aboutsummaryrefslogtreecommitdiff
path: root/src
Commit message (Collapse)AuthorAge
...
* Re-bin segment when memory pages are freed.Thomas Munro2023-07-04
| | | | | | | | | | | | | | | | | | | | | It's OK to be lazy about re-binning memory segments when allocating, because that can only leave segments in a bin that's too high. We'll search higher bins if necessary while allocating next time, and also eventually re-bin, so no memory can become unreachable that way. However, when freeing memory, the largest contiguous range of free pages might go up, so we should re-bin eagerly to make sure we don't leave the segment in a bin that is too low for get_best_segment() to find. The re-binning code is moved into a function of its own, so it can be called whenever free pages are returned to the segment's free page map. Back-patch to all supported releases. Author: Dongming Liu <ldming101@gmail.com> Reviewed-by: Robert Haas <robertmhaas@gmail.com> (earlier version) Reviewed-by: Thomas Munro <thomas.munro@gmail.com> Discussion: https://postgr.es/m/CAL1p7e8LzB2LSeAXo2pXCW4%2BRya9s0sJ3G_ReKOU%3DAjSUWjHWQ%40mail.gmail.com
* Remove trailing zero words from BitmapsetsDavid Rowley2023-07-04
| | | | | | | | | | | | | | | | | | | | | Prior to this, Bitmapsets could contain trailing words which had no members set. Many places in bitmapset.c had to loop over trailing words to check for empty words. If we ensure we always remove these trailing zero words then we can perform various optimizations such as fast pathing bms_is_subset to return false when 'a' has more words than 'b'. A similar optimization is possible in bms_equal. Both of these together can yield quite significant performance increases in the query planner when querying a partitioned table with around 100 or more partitions. While we're at it, since the minimum number of words a Bitmapset can contain is 1, we can make use of do/while loops instead of for loops when looping over all words in a set. This means checking the loop condition 1 less time, which for single-word sets cuts the loop condition checks in half. Author: David Rowley Reviewed-by: Yuya Watari Discussion: https://postgr.es/m/CAApHDvr5O41MuUjw0DQKqmAnv7QqvmLqXReEd5o4nXTzWp8-+w@mail.gmail.com
* Increase size of bgw_library_name.Nathan Bossart2023-07-03
| | | | | | | | | | | This commit increases the size of the bgw_library_name member of the BackgroundWorker struct from BGW_MAXLEN (96) bytes to MAXPGPATH (default of 1024) bytes so that it can store longer file names (e.g., absolute paths). Author: Yurii Rashkovskii Reviewed-by: Daniel Gustafsson, Aleksander Alekseev Discussion: https://postgr.es/m/CA%2BRLCQyjFV5Y8tG5QgUb6gjteL4S3p%2B1gcyqWTqigyM93WZ9Pg%40mail.gmail.com
* Fix race in SSI interaction with gin fast path.Thomas Munro2023-07-04
| | | | | | | | | | | | | | | | | | The ginfast.c code previously checked for conflicts in before locking the relevant buffer, leaving a window where a RW conflict could be missed. Re-order. There was also a place where buffer ID and block number were confused while trying to predicate-lock a page, noted by visual inspection. Back-patch to all supported releases. Fixes one more problem discovered with the reproducer from bug #17949, in this case when Dmitry tried other index types. Reported-by: Artem Anisimov <artem.anisimov.255@gmail.com> Reported-by: Dmitry Dolgov <9erthalion6@gmail.com> Reviewed-by: Heikki Linnakangas <hlinnaka@iki.fi> Discussion: https://postgr.es/m/17949-a0f17035294a55e2%40postgresql.org
* Fix race in SSI interaction with bitmap heap scan.Thomas Munro2023-07-04
