Formerly, this function would always return "true" for an appendrel child
relation, because it would think that the appendrel parent was a potential
join target for the child. In principle that should only lead to some
inefficiency in planning, but fuzz testing by Andreas Seltenreich disclosed
that it could lead to "could not find pathkey item to sort" planner errors
in odd corner cases. Specifically, we would think that all columns of a
child table's multicolumn index were interesting pathkeys, causing us to
generate a MergeAppend path that sorts by all the columns. However, if any
of those columns weren't actually used above the level of the appendrel,
they would not get added to that rel's targetlist, which would result in
being unable to resolve the MergeAppend's sort keys against its targetlist
during createplan.c.
Backpatch to 9.3. In older versions, columns of an appendrel get added
to its targetlist even if they're not mentioned above the scan level,
so that the failure doesn't occur. It might be worth back-patching this
fix to older versions anyway, but I'll refrain for the moment.
The tuplesort/tuplestore memory management logic assumed that the chunk
allocation overhead for its memtuples array could not increase when
increasing the array size. This is and always was true for tuplesort,
but we (I, I think) blindly copied that logic into tuplestore.c without
noticing that the assumption failed to hold for the much smaller array
elements used by tuplestore. Given rather small work_mem, this could
result in an improper complaint about "unexpected out-of-memory situation",
as reported by Brent DeSpain in bug #13530.
The easiest way to fix this is just to increase tuplestore's initial
array size so that the assumption holds. Rather than relying on magic
constants, though, let's export a #define from aset.c that represents
the safe allocation threshold, and make tuplestore's calculation depend
on that.
Do the same in tuplesort.c to keep the logic looking parallel, even though
tuplesort.c isn't actually at risk at present. This will keep us from
breaking it if we ever muck with the allocation parameters in aset.c.
Back-patch to all supported versions. The error message doesn't occur
pre-9.3, not so much because the problem can't happen as because the
pre-9.3 tuplestore code neglected to check for it. (The chance of
trouble is a great deal larger as of 9.3, though, due to changes in the
array-size-increasing strategy.) However, allowing LACKMEM() to become
true unexpectedly could still result in less-than-desirable behavior,
so let's patch it all the way back.
Commit c9b0cbe98b accidentally broke the
order of operations during postmaster shutdown: it resulted in removing
the per-socket lockfiles after, not before, postmaster.pid. This creates
a race-condition hazard for a new postmaster that's started immediately
after observing that postmaster.pid has disappeared; if it sees the
socket lockfile still present, it will quite properly refuse to start.
This error appears to be the explanation for at least some of the
intermittent buildfarm failures we've seen in the pg_upgrade test.
Another problem, which has been there all along, is that the postmaster
has never bothered to close() its listen sockets, but has just allowed them
to close at process death. This creates a different race condition for an
incoming postmaster: it might be unable to bind to the desired listen
address because the old postmaster is still incumbent. This might explain
some odd failures we've seen in the past, too. (Note: this is not related
to the fact that individual backends don't close their client communication
sockets. That behavior is intentional and is not changed by this patch.)
Fix by adding an on_proc_exit function that closes the postmaster's ports
explicitly, and (in 9.3 and up) reshuffling the responsibility for where
to unlink the Unix socket files. Lock file unlinking can stay where it
is, but teach it to unlink the lock files in reverse order of creation.
SP-GiST initialized an all-zeros page at vacuum, but that was not
WAL-logged, which is not safe. You might get a torn page write, when it gets
flushed to disk, and end-up with a half-initialized index page. To fix,
leave it in the all-zeros state, and add it to the FSM. It will be
initialized when reused. Also don't set the page-deleted flag when recycling
an empty page. That was also not WAL-logged, and a torn write of that would
cause the page to have an invalid checksum.
Backpatch to 9.2, where SP-GiST indexes were added.
Commit f3b5565dd4 was a couple of bricks shy
of a load; specifically, it missed putting pg_trigger_tgrelid_tgname_index
into the relcache init file, because that index is not used by any
syscache. However, we have historically nailed that index into cache for
performance reasons. The upshot was that load_relcache_init_file always
decided that the init file was busted and silently ignored it, resulting
in a significant hit to backend startup speed.
To fix, reinstantiate RelationIdIsInInitFile() as a wrapper around
RelationSupportsSysCache(), which can know about additional relations
that should be in the init file despite being unknown to syscache.c.
Also install some guards against future mistakes of this type: make
write_relcache_init_file Assert that all nailed relations get written to
the init file, and make load_relcache_init_file emit a WARNING if it takes
the "wrong number of nailed relations" exit path. Now that we remove the
init files during postmaster startup, that case should never occur in the
field, even if we are starting a minor-version update that added or removed
rels from the nailed set. So the warning shouldn't ever be seen by end
users, but it will show up in the regression tests if somebody breaks this
logic.
