msg.credativ.de>
Commit 314cbfc5da redefined the signature of this hook as
typedef int (*walrcv_receive_type) (char **buffer, int *wait_fd);
But in fact the type of the "wait_fd" variable ought to be pgsocket,
which is what WaitLatchOrSocket expects, and which is necessary if
we want to be able to assign PGINVALID_SOCKET to it on Windows.
So fix that.
It was declared as "pid_t", which would be fine except that none of
the places that printed it in error messages took any thought for the
possibility that it's not equivalent to "int". This leads to warnings
on some buildfarm members, and could possibly lead to actually wrong
error messages on those platforms. There doesn't seem to be any very
good reason not to just make it "int"; it's only ever assigned from
MyProcPid, which is int. If we want to cope with PIDs that are wider
than int, this is not the place to start.
Also, fix the comment, which seems to perhaps be a leftover from a time
when the field was only a bool?
Per buildfarm. Back-patch to 9.5 which has same issue.
For a long time, opclasscmds.c explained that "we do not create a
dependency link to the AM [for an opclass or opfamily], because we don't
currently support DROP ACCESS METHOD". Commit 473b932870 invented
DROP ACCESS METHOD, but it batted only 1 for 2 on adding the dependency
links, and 0 for 2 on updating the comments about the topic.
In passing, undo the same commit's entirely inappropriate decision to
blow away an existing index as a side-effect of create_am.sql.
As part of reserving the pg_* namespace for default roles and in line
with SET ROLE and other previous efforts, disallow settings the role
to a default/reserved role using SET SESSION AUTHORIZATION.
These checks and restrictions on what is allowed regarding default /
reserved roles are under debate, but it seems prudent to ensure that
the existing checks at least cover the intended cases while the
debate rages on. On me to clean it up if the consensus decision is
to remove these checks.
Logical messages, added in 3fe3511d05, during decoding failed to filter
messages emitted in other databases and messages emitted "under" a
replication origin the output plugin isn't interested in.
Add tests to verify that both types of filtering actually work. While
touching message.sql remove hunk obsoleted by d25379e.
Bump XLOG_PAGE_MAGIC because xl_logical_message changed and because
3fe3511d05 had omitted doing so. 3fe3511d05 additionally didn't bump
catversion, but 7a542700d has done so since.
Author: Petr Jelinek
Reported-By: Andres Freund
Discussion: 20160406142513.wotqy3ba3kanr423@alap3.anarazel.de
The current definition of init_spin_delay (introduced recently in
48354581a) wasn't C89 compliant. It's not legal to refer to refer to
non-constant expressions, and the ptr argument was one. This, as
reported by Tom, lead to a failure on buildfarm animal pademelon.
The pointer, especially on system systems with ASLR, isn't super helpful
anyway, though. So instead of making init_spin_delay into an inline
function, make s_lock_stuck() report the function name in addition to
file:line and change init_spin_delay() accordingly. While not a direct
replacement, the function name is likely more useful anyway (line
numbers are often hard to interpret in third party reports).
This also fixes what file/line number is reported for waits via
s_lock().
As PG_FUNCNAME_MACRO is now used outside of elog.h, move it to c.h.
Reported-By: Tom Lane
Discussion: 4369.1460435533@sss.pgh.pa.us
The recent patch to make Pin/UnpinBuffer lockfree in the hot
path (48354581a), accidentally used pg_atomic_fetch_or_u32() in
MarkLocalBufferDirty(). Other code operating on local buffers was
careful to only use pg_atomic_read/write_u32 which just read/write from
memory; to avoid unnecessary overhead.
On its own that'd just make MarkLocalBufferDirty() slightly less
efficient, but in addition InitLocalBuffers() doesn't call
pg_atomic_init_u32() - thus the spinlock fallback for the atomic
operations isn't initialized. That in turn caused, as reported by Tom,
buildfarm animal gaur to fail. As those errors are actually useful
against this type of error, continue to omit - intentionally this time -
initialization of the atomic variable.
In addition, add an explicit note about only using pg_atomic_read/write
on local buffers's state to BufferDesc's description.
Reported-By: Tom Lane
Discussion: 1881.1460431476@sss.pgh.pa.us
It's silly to define these counts as narrower than they might someday
need to be. Also, I believe that the BLCKSZ * nflush calculation in
mdwriteback was capable of overflowing an int.
Commit 428b1d6b29 introduced the use of
msync() for flushing dirty data from the kernel's file buffers. Several
portability issues were overlooked, though:
* Not all implementations of mmap() think that nbytes == 0 means "map
the whole file". To fix, use lseek() to find out the true length.
Fix callers of pg_flush_data to be aware that nbytes == 0 may result
in trashing the file's seek position.