| | | | | | | | | | | | | | | | | | | | | | | | | | | | When performing a bitmap heap scan, we don't want to miss concurrent writes that occurred after we observed the heap's rs_nblocks, but before we took predicate locks on index pages. Therefore, we can't skip fetching any heap tuples that are referenced by the index, because we need to test them all with CheckForSerializableConflictOut(). The old optimization that would ignore any references to blocks >= rs_nblocks gets in the way of that requirement, because it means that concurrent writes in that window are ignored. Removing that optimization shouldn't affect correctness at any isolation level, because any new tuples shouldn't be visible to an MVCC snapshot. There also shouldn't be any error-causing references to heap blocks past the end, because we should have held at least an AccessShareLock on the table before the index scan. It can't get smaller while our transaction is running. For now, though, we'll keep the optimization at lower levels to avoid making unnecessary changes in a bug fix. Back-patch to all supported releases. In release 11, the code is in a different place but not fundamentally different. Fixes one aspect of bug #17949. Reported-by: Artem Anisimov <artem.anisimov.255@gmail.com> Reviewed-by: Dmitry Dolgov <9erthalion6@gmail.com> Reviewed-by: Heikki Linnakangas <hlinnaka@iki.fi> Discussion: https://postgr.es/m/17949-a0f17035294a55e2%40postgresql.org
* Fix race in SSI interaction with empty btrees.Thomas Munro2023-07-04
| | | | | | | | | | | | | | | | | When predicate-locking btrees, we have a special case for completely empty btrees, since there is no page to lock. This was racy, because, without buffer lock held, a matching key could be inserted between the _bt_search() and the PredicateLockRelation() calls. Fix, by rechecking _bt_search() after taking the relation-level SIREAD lock, if using SERIALIZABLE isolation and an empty btree is discovered. Back-patch to all supported releases. Fixes one aspect of bug #17949. Reported-by: Artem Anisimov <artem.anisimov.255@gmail.com> Reviewed-by: Dmitry Dolgov <9erthalion6@gmail.com> Reviewed-by: Heikki Linnakangas <hlinnaka@iki.fi> Discussion: https://postgr.es/m/17949-a0f17035294a55e2%40postgresql.org
* Don't truncate database and user names in startup packets.Nathan Bossart2023-07-03
| | | | | | | | | | | | | | | | | | | | Unlike commands such as CREATE DATABASE, ProcessStartupPacket() does not perform multibyte-aware truncation of overlength names. This means that connection attempts might fail even if the user provides the same overlength names that were used in CREATE DATABASE, CREATE ROLE, etc. Ideally, we'd do the same multibyte- aware truncation in both code paths, but it doesn't seem worth the added complexity of trying to discover the encoding of the names. Instead, let's simply skip truncating the names in the startup packet and let the user/database lookup fail later on. With this change, users must provide the exact names stored in the catalogs, even if the names were truncated. This reverts commit d18c1d1f51. Author: Bertrand Drouvot Reviewed-by: Kyotaro Horiguchi, Tom Lane Discussion: https://postgr.es/m/07436793-1426-29b2-f924-db7422a05fb7%40gmail.com
* Consider fillfactor when estimating relation sizeTomas Vondra2023-07-03
| | | | | | | | | | | | | | | | | When table_block_relation_estimate_size() estimated the number of tuples in a relation without statistics (e.g. right after load), it did not consider fillfactor when calculating density. With non-default fillfactor values, this may result in significant overestimate of the number of tuples - up to 10x with the minimum 10% fillfactor. This may have unexpected consequences, e.g. when creating hash indexes. This considers the current fillfactor value in the "no statistics" code path. If the fillfactor changes after loading data into the table, the estimate may be off. But that seems much less likely than changing the fillfactor before the data load. Reviewed-by: Corey Huinker, Peter Eisentraut Discussion: https://postgr.es/m/cf154ef9-6bac-d268-b735-67a3443debba@enterprisedb.com
* Improve pg_basebackup long file name test Windows robustnessAndrew Dunstan2023-07-03
| | | | | | | | | | Creation of a file with a very long name can create problems on Windows due to its file path limits. Work around that by creating the file via a symlink with a shorter name. Error displayed by buildfarm animal fairywren.o Backpatch to all live branches
* Fix code indentation violationsTomas Vondra2023-07-03
| | | | | | | Commits ce5aaea8cd, 2b8b2852bb and 28d03feac3 violated the expected code indentation rules, upsetting the new buildfarm member "koel." Discussion: https://postgr.es/m/ZKIU4mhWpgJOM0W0%40paquier.xyz
* A minor simplification for List manipulationPeter Eisentraut2023-07-03