Back-patch to all supported branches, like the previous commit.
When we invalidate the relcache entry for a system catalog or index, we
must also delete the relcache "init file" if the init file contains a copy
of that rel's entry. The old way of doing this relied on a specially
maintained list of the OIDs of relations present in the init file: we made
the list either when reading the file in, or when writing the file out.
The problem is that when writing the file out, we included only rels
present in our local relcache, which might have already suffered some
deletions due to relcache inval events. In such cases we correctly decided
not to overwrite the real init file with incomplete data --- but we still
used the incomplete initFileRelationIds list for the rest of the current
session. This could result in wrong decisions about whether the session's
own actions require deletion of the init file, potentially allowing an init
file created by some other concurrent session to be left around even though
it's been made stale.
Since we don't support changing the schema of a system catalog at runtime,
the only likely scenario in which this would cause a problem in the field
involves a "vacuum full" on a catalog concurrently with other activity, and
even then it's far from easy to provoke. Remarkably, this has been broken
since 2002 (in commit 7863404417), but we had
never seen a reproducible test case until recently. If it did happen in
the field, the symptoms would probably involve unexpected "cache lookup
failed" errors to begin with, then "could not open file" failures after the
next checkpoint, as all accesses to the affected catalog stopped working.
Recovery would require manually removing the stale "pg_internal.init" file.
To fix, get rid of the initFileRelationIds list, and instead consult
syscache.c's list of relations used in catalog caches to decide whether a
relation is included in the init file. This should be a tad more efficient
anyway, since we're replacing linear search of a list with ~100 entries
with a binary search. It's a bit ugly that the init file contents are now
so directly tied to the catalog caches, but in practice that won't make
much difference.
Back-patch to all supported branches.
When the inner side of a nestloop SEMI or ANTI join is an indexscan that
uses all the join clauses as indexquals, it can be presumed that both
matched and unmatched outer rows will be processed very quickly: for
matched rows, we'll stop after fetching one row from the indexscan, while
for unmatched rows we'll have an indexscan that finds no matching index
entries, which should also be quick. The planner already knew about this,
but it was nonetheless charging for at least one full run of the inner
indexscan, as a consequence of concerns about the behavior of materialized
inner scans --- but those concerns don't apply in the fast case. If the
inner side has low cardinality (many matching rows) this could make an
indexscan plan look far more expensive than it actually is. To fix,
rearrange the work in initial_cost_nestloop/final_cost_nestloop so that we
don't add the inner scan cost until we've inspected the indexquals, and
then we can add either the full-run cost or just the first tuple's cost as
appropriate.
Experimentation with this fix uncovered another problem: add_path and
friends were coded to disregard cheap startup cost when considering
parameterized paths. That's usually okay (and desirable, because it thins
the path herd faster); but in this fast case for SEMI/ANTI joins, it could
result in throwing away the desired plain indexscan path in favor of a
bitmap scan path before we ever get to the join costing logic. In the
many-matching-rows cases of interest here, a bitmap scan will do a lot more
work than required, so this is a problem. To fix, add a per-relation flag
consider_param_startup that works like the existing consider_startup flag,
but applies to parameterized paths, and set it for relations that are the
inside of a SEMI or ANTI join.
To make this patch reasonably safe to back-patch, care has been taken to
avoid changing the planner's behavior except in the very narrow case of
SEMI/ANTI joins with inner indexscans. There are places in
compare_path_costs_fuzzily and add_path_precheck that are not terribly
consistent with the new approach, but changing them will affect planner
decisions at the margins in other cases, so we'll leave that for a
HEAD-only fix.
Back-patch to 9.3; before that, the consider_startup flag didn't exist,
meaning that the second aspect of the patch would be too invasive.
Per a complaint from Peter Holzer and analysis by Tomas Vondra.
Commit 2ce439f337 introduced a rather serious
regression, namely that if its scan of the data directory came across any
un-fsync-able files, it would fail and thereby prevent database startup.
Worse yet, symlinks to such files also caused the problem, which meant that
crash restart was guaranteed to fail on certain common installations such
as older Debian.
After discussion, we agreed that (1) failure to start is worse than any
consequence of not fsync'ing is likely to be, therefore treat all errors
in this code as nonfatal; (2) we should not chase symlinks other than
those that are expected to exist, namely pg_xlog/ and tablespace links
under pg_tblspc/. The latter restriction avoids possibly fsync'ing a
much larger part of the filesystem than intended, if the user has left
random symlinks hanging about in the data directory.
This commit takes care of that and also does some code beautification,
mainly moving the relevant code into fd.c, which seems a much better place
for it than xlog.c, and making sure that the conditional compilation for
the pre_sync_fname pass has something to do with whether pg_flush_data
works.