* Not all implementations of mmap() will accept partial-page mmap
requests. To fix, round down the length request to whatever sysconf()
says the page size is. (I think this is OK from a portability standpoint,
because sysconf() is required by SUS v2, and we aren't trying to compile
this part on Windows anyway. Buildfarm should let us know if not.)
* On 32-bit machines, the file size might exceed the available free
address space, or even exceed what will fit in size_t. Check for
the latter explicitly to avoid passing a false request size to mmap().
If mmap fails, silently fall through to the next implementation method,
rather than bleating to the postmaster log and giving up.
* mmap'ing directories fails on some platforms, and even if it works,
msync'ing the directory is quite unlikely to help, as for that matter are
the other flush implementations. In pre_sync_fname(), just skip flush
attempts on directories.
In passing, copy-edit the comments a bit.
Stas Kelvich and myself
I've seen one too many "could not bind IPv4 socket: No error" log entries
from the Windows buildfarm members. Per previous discussion, this is
likely caused by the fact that we're doing nothing to translate
WSAGetLastError() to errno. Put in a wrapper layer to do that.
If this works as expected, it should get back-patched, but let's see what
happens in the buildfarm first.
Discussion: <4065.1452450340@sss.pgh.pa.us>
The original patch kind of ignored the fact that we were doing something
different from a costing point of view, but nobody noticed. This patch
fixes that oversight.
David Rowley
That unused function was introduced as a sample because synchronous
replication or replication monitoring tools might need it in the future.
Recently commit 989be08 added the function SyncRepGetOldestSyncRecPtr
which provides almost the same functionality for multiple synchronous
standbys feature. So it's time to remove that unused sample function.
This commit does that.
Per discussion, this gives potential users of the hook more flexibility,
because they can build custom Paths that implement only one stage of
upper processing atop core-provided Paths for earlier stages.
On a big NUMA machine with 1000 connections in saturation load
there was a performance regression due to spinlock contention, for
acquiring values which were never used. Just fill with dummy
values if we're not going to use them.
This patch has not been benchmarked yet on a big NUMA machine, but
it seems like a good idea on general principle, and it seemed to
prevent an apparent 2.2% regression on a single-socket i7 box
running 200 connections at saturation load.
Rename this function to GenericXLogRegisterBuffer() to make it clearer
what it does, and leave room for other sorts of "register" actions in
future. Also, replace its "bool isNew" argument with an integer flags
argument, so as to allow adding more flags in future without an API
break.
Alexander Korotkov, adjusted slightly by me
The previous coding could allow the contents of the "hole" between pd_lower
and pd_upper to diverge during replay from what it had been when the update
was originally applied. This would pose a problem if checksums were in
use, and in any case would complicate forensic comparisons between master
and slave servers. So force the "hole" to contain zeroes, both at initial
application of a generically-logged action, and at replay.
Alexander Korotkov, adjusted slightly by me
When IF NOT EXISTS was added to CREATE TABLE AS, this logic didn't get
the memo, possibly resulting in an Assert failure. It looks like there
would have been no ill effects in a non-Assert build, though. Back-patch
to 9.5 where the IF NOT EXISTS option was added.
Stas Kelvich
I was initially concerned that the some of the hundreds of
references to BufferGetPage() where the literal
BGP_NO_SNAPSHOT_TEST were passed might not optimize as well as a
macro, leading to some hard-to-find performance regressions in
corner cases. Inspection of disassembled code has shown identical
code at all inspected locations, and the size difference doesn't
amount to even one byte per such call. So make it readable.
Per gripes from Álvaro Herrera and Tom Lane
It was incorrectly declared as a volatile pointer to a non-volatile
structure. Eliminate the OldSnapshotControl struct definition; it
is really not needed. Pointed out by Tom Lane.
While at it, add OldSnapshotControlData to pgindent's list of
structures.
The existing code would either Assert or generate an invalid
SyncRepConfig variable, neither of which is desirable. A regular
error should be thrown instead.
This commit silences compiler warning in non assertion-enabled builds.
Per report from Jeff Janes.
Suggested fix by Tom Lane.
Coverity complained about an apparent missing "break" in a switch
added by bb140506df. The human-readable comments are pretty
clear that this is intentional, but add a standard /* FALL THRU */
comment to make it clear to tools too.
Coverity complained that the code added by 015e88942a lacked an
error check for SearchSysCache1 failures, which it should have. But
the code was pretty duff in other ways too, including failure to think
about whether it could really cope with arrays of different lengths.
Previously we used a spinlock, in adition to the atomically manipulated
->state field, to protect the wait queue. But it's pretty simple to
instead perform the locking using a flag in state.
Due to 6150a1b0 BufferDescs, on platforms (like PPC) with > 1 byte
spinlocks, increased their size above 64byte. As 64 bytes are the size
we pad allocated BufferDescs to, this can increase false sharing;
causing performance problems in turn. Together with the previous commit
this reduces the size to <= 64 bytes on all common platforms.