| | | | | | | | | Fix one place that was using lfirst(list_head(list)) by using linitial(list) instead. They are equivalent but the latter is simpler. We did the same in 9d299a49. Author: Richard Guo <guofenglinux@gmail.com> Discussion: https://www.postgresql.org/message-id/flat/CAMbWs49dJnpezDQDDxCPKq7+=_3NyqLqGqnhqCjd+dYe4MS15w@mail.gmail.com
* Update PG_CACHE_LINE_SIZE description.Heikki Linnakangas2023-07-03
| | | | | | | | | PG_CACHE_LINE_SIZE was originally only used in xlog.c, but this hasn't been true for a very long time and is now wildly used, so modify its description to not mention any explicit source code file. Author: Julien Rouhaud Discussion: https://www.postgresql.org/message-id/20230701074936.p3qcssl4t7murt2q@jrouhaud
* Take pg_attribute out of VacAttrStatsPeter Eisentraut2023-07-03
| | | | | | | | | | | | | The VacAttrStats structure contained the whole Form_pg_attribute for a column, but it actually only needs attstattarget from there. So remove the Form_pg_attribute field and make a separate field for attstattarget. This simplifies some code for extended statistics that doesn't deal with a column but an expression, which had to fake up pg_attribute rows to satisfy internal APIs. Also, we can remove some comments that essentially said "don't look at pg_attribute directly". Reviewed-by: Tomas Vondra <tomas.vondra@enterprisedb.com> Discussion: https://www.postgresql.org/message-id/flat/d6069765-5971-04d3-c10d-e4f7b2e9c459%40eisentraut.org
* Add macro for maximum statistics targetPeter Eisentraut2023-07-03
| | | | | | | | The number of places where 10000 was hardcoded had grown a bit beyond the comfort level. Introduce a macro MAX_STATISTICS_TARGET instead. Reviewed-by: Tomas Vondra <tomas.vondra@enterprisedb.com> Discussion: https://www.postgresql.org/message-id/flat/d6069765-5971-04d3-c10d-e4f7b2e9c459%40eisentraut.org
* Change type of pg_statistic_ext.stxstattargetPeter Eisentraut2023-07-03
| | | | | | | | Change from int32 to int16, to match attstattarget (changed in 90189eefc1). Reviewed-by: Tomas Vondra <tomas.vondra@enterprisedb.com> Discussion: https://www.postgresql.org/message-id/flat/d6069765-5971-04d3-c10d-e4f7b2e9c459%40eisentraut.org
* Remove support for OpenSSL 1.0.1Michael Paquier2023-07-03
| | | | | | | | | | | | | | Here are some notes about this change: - As X509_get_signature_nid() should always exist (OpenSSL and LibreSSL), hence HAVE_X509_GET_SIGNATURE_NID is now gone. - OPENSSL_API_COMPAT is bumped to 0x10002000L. - One comment related to 1.0.1e introduced by 74242c2 is removed. Upstream OpenSSL still provides long-term support for 1.0.2 in a closed fashion, so removing it is out of scope for a few years, at least. Reviewed-by: Jacob Champion, Daniel Gustafsson Discussion: https://postgr.es/m/ZG3JNursG69dz1lr@paquier.xyz
* Refactor some code related to wait events "BufferPin" and "Extension"Michael Paquier2023-07-03
| | | | | | | | | | | | | | | | | | The following changes are done: - Addition of WaitEventBufferPin and WaitEventExtension, that hold a list of wait events related to each category. - Addition of two functions that encapsulate the list of wait events for each category. - Rename BUFFER_PIN to BUFFERPIN (only this wait event class used an underscore, requiring a specific rule in the automation script). These changes make a bit easier the automatic generation of all the code and documentation related to wait events, as all the wait event categories are now controlled by consistent structures and functions. Author: Bertrand Drouvot Discussion: https://postgr.es/m/c6f35117-4b20-4c78-1df5-d3056010dcf5@gmail.com Discussion: https://postgr.es/m/77a86b3a-c4a8-5f5d-69b9-d70bbf2e9b98@gmail.com
* Make PG_TEST_NOCLEAN work for temporary directories in TAP testsMichael Paquier2023-07-03
| | | | | | | | | | | | | | | | When set, this environment variable was only effective for data directories but not for all the other temporary files created by PostgreSQL::Test::Utils. Keeping the temporary files after a successful run can be useful for debugging purposes. The documentation is updated to reflect the new behavior, with contents available in doc/ since v16 and in src/test/perl/README since v15. Author: Jacob Champion Reviewed-by: Daniel Gustafsson Discussion: https://postgr.es/m/CAAWbhmgHtDH1SGZ+Fw05CsXtE0mzTmjbuUxLB9mY9iPKgM6cUw@mail.gmail.com Discussion: https://postgr.es/m/YyPd9unV14SX2bLF@paquier.xyz Backpatch-through: 11