I also relocated the call site in xlog.c down a few lines; it seems a
bit silly to be doing this before ValidateXLOGDirectoryStructure().
The similar logic in initdb.c ought to be made to match this, but that
change is noncritical and will be dealt with separately.
Back-patch to all active branches, like the prior commit.
Abhijit Menon-Sen and Tom Lane
This reverts commit 16304a0134, except
for its changes in src/port/snprintf.c; as well as commit
cac18a76bb which is no longer needed.
Fujii Masao reported that the previous commit caused failures in psql on
OS X, since if one exits the pager program early while viewing a query
result, psql sees an EPIPE error from fprintf --- and the wrapper function
thought that was reason to panic. (It's a bit surprising that the same
does not happen on Linux.) Further discussion among the security list
concluded that the risk of other such failures was far too great, and
that the one-size-fits-all approach to error handling embodied in the
previous patch is unlikely to be workable.
This leaves us again exposed to the possibility of the type of failure
envisioned in CVE-2015-3166. However, that failure mode is strictly
hypothetical at this point: there is no concrete reason to believe that
an attacker could trigger information disclosure through the supposed
mechanism. In the first place, the attack surface is fairly limited,
since so much of what the backend does with format strings goes through
stringinfo.c or psprintf(), and those already had adequate defenses.
In the second place, even granting that an unprivileged attacker could
control the occurrence of ENOMEM with some precision, it's a stretch to
believe that he could induce it just where the target buffer contains some
valuable information. So we concluded that the risk of non-hypothetical
problems induced by the patch greatly outweighs the security risks.
We will therefore revert, and instead undertake closer analysis to
identify specific calls that may need hardening, rather than attempt a
universal solution.
We have kept the portion of the previous patch that improved snprintf.c's
handling of errors when it calls the platform's sprintf(). That seems to
be an unalloyed improvement.
Security: CVE-2015-3166
All known standard library implementations of these functions can fail
with ENOMEM. A caller neglecting to check for failure would experience
missing output, information exposure, or a crash. Check return values
within wrappers and code, currently just snprintf.c, that bypasses the
wrappers. The wrappers do not return after an error, so their callers
need not check. Back-patch to 9.0 (all supported versions).
Popular free software standard library implementations do take pains to
bypass malloc() in simple cases, but they risk ENOMEM for floating point
numbers, positional arguments, large field widths, and large precisions.
No specification demands such caution, so this commit regards every call
to a printf family function as a potential threat.
Injecting the wrappers implicitly is a compromise between patch scope
and design goals. I would prefer to edit each call site to name a
wrapper explicitly. libpq and the ECPG libraries would, ideally, convey
errors to the caller rather than abort(). All that would be painfully
invasive for a back-patched security fix, hence this compromise.
Security: CVE-2015-3166
The logic introduced in commit b69bf30b9b
and repaired in commits 669c7d20e6 and
7be47c56af helps to ensure that we don't
overwrite old multixact member information while it is still needed,
but a user who creates many large multixacts can still exhaust the
member space (and thus start getting errors) while autovacuum stands
idly by.
To fix this, progressively ramp down the effective value (but not the
actual contents) of autovacuum_multixact_freeze_max_age as member space
utilization increases. This makes autovacuum more aggressive and also
reduces the threshold for a manual VACUUM to perform a full-table scan.
This patch leaves unsolved the problem of ensuring that emergency
autovacuums are triggered even when autovacuum=off. We'll need to fix
that via a separate patch.
Thomas Munro and Robert Haas
Otherwise, if there's another crash, some writes from after the first
crash might make it to disk while writes from before the crash fail
to make it to disk. This could lead to data corruption.
Back-patch to all supported versions.
Abhijit Menon-Sen, reviewed by Andres Freund and slightly revised
by me.
After a timeline switch, we would leave behind recycled WAL segments that
are in the future, but on the old timeline. After promotion, and after they
become old enough to be recycled again, we would notice that they don't have
a .ready or .done file, create a .ready file for them, and archive them.
That's bogus, because the files contain garbage, recycled from an older
timeline (or prealloced as zeros). We shouldn't archive such files.
This could happen when we're following a timeline switch during replay, or
when we switch to new timeline at end-of-recovery.
To fix, whenever we switch to a new timeline, scan the data directory for
WAL segments on the old timeline, but with a higher segment number, and
remove them. Those don't belong to our timeline history, and are most
likely bogus recycled or preallocated files. They could also be valid files
that we streamed from the primary ahead of time, but in any case, they're
not needed to recover to the new timeline.