Author: Andres Freund
Discussion: CAA4eK1+ZeB8PMwwktf+3bRS0Pt4Ux6Rs6Aom0uip8c6shJWmyg@mail.gmail.com20160327121858.zrmrjegmji2ymnvr@alap3.anarazel.de
Pinning/Unpinning a buffer is a very frequent operation; especially in
read-mostly cache resident workloads. Benchmarking shows that in various
scenarios the spinlock protecting a buffer header's state becomes a
significant bottleneck. The problem can be reproduced with pgbench -S on
larger machines, but can be considerably worse for queries which touch
the same buffers over and over at a high frequency (e.g. nested loops
over a small inner table).
To allow atomic operations to be used, cram BufferDesc's flags,
usage_count, buf_hdr_lock, refcount into a single 32bit atomic variable;
that allows to manipulate them together using 32bit compare-and-swap
operations. This requires reducing MAX_BACKENDS to 2^18-1 (which could
be lifted by using a 64bit field, but it's not a realistic configuration
atm).
As not all operations can easily implemented in a lockfree manner,
implement the previous buf_hdr_lock via a flag bit in the atomic
variable. That way we can continue to lock the header in places where
it's needed, but can get away without acquiring it in the more frequent
hot-paths. There's some additional operations which can be done without
the lock, but aren't in this patch; but the most important places are
covered.
As bufmgr.c now essentially re-implements spinlocks, abstract the delay
logic from s_lock.c into something more generic. It now has already two
users, and more are coming up; there's a follupw patch for lwlock.c at
least.
This patch is based on a proof-of-concept written by me, which Alexander
Korotkov made into a fully working patch; the committed version is again
revised by me. Benchmarking and testing has, amongst others, been
provided by Dilip Kumar, Alexander Korotkov, Robert Haas.
On a large x86 system improvements for readonly pgbench, with a high
client count, of a factor of 8 have been observed.
Author: Alexander Korotkov and Andres Freund
Discussion: 2400449.GjM57CE0Yg@dinodell
I used the wrong variable here. Doesn't make a difference today because
the only plausible caller passes a non-NULL variable, but someday it
will be wrong, and even today's correctness is subtle: the caller that
does pass a NULL is never invoked because of object type constraints.
Surely not a condition to rely on.
Noted by Coverity
Since we're requiring pages handled by generic_xlog.c to be standard
format, specify REGBUF_STANDARD when doing a full-page image, so that
xloginsert.c can compress out the "hole" between pd_lower and pd_upper.
Given the current API in which this path will be taken only for a newly
initialized page, the hole is likely to be particularly large in such
cases, so that this oversight could easily be performance-significant.
I don't notice any particular change in the runtime of contrib/bloom's
regression test, though.
Make the inner comparison loops of computeDelta() as tight as possible by
pulling considerations of valid and invalid ranges out of the inner loops,
and extending a match or non-match detection as far as possible before
deciding what to do next. To keep this tractable, give up the possibility
of merging fragments across the pd_lower to pd_upper gap. The fraction of
pages where that could happen (ie, there are 4 or fewer bytes in the gap,
*and* data changes immediately adjacent to it on both sides) is too small
to be worth spending cycles on.
Also, avoid two BLCKSZ-length memcpy()s by computing the delta before
moving data into the target buffer, instead of after. This doesn't save
nearly as many cycles as being tenser about computeDelta(), but it still
seems worth doing.
On my machine, this patch cuts a full 40% off the runtime of
contrib/bloom's regression test.
This routine is unsafe as implemented, because it invalidates the page
image pointers returned by previous GenericXLogRegister() calls.
Rather than complicate the API or the implementation to avoid that,
let's just get rid of it; the use-case for having it seems much
too thin to justify a lot of work here.
While at it, do some wordsmithing on the SGML docs for generic WAL.
Improve commentary, use more specific names for the delta fields,
const-ify pointer arguments where possible, avoid assuming that
initializing only the first element of a local array will guarantee
that the remaining elements end up as we need them. (I think that
code in generic_redo actually worked, but only because InvalidBuffer
is zero; this is a particularly ugly way of depending on that ...)
Previously bcac23d exposed a subset of support functions, namely the
ones Kaigai found useful. In
20160304193704.elq773pyg5fyl3mi@alap3.anarazel.de I mentioned that
there's some functions missing to use the facility in an external
project.
To avoid having to add functions piecemeal, add all the functions which
are used to define READ_* and WRITE_* macros; users of the extensible
node functionality are likely to need these. Additionally expose
outDatum(), which doesn't have it's own WRITE_ macro, as it needs
information from the embedding struct.