* Remove redundant PARTITION BY columns from WindowClausesDavid Rowley2023-07-03
| | | | | | | | | | | | | | | | | | | Here we adjust the query planner to have it remove items from a window clause's PARTITION BY clause in cases where the pathkey for a column in the PARTITION BY clause is redundant. Doing this allows the optimization added in 9d9c02ccd to stop window aggregation early rather than going into "pass-through" mode to find tuples belonging to the next partition. Also, when we manage to remove all PARTITION BY columns, we now no longer needlessly check that the current tuple belongs to the same partition as the last tuple in nodeWindowAgg.c. If the pathkey was redundant then all tuples must contain the same value for the given redundant column, so there's no point in checking that during execution. Author: David Rowley Reviewed-by: Richard Guo Discussion: https://postgr.es/m/CAApHDvo2ji+hdxrxfXtRtsfSVw3to2o1nCO20qimw0dUGK8hcQ@mail.gmail.com
* Silence "missing contrecord" error.Thomas Munro2023-07-03
| | | | | | | | | | | | | | | | | Commit dd38ff28ad added a new error message "missing contrecord" when we fail to reassemble a record. Unfortunately that caused noisy messages to be logged by pg_waldump at end of segment, and by walsender when asked to shut down on a segment boundary. Remove the new error message, so that this condition signals end-of- WAL without a message. It's arguably a reportable condition that should not be silenced while performing crash recovery, but fixing that without introducing noise in the other cases will require more research. Back-patch to 15. Reported-by: Tomas Vondra <tomas.vondra@enterprisedb.com> Discussion: https://postgr.es/m/6a1df56e-4656-b3ce-4b7a-a9cb41df8189%40enterprisedb.com
* Fix oversight in handling of modifiedCols since f24523672dTomas Vondra2023-07-02
| | | | | | | | | | | | | | | | | | | Commit f24523672d fixed a memory leak by moving the modifiedCols bitmap into the per-row memory context. In the case of AFTER UPDATE triggers, the bitmap is however referenced from an event kept until the end of the query, resulting in a use-after-free bug. Fixed by copying the bitmap into the AfterTriggerEvents memory context, which is the one where we keep the trigger events. There's only one place that needs to do the copy, but the memory context may not exist yet. Doing that in a separate function seems more readable. Report by Alexander Pyhalov, fix by me. Backpatch to 13, where the bitmap was added to the event by commit 71d60e2aa0. Reported-by: Alexander Pyhalov Backpatch-through: 13 Discussion: https://postgr.es/m/acddb17c89b0d6cb940eaeda18c08bbe@postgrespro.ru
* Fix memory leak in Incremental Sort rescansTomas Vondra2023-07-02
| | | | | | | | | | | | | | | | | | | | | | | | The Incremental Sort had a couple issues, resulting in leaking memory during rescans, possibly triggering OOM. The code had a couple of related flaws: 1. During rescans, the sort states were reset but then also set to NULL (despite the comment saying otherwise). ExecIncrementalSort then sees NULL and initializes a new sort state, leaking the memory used by the old one. 2. Initializing the sort state also automatically rebuilt the info about presorted keys, leaking the already initialized info. presorted_keys was also unnecessarily reset to NULL. Patch by James Coleman, based on patches by Laurenz Albe and Tom Lane. Backpatch to 13, where Incremental Sort was introduced. Author: James Coleman, Laurenz Albe, Tom Lane Reported-by: Laurenz Albe, Zu-Ming Jiang Backpatch-through: 13 Discussion: https://postgr.es/m/b2bd02dff61af15e3526293e2771f874cf2a3be7.camel%40cybertec.at Discussion: https://postgr.es/m/db03c582-086d-e7cd-d4a1-3bc722f81765%40inf.ethz.ch
* Improve BRIN minmax-multi opclass test coverageTomas Vondra2023-07-02