In investigating yesterday's crash report from Hugo Osvaldo Barrera, I only
looked back as far as commit f3aec2c7f5 where the breakage occurred
(which is why I thought the IPv4-in-IPv6 business was undocumented). But
actually the logic dates back to commit 3c9bb8886d and was simply
broken by erroneous refactoring in the later commit. A bit of archives
excavation shows that we added the whole business in response to a report
that some 2003-era Linux kernels would report IPv4 connections as having
IPv4-in-IPv6 addresses. The fact that we've had no complaints since 9.0
seems to be sufficient confirmation that no modern kernels do that, so
let's just rip it all out rather than trying to fix it.
Do this in the back branches too, thus essentially deciding that our
effective behavior since 9.0 is correct. If there are any platforms on
which the kernel reports IPv4-in-IPv6 addresses as such, yesterday's fix
would have made for a subtle and potentially security-sensitive change in
the effective meaning of IPv4 pg_hba.conf entries, which does not seem like
a good thing to do in minor releases. So let's let the post-9.0 behavior
stand, and change the documentation to match it.
In passing, I failed to resist the temptation to wordsmith the description
of pg_hba.conf IPv4 and IPv6 address entries a bit. A lot of this text
hasn't been touched since we were IPv4-only.
If any error occurred while we were in the middle of reading a protocol
message from the client, we could lose sync, and incorrectly try to
interpret a part of another message as a new protocol message. That will
usually lead to an "invalid frontend message" error that terminates the
connection. However, this is a security issue because an attacker might
be able to deliberately cause an error, inject a Query message in what's
supposed to be just user data, and have the server execute it.
We were quite careful to not have CHECK_FOR_INTERRUPTS() calls or other
operations that could ereport(ERROR) in the middle of processing a message,
but a query cancel interrupt or statement timeout could nevertheless cause
it to happen. Also, the V2 fastpath and COPY handling were not so careful.
It's very difficult to recover in the V2 COPY protocol, so we will just
terminate the connection on error. In practice, that's what happened
previously anyway, as we lost protocol sync.
To fix, add a new variable in pqcomm.c, PqCommReadingMsg, that is set
whenever we're in the middle of reading a message. When it's set, we cannot
safely ERROR out and continue running, because we might've read only part
of a message. PqCommReadingMsg acts somewhat similarly to critical sections
in that if an error occurs while it's set, the error handler will force the
connection to be terminated, as if the error was FATAL. It's not
implemented by promoting ERROR to FATAL in elog.c, like ERROR is promoted
to PANIC in critical sections, because we want to be able to use
PG_TRY/CATCH to recover and regain protocol sync. pq_getmessage() takes
advantage of that to prevent an OOM error from terminating the connection.
To prevent unnecessary connection terminations, add a holdoff mechanism
similar to HOLD/RESUME_INTERRUPTS() that can be used hold off query cancel
interrupts, but still allow die interrupts. The rules on which interrupts
are processed when are now a bit more complicated, so refactor
ProcessInterrupts() and the calls to it in signal handlers so that the
signal handlers always call it if ImmediateInterruptOK is set, and
ProcessInterrupts() can decide to not do anything if the other conditions
are not met.
Reported by Emil Lenngren. Patch reviewed by Noah Misch and Andres Freund.
Backpatch to all supported versions.
Security: CVE-2015-0244
calc_rangesel() failed outright when comparing range variables to empty
constant ranges with < or >=, as a result of missing cases in a switch.
It also produced a bogus estimate for > comparison to an empty range.
On top of that, the >= and > cases were mislabeled throughout. For
nonempty constant ranges, they managed to produce the right answers
anyway as a result of counterbalancing typos.
Also, default_range_selectivity() omitted cases for elem <@ range,
range &< range, and range &> range, so that rather dubious defaults
were applied for these operators.
In passing, rearrange the code in rangesel() so that the elem <@ range
case is handled in a less opaque fashion.
Report and patch by Emre Hasegeli, some additional work by me
As of 9.3, ruleutils.c goes to some lengths to ensure that table and column
aliases used in its output are unique. Of course this takes more time than
was required before, which in itself isn't fatal. However, EXPLAIN was set
up so that recalculation of the unique aliases was repeated for each
subexpression printed in a plan. That results in O(N^2) time and memory
consumption for large plan trees, which did not happen in older branches.
Fortunately, the expensive work is the same across a whole plan tree,
so there is no need to repeat it; we can do most of the initialization
just once per query and re-use it for each subexpression. This buys
back most (not all) of the performance loss since 9.2.
We need an extra ExplainState field to hold the precalculated deparse
context. That's no problem in HEAD, but in the back branches, expanding
sizeof(ExplainState) seems risky because third-party extensions might
have local variables of that struct type. So, in 9.4 and 9.3, introduce
an auxiliary struct to keep sizeof(ExplainState) the same. We should
refactor the APIs to avoid such local variables in future, but that's
material for a separate HEAD-only commit.