Discussion: 20160304193704.elq773pyg5fyl3mi@alap3.anarazel.de
This creates an initial set of default roles which administrators may
use to grant access to, historically, superuser-only functions. Using
these roles instead of granting superuser access reduces the number of
superuser roles required for a system. Documention for each of the
default roles has been added to user-manag.sgml.
Bump catversion to 201604082, as we had a commit that bumped it to
201604081 and another that set it back to 201604071...
Reviews by José Luis Tallón and Robert Haas
This will prevent users from creating roles which begin with "pg_" and
will check for those roles before allowing an upgrade using pg_upgrade.
This will allow for default roles to be provided at initdb time.
Reviews by José Luis Tallón and Robert Haas
This feature is controlled by a new old_snapshot_threshold GUC. A
value of -1 disables the feature, and that is the default. The
value of 0 is just intended for testing. Above that it is the
number of minutes a snapshot can reach before pruning and vacuum
are allowed to remove dead tuples which the snapshot would
otherwise protect. The xmin associated with a transaction ID does
still protect dead tuples. A connection which is using an "old"
snapshot does not get an error unless it accesses a page modified
recently enough that it might not be able to produce accurate
results.
This is similar to the Oracle feature, and we use the same SQLSTATE
and error message for compatibility.
This patch is a no-op patch which is intended to reduce the chances
of failures of omission once the functional part of the "snapshot
too old" patch goes in. It adds parameters for snapshot, relation,
and an enum to specify whether the snapshot age check needs to be
done for the page at this point. This initial patch passes NULL
for the first two new parameters and BGP_NO_SNAPSHOT_TEST for the
third. The follow-on patch will change the places where the test
needs to be made.
These parameters are available for SSPI authentication only, to make
it possible to make it behave more like "normal gssapi", while
making it possible to maintain compatibility.
compat_realm is on by default, but can be turned off to make the
authentication use the full Kerberos realm instead of the NetBIOS name.
upn_username is off by default, and can be turned on to return the users
Kerberos UPN rather than the SAM-compatible name (a user in Active
Directory can have both a legacy SAM-compatible username and a new
Kerberos one. Normally they are the same, but not always)
Author: Christian Ullrich
Reviewed by: Robbie Harwood, Alvaro Herrera, me
OpenSSL has an unfortunate tendency to mix per-session state error
handling with per-thread error handling. This can cause problems when
programs that link to libpq with OpenSSL enabled have some other use of
OpenSSL; without care, one caller of OpenSSL may cause problems for the
other caller. Backend code might similarly be affected, for example
when a third party extension independently uses OpenSSL without taking
the appropriate precautions.
To fix, don't trust other users of OpenSSL to clear the per-thread error
queue. Instead, clear the entire per-thread queue ahead of certain I/O
operations when it appears that there might be trouble (these I/O
operations mostly need to call SSL_get_error() to check for success,
which relies on the queue being empty). This is slightly aggressive,
but it's pretty clear that the other callers have a very dubious claim
to ownership of the per-thread queue. Do this is both frontend and
backend code.
Finally, be more careful about clearing our own error queue, so as to
not cause these problems ourself. It's possibly that control previously
did not always reach SSLerrmessage(), where ERR_get_error() was supposed
to be called to clear the queue's earliest code. Make sure
ERR_get_error() is always called, so as to spare other users of OpenSSL
the possibility of similar problems caused by libpq (as opposed to
problems caused by a third party OpenSSL library like PHP's OpenSSL
extension). Again, do this is both frontend and backend code.
See bug #12799 and https://bugs.php.net/bug.php?id=68276
Based on patches by Dave Vitek and Peter Eisentraut.
From: Peter Geoghegan <pg@bowt.ie>
Create a "bsd" auth method that works the same as "password" so far as
clients are concerned, but calls the BSD Authentication service to
check the password. This is currently only available on OpenBSD.
Marisa Emerson, reviewed by Thomas Munro
This allows parallel aggregation to use them. It may seem surprising
that we use float8_combine for both float4_accum and float8_accum
transition functions, but that's because those functions differ only
in the type of the non-transition-state argument.
Haribabu Kommi, reviewed by David Rowley and Tomas Vondra
As noticed by Tom Lane changing operation's number in commit
bb140506df causes on-disk format incompatibility.
Revert to previous numbering, that is reason to add special array to store
priorities of operation. Also it reverts order of tsquery to previous.
Author: Dmitry Ivanov
Now indexes (but only B-tree for now) can contain "extra" column(s) which
doesn't participate in index structure, they are just stored in leaf
tuples. It allows to use index only scan by using single index instead
of two or more indexes.
Author: Anastasia Lubennikova with minor editorializing by me
Reviewers: David Rowley, Peter Geoghegan, Jeff Janes