| | | | | | | | | | | | | | | | | | | | | | Per the code coverage report, the existing regression tests did not exercice some a couple important BRIN minmax-multi code paths. - The tests focused on testing planning with a range of scan key strategies, but not the execution. Fixed by adding queries that actually test query execution for both equality and inequality. - All tests created indexes after inserting data, but this only exercises the CREATE INDEX strategy that sees all values at once, not incremental summary updates. The new tests flip the order and create the index before adding data. - The assert check(s) validating correctness of expanded ranges were present only in the "union" code path, which is not covered by regression tests at all (as it requires concurrency etc.). Fixed by adding the asserts to a couple more places. Reviewed-by: Heikki Linnakangas Discussion: https://postgr.es/m/57020b2e-d9c9-9bc7-4892-b36d9bb07563%40enterprisedb.com
* Introduce bloom_filter_size for BRIN bloom opclassTomas Vondra2023-07-02
| | | | | | | | | Move the calculation of Bloom filter parameters (for BRIN indexes) into a separate function to make reuse easier. At the moment we only call it from one place, but that may change and it's easier to read anyway. Reviewed-by: Heikki Linnakangas Discussion: https://postgr.es/m/0e1f3350-c9cf-ab62-43a5-5dae314de89c%40enterprisedb.com
* Minor cleanups in the BRIN codeTomas Vondra2023-07-02
| | | | | | | | | | | | BRIN bloom and minmax-multi opclasses were somewhat inconsistent when dealing with bool variables, assigning to them Datum values etc. While not a bug, it makes the code harder to understand, so fix that. While at it, update an incorrect comment copied to bloom opclass from minmax, talking about strategies not supported by bloom. Reviewed-by: Heikki Linnakangas Discussion: https://postgr.es/m/0e1f3350-c9cf-ab62-43a5-5dae314de89c%40enterprisedb.com
* Trust signalfd on illumos, again.Thomas Munro2023-07-02
| | | | | | | | | | | | | | | | | Commit 3ab4fc5d avoided choosing signalfd by default on illumos, because it triggered kernel panics. That was fixed, so we can remove a kludge from our code. Users/packagers can still override the default choice at compile time if desired, and we'll leave the back-branches unchanged so they keep choosing self-pipe by default, but we'll default to signalfd (like we do for Linux) in 17. Fixed kernels should be everywhere by the time 17 ships. The illumos issues were: * https://www.illumos.org/issues/13700 * https://www.illumos.org/issues/14892 Discussion: https://postgr.es/m/CA+hUKG+NK-K_G_i1H3OpDTwYPEsiwQi_jw58PGcW2H+-N2eVCA@mail.gmail.com
* Remove redundant check for fast_forward.Heikki Linnakangas2023-06-30
| | | | | | | We already checked for it earlier in the function. Reviewed-by: Andres Freund Discussion: https://www.postgresql.org/message-id/1ba2899e-77f8-7866-79e5-f3b7d1251a3e@iki.fi
* Improve comment on why we need ctid->(cmin,cmax) mapping.Heikki Linnakangas2023-06-30
| | | | | | | Combocids are only part of the problem. Explain the problem in more detail. Reviewed-by: Andres Freund Discussion: https://www.postgresql.org/message-id/1ba2899e-77f8-7866-79e5-f3b7d1251a3e@iki.fi
* Introduce long options in pg_archivecleanupMichael Paquier2023-06-30
| | | | | | | | | | | | | | This patch is a preliminary refactoring for an upcoming patch aimed at adding new options to this tool, and using long options for these is more user-friendly. The existing short options gain long flavors, as of: * -d/--debug * -n/--dry-run * -x/--strip-extension Author: Atsushi Torikoshi Reviewed-by: Fujii Masao, Kyotaro Horiguchi, Michael Paquier Discussion: https://postgr.es/m/d660ef741ce3d82f3b4283f1cafd576c@oss.nttdata.com
* Fix marking of indisvalid for partitioned indexes at creationMichael Paquier2023-06-30