Per gripe from Alexey Bashtanov. Back-patch to 9.3 where the issue
was introduced.
Darwin --enable-nls builds use a substitute setlocale() that may start a
thread. Buildfarm member orangutan experienced BackendList corruption
on account of different postmaster threads executing signal handlers
simultaneously. Furthermore, a multithreaded postmaster risks undefined
behavior from sigprocmask() and fork(). Emit LOG messages about the
problem and its workaround. Back-patch to 9.0 (all supported versions).
Backpatch to 9.3 where src/common was introduce, because a bugfix that
needs to be backpatched, requires the function. Earlier branches will
have to duplicate the code.
MapArrayTypeName would copy up to NAMEDATALEN-1 bytes of the base type
name, which of course is wrong: after prepending '_' there is only room for
NAMEDATALEN-2 bytes. Aside from being the wrong result, this case would
lead to overrunning the statically allocated work buffer. This would be a
security bug if the function were ever used outside bootstrap mode, but it
isn't, at least not in any currently supported branches.
Aside from fixing the off-by-one loop logic, this patch gets rid of the
static work buffer by having MapArrayTypeName pstrdup its result; the sole
caller was already doing that, so this just requires moving the pstrdup
call. This saves a few bytes but mainly it makes the API a lot cleaner.
Back-patch on the off chance that there is some third-party code using
MapArrayTypeName with less-secure input. Pushing pstrdup into the function
should not cause any serious problems for such hypothetical code; at worst
there might be a short term memory leak.
Per Coverity scanning.
We expose a function IsValidJsonNumber that internally calls the lexer
for json numbers. That allows us to use the same test everywhere,
instead of inventing a broken test for hstore conversions. The new
function is also used in datum_to_json, replacing the code that is now
moved to the new function.
Backpatch to 9.3 where hstore_to_json_loose was introduced.
There was a window in RestoreBackupBlock where a page would be zeroed out,
but not yet locked. If a backend pinned and locked the page in that window,
it saw the zeroed page instead of the old page or new page contents, which
could lead to missing rows in a result set, or errors.
To fix, replace RBM_ZERO with RBM_ZERO_AND_LOCK, which atomically pins,
zeroes, and locks the page, if it's not in the buffer cache already.
In stable branches, the old RBM_ZERO constant is renamed to RBM_DO_NOT_USE,
to avoid breaking any 3rd party extensions that might use RBM_ZERO. More
importantly, this avoids renumbering the other enum values, which would
cause even bigger confusion in extensions that use ReadBufferExtended, but
haven't been recompiled.
Backpatch to all supported versions; this has been racy since hot standby
was introduced.
This only happens if a client issues a Parse message with an empty query
string, which is a bit odd; but since it is explicitly called out as legal
by our FE/BE protocol spec, we'd probably better continue to allow it.
Fix by adding tests everywhere that the raw_parse_tree field is passed to
functions that don't or shouldn't accept NULL. Also make it clear in the
relevant comments that NULL is an expected case.
This reverts commits a73c9dbab0 and
2e9650cbcf, which fixed specific crash
symptoms by hacking things at what now seems to be the wrong end, ie the
callee functions. Making the callees allow NULL is superficially more
robust, but it's not always true that there is a defensible thing for the
callee to do in such cases. The caller has more context and is better
able to decide what the empty-query case ought to do.
Per followup discussion of bug #11335. Back-patch to 9.2. The code
before that is sufficiently different that it would require development
of a separate patch, which doesn't seem worthwhile for what is believed
to be an essentially cosmetic change.
At one time it wasn't terribly important what column names were associated
with the fields of a composite Datum, but since the introduction of
operations like row_to_json(), it's important that looking up the rowtype
ID embedded in the Datum returns the column names that users would expect.
However, that doesn't work terribly well: you could get the column names
of the underlying table, or column aliases from any level of the query,
depending on minor details of the plan tree. You could even get totally
empty field names, which is disastrous for cases like row_to_json().
It seems unwise to change this behavior too much in stable branches,
however, since users might not have noticed that they weren't getting
the least-unintuitive choice of field names. Therefore, in the back
branches, only change the results when the child plan has returned an
actually-empty field name. (We assume that can't happen with a named
rowtype, so this also dodges the issue of possibly producing RECORD-typed
output from a Var with a named composite result type.) As in the sister
patch for HEAD, we can get a better name to use from the Var's
corresponding RTE. There is no need to touch the RowExpr code since it
was already using a copy of the RTE's alias list for RECORD cases.
Back-patch as far as 9.2. Before that we did not have row_to_json()
so there were no core functions potentially affected by bogus field
names. While 9.1 and earlier do have contrib's hstore(record) which
is also affected, those versions don't seem to produce empty field names
(at least not in the known problem cases), so we'll leave them alone.
As noted by Noah Misch, my initial cut at fixing bug #11638 didn't cover
all cases where ANALYZE might be invoked in an unsafe context. We need to
test the result of IsInTransactionChain not IsTransactionBlock; which is
notationally a pain because IsInTransactionChain requires an isTopLevel
flag, which would have to be passed down through several levels of callers.
I chose to pass in_outer_xact (ie, the result of IsInTransactionChain)
rather than isTopLevel per se, as that seemed marginally more apropos
for the intermediate functions to know about.
CREATE DATABASE and ALTER DATABASE .. SET TABLESPACE copy the source
database directory on the filesystem level. To ensure the on disk
state is consistent they block out users of the affected database and
force a checkpoint to flush out all data to disk. Unfortunately, up to
now, that checkpoint didn't flush out dirty buffers from unlogged
relations.
That bug means there could be leftover dirty buffers in either the
template database, or the database in its old location. Leading to
problems when accessing relations in an inconsistent state; and to
possible problems during shutdown in the SET TABLESPACE case because
buffers belonging files that don't exist anymore are flushed.
This was reported in bug #10675 by Maxim Boguk.
Fix by Pavan Deolasee, modified somewhat by me. Reviewed by MauMau and
Fujii Masao.
Backpatch to 9.1 where unlogged tables were introduced.
Follow our usual style of providing an "extern" for a standard library
function only when we're also providing the implementation. This avoids
issues when the system headers declare the function slightly differently
than we do, as noted by Caleb Welton.
We might have to go to the extent of probing to see if the system headers
declare the function, but let's not do that until it's demonstrated to be
necessary.
Oversight in commit 9e6b1bf258. Back-patch
to all supported branches, as that was.
Up to now, PG has assumed that any given timezone abbreviation (such as
"EDT") represents a constant GMT offset in the usage of any particular
region; we had a way to configure what that offset was, but not for it
to be changeable over time. But, as with most things horological, this
view of the world is too simplistic: there are numerous regions that have
at one time or another switched to a different GMT offset but kept using
the same timezone abbreviation. Almost the entire Russian Federation did
that a few years ago, and later this month they're going to do it again.
And there are similar examples all over the world.
To cope with this, invent the notion of a "dynamic timezone abbreviation",
which is one that is referenced to a particular underlying timezone
(as defined in the IANA timezone database) and means whatever it currently
means in that zone. For zones that use or have used daylight-savings time,
the standard and DST abbreviations continue to have the property that you
can specify standard or DST time and get that time offset whether or not
DST was theoretically in effect at the time. However, the abbreviations
mean what they meant at the time in question (or most recently before that
time) rather than being absolutely fixed.
The standard abbreviation-list files have been changed to use this behavior
for abbreviations that have actually varied in meaning since 1970. The
old simple-numeric definitions are kept for abbreviations that have not
changed, since they are a bit faster to resolve.
While this is clearly a new feature, it seems necessary to back-patch it
into all active branches, because otherwise use of Russian zone
abbreviations is going to become even more problematic than it already was.
This change supersedes the changes in commit 513d06ded et al to modify the
fixed meanings of the Russian abbreviations; since we've not shipped that
yet, this will avoid an undesirably incompatible (not to mention incorrect)
change in behavior for timestamps between 2011 and 2014.
This patch makes some cosmetic changes in ecpglib to keep its usage of
datetime lookup tables as similar as possible to the backend code, but
doesn't do anything about the increasingly obsolete set of timezone
abbreviation definitions that are hard-wired into ecpglib. Whatever we
do about that will likely not be appropriate material for back-patching.
Also, a potential free() of a garbage pointer after an out-of-memory
failure in ecpglib has been fixed.
This patch also fixes pre-existing bugs in DetermineTimeZoneOffset() that
caused it to produce unexpected results near a timezone transition, if
both the "before" and "after" states are marked as standard time. We'd
only ever thought about or tested transitions between standard and DST
time, but that's not what's happening when a zone simply redefines their
base GMT offset.
In passing, update the SGML documentation to refer to the Olson/zoneinfo/
zic timezone database as the "IANA" database, since it's now being
maintained under the auspices of IANA.
As of commit a87c72915 (which later got backpatched as far as 9.1),
we're explicitly supporting the notion that append relations can be
nested; this can occur when UNION ALL constructs are nested, or when
a UNION ALL contains a table with inheritance children.
Bug #11457 from Nelson Page, as well as an earlier report from Elvis
Pranskevichus, showed that there were still nasty bugs associated with such
cases: in particular the EquivalenceClass mechanism could try to generate
"join" clauses connecting an appendrel child to some grandparent appendrel,
which would result in assertion failures or bogus plans.