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The logic that introduced partitioned indexes missed a few things when invalidating a partitioned index when these are created, still the code is written to handle recursions: 1) If created from scratch because a mapping index could not be found, the new index created could be itself invalid, if for example it was a partitioned index with one of its leaves invalid. 2) A CCI was missing when indisvalid is set for a parent index, leading to inconsistent trees when recursing across more than one level for a partitioned index creation if an invalidation of the parent was required. This could lead to the creation of a partition index tree where some of the partitioned indexes are marked as invalid, but some of the parents are marked valid, which is not something that should happen (as validatePartitionedIndex() defines, indisvalid is switched to true for a partitioned index iff all its partitions are themselves valid). This patch makes sure that indisvalid is set to false on a partitioned index if at least one of its partition is invalid. The flag is set to true if *all* its partitions are valid. The regression test added in this commit abuses of a failed concurrent index creation, marked as invalid, that maps with an index created on its partitioned table afterwards. Reported-by: Alexander Lakhin Reviewed-by: Alexander Lakhin Discussion: https://postgr.es/m/14987634-43c0-0cb3-e075-94d423607e08@gmail.com Backpatch-through: 11
* Add tab completion for CREATE SCHEMA in psqlMichael Paquier2023-06-30
| | | | | | | | | | | | | | | | | The following patterns are added for CREATE SCHEMA: - AUTHORIZATION, without a schema name or after a schema name. - Possible list of owner roles after AUTHORIZATION. - CREATE and GRANT within the supported set of commands. - Correct object types supported in an embedded CREATE SCHEMA command. While on it, this commit adjusts the completion done after CREATE UNLOGGED: - Addition of SEQUENCE. - Avoid suggesting MATERIALIZED VIEW in CREATE TABLE. Author: Dagfinn Ilmari Mannsåker Reviewed-by: Suraj Khamkar, Michael Paquier Discussion: https://postgr.es/m/8735snihmz.fsf@wibble.ilmari.org
* Use named captures in Catalog::ParseHeader()Michael Paquier2023-06-30
| | | | | | | | | | | | | | | Using at least perl 5.14 is required since 4c15327, meaning that it is possible to use named captures and the %+ hash instead of having to count parenthesis groups manually. While on it, CATALOG is made more flexible in its handling of whitespaces for parameter lists (see the addition of \s* in this case). The generated postgres.bki remains exactly the same before and after this commit. Author: Dagfinn Ilmari Mannsåker Reviewed-by: John Naylor Discussion: https://postgr.es/m/87y1l3s7o9.fsf@wibble.ilmari.org
* Stamp HEAD as 17devel.Tom Lane2023-06-29
| | | | Let the hacking begin ...
* Fix pg_depend entry to AMs after ALTER TABLE .. SET ACCESS METHODMichael Paquier2023-06-30
| | | | | | | | | | | | | | | | | | | | | | ALTER TABLE .. SET ACCESS METHOD was not registering a dependency to the new access method with the relation altered in its rewrite phase, making possible the drop of an access method even if there are relations that depend on it. During the rewrite, a temporary relation is created to build the new relation files before swapping the new and old files, and, while the temporary relation was registering a correct dependency to the new AM, the old relation did not do that. A dependency on the access method is added when the relation files are swapped, which is the point where pg_class is updated. Materialized views and tables use the same code path, hence both were impacted. Backpatch down to 15, where this command has been introduced. Reported-by: Alexander Lakhin Reviewed-by: Nathan Bossart, Andres Freund Discussion: https://postgr.es/m/18000-9145c25b1af475ca@postgresql.org Backpatch-through: 15
* Defend against bogus parameterization of join input paths.Tom Lane2023-06-29
| | | | | | | | | | | | | | | | | | | An outer join cannot be formed using an input path that is parameterized by a value that is supposed to be nulled by the outer join. This is obviously nonsensical, and it could lead to a bad plan being selected; although currently it seems that we'll hit various sanity-check assertions first. I think that such cases were formerly prevented by the delay_upper_joins mechanism, but now that that's gone we need an explicit check. (Perhaps we should avoid generating baserel paths that could lead to this situation in the first place; but it seems like having a defense at the join level would be a good idea anyway.) Richard Guo and Tom Lane, per