Upon investigation I concluded that all current callers of
find_childrel_appendrelinfo() need to be fixed to explicitly consider
multiple levels of parent appendrels. The most complex fix was in
processing of "broken" EquivalenceClasses, which are ECs for which we have
been unable to generate all the derived equality clauses we would like to
because of missing cross-type equality operators in the underlying btree
operator family. That code path is more or less entirely untested by
the regression tests to date, because no standard opfamilies have such
holes in them. So I wrote a new regression test script to try to exercise
it a bit, which turned out to be quite a worthwhile activity as it exposed
existing bugs in all supported branches.
The present patch is essentially the same as far back as 9.2, which is
where parameterized paths were introduced. In 9.0 and 9.1, we only need
to back-patch a small fragment of commit 5b7b5518d, which fixes failure to
propagate out the original WHERE clauses when a broken EC contains constant
members. (The regression test case results show that these older branches
are noticeably stupider than 9.2+ in terms of the quality of the plans
generated; but we don't really care about plan quality in such cases,
only that the plan not be outright wrong. A more invasive fix in the
older branches would not be a good idea anyway from a plan-stability
standpoint.)
x86's memory barrier assembly was marked as clobbering "memory" but
not "cc" even though 'addl' sets various flags. As it turns out gcc on
x86 implicitly assumes "cc" on every inline assembler statement, so
it's not a bug. But as that's poorly documented and might get copied
to architectures or compilers where that's not the case, it seems
better to be precise.
Discussion: 20140919100016.GH4277@alap3.anarazel.de
To keep the code common, backpatch to 9.2 where explicit memory
barriers were introduced.
Some Sparc CPUs can be run in various coherence models, ranging from
RMO (relaxed) over PSO (partial) to TSO (total). Solaris has always
run CPUs in TSO mode while in userland, but linux didn't use to and
the various *BSDs still don't. Unfortunately the sparc TAS/S_UNLOCK
were only correct under TSO. Fix that by adding the necessary memory
barrier instructions. On sparcv8+, which should be all relevant CPUs,
these are treated as NOPs if the current consistency model doesn't
require the barriers.
Discussion: 20140630222854.GW26930@awork2.anarazel.de
Will be backpatched to all released branches once a few buildfarm
cycles haven't shown up problems. As I've no access to sparc, this is
blindly written.
There were several oversights in recovery code where COMMIT/ABORT PREPARED
records were ignored:
* pg_last_xact_replay_timestamp() (wasn't updated for 2PC commits)
* recovery_min_apply_delay (2PC commits were applied immediately)
* recovery_target_xid (recovery would not stop if the XID used 2PC)
The first of those was reported by Sergiy Zuban in bug #11032, analyzed by
Tom Lane and Andres Freund. The bug was always there, but was masked before
commit d19bd29f07, because COMMIT PREPARED
always created an extra regular transaction that was WAL-logged.
Backpatch to all supported versions (older versions didn't have all the
features and therefore didn't have all of the above bugs).
Instead of truncating pg_multixact at vacuum time, do it only at
checkpoint time. The reason for doing it this way is twofold: first, we
want it to delete only segments that we're certain will not be required
if there's a crash immediately after the removal; and second, we want to
do it relatively often so that older files are not left behind if
there's an untimely crash.
Per my proposal in
http://www.postgresql.org/message-id/20140626044519.GJ7340@eldon.alvh.no-ip.org
we now execute the truncation in the checkpointer process rather than as
part of vacuum. Vacuum is in only charge of maintaining in shared
memory the value to which it's possible to truncate the files; that
value is stored as part of checkpoints also, and so upon recovery we can
reuse the same value to re-execute truncate and reset the
oldest-value-still-safe-to-use to one known to remain after truncation.
Per bug reported by Jeff Janes in the course of his tests involving
bug #8673.
While at it, update some comments that hadn't been updated since
multixacts were changed.
Backpatch to 9.3, where persistency of pg_multixact files was
introduced by commit 0ac5ad5134.
ExecMakeTableFunctionResult evaluated the arguments for a function-in-FROM
in the query-lifespan memory context. This is insignificant in simple
cases where the function relation is scanned only once; but if the function
is in a sub-SELECT or is on the inside of a nested loop, any memory
consumed during argument evaluation can add up quickly. (The potential for
trouble here had been foreseen long ago, per existing comments; but we'd
not previously seen a complaint from the field about it.) To fix, create
an additional temporary context just for this purpose.
Per an example from MauMau. Back-patch to all active branches.
This function is pervasive on free software operating systems; import
NetBSD's implementation. Back-patch to 8.4, like the commit that will
harness it.
In a50d976254 I already changed this, but got it wrong for the case
where the number of members is larger than the number of entries that
fit in the last page of the last segment.
As reported by Serge Negodyuck in a followup to bug #8673.
In yesterday's commit 2dc4f011fd, I tried
to force buffering of stdout/stderr in initdb to be what it is by
default when the program is run interactively on Unix (since that's how
most manual testing is done). This tripped over the fact that Windows
doesn't support _IOLBF mode. We dealt with that a long time ago in
syslogger.c by falling back to unbuffered mode on Windows. Export that
solution in port.h and use it in initdb.
Back-patch to 8.4, like the previous commit.
To lock a prepared transaction's shared memory entry, we used to mark it
with the XID of the backend. When the XID was no longer active according
to the proc array, the entry was implicitly considered as not locked
anymore. However, when preparing a transaction, the backend's proc array
entry was cleared before transfering the locks (and some other state) to
the prepared transaction's dummy PGPROC entry, so there was a window where
another backend could finish the transaction before it was in fact fully
prepared.
To fix, rewrite the locking mechanism of global transaction entries. Instead
of an XID, just have simple locked-or-not flag in each entry (we store the
locking backend's backend id rather than a simple boolean, but that's just
for debugging purposes). The backend is responsible for explicitly unlocking
the entry, and to make sure that that happens, install a callback to unlock
it on abort or process exit.
Backpatch to all supported versions.
This was not changed in HEAD, but will be done later as part of a
pgindent run. Future pgindent runs will also do this.
Report by Tom Lane
Backpatch through all supported branches, but not HEAD
If we have an array of records stored on disk, the individual record fields
cannot contain out-of-line TOAST pointers: the tuptoaster.c mechanisms are
only prepared to deal with TOAST pointers appearing in top-level fields of
a stored row. The same applies for ranges over composite types, nested
composites, etc. However, the existing code only took care of expanding
sub-field TOAST pointers for the case of nested composites, not for other
structured types containing composites. For example, given a command such
as
UPDATE tab SET arraycol = ARRAY[(ROW(x,42)::mycompositetype] ...
where x is a direct reference to a field of an on-disk tuple, if that field
is long enough to be toasted out-of-line then the TOAST pointer would be
inserted as-is into the array column. If the source record for x is later
deleted, the array field value would become a dangling pointer, leading
to errors along the line of "missing chunk number 0 for toast value ..."
when the value is referenced. A reproducible test case for this was
provided by Jan Pecek, but it seems likely that some of the "missing chunk
number" reports we've heard in the past were caused by similar issues.
Code-wise, the problem is that PG_DETOAST_DATUM() is not adequate to
produce a self-contained Datum value if the Datum is of composite type.
Seen in this light, the problem is not just confined to arrays and ranges,
but could also affect some other places where detoasting is done in that
way, for example form_index_tuple().
I tried teaching the array code to apply toast_flatten_tuple_attribute()
along with PG_DETOAST_DATUM() when the array element type is composite,
but this was messy and imposed extra cache lookup costs whether or not any
TOAST pointers were present, indeed sometimes when the array element type
isn't even composite (since sometimes it takes a typcache lookup to find
that out). The idea of extending that approach to all the places that
currently use PG_DETOAST_DATUM() wasn't attractive at all.
This patch instead solves the problem by decreeing that composite Datum
values must not contain any out-of-line TOAST pointers in the first place;
that is, we expand out-of-line fields at the point of constructing a
composite Datum, not at the point where we're about to insert it into a
larger tuple. This rule is applied only to true composite Datums, not
to tuples that are being passed around the system as tuples, so it's not
as invasive as it might sound at first. With this approach, the amount
of code that has to be touched for a full solution is greatly reduced,
and added cache lookup costs are avoided except when there actually is
a TOAST pointer that needs to be inlined.
The main drawback of this approach is that we might sometimes dereference
a TOAST pointer that will never actually be used by the query, imposing a
rather large cost that wasn't there before. On the other side of the coin,
if the field value is used multiple times then we'll come out ahead by
avoiding repeat detoastings. Experimentation suggests that common SQL
coding patterns are unaffected either way, though. Applications that are
very negatively affected could be advised to modify their code to not fetch
columns they won't be using.
In future, we might consider reverting this solution in favor of detoasting
only at the point where data is about to be stored to disk, using some
method that can drill down into multiple levels of nested structured types.
That will require defining new APIs for structured types, though, so it
doesn't seem feasible as a back-patchable fix.
Note that this patch changes HeapTupleGetDatum() from a macro to a function
call; this means that any third-party code using that macro will not get
protection against creating TOAST-pointer-containing Datums until it's
recompiled. The same applies to any uses of PG_RETURN_HEAPTUPLEHEADER().
It seems likely that this is not a big problem in practice: most of the
tuple-returning functions in core and contrib produce outputs that could
not possibly be toasted anyway, and the same probably holds for third-party
extensions.
This bug has existed since TOAST was invented, so back-patch to all
supported branches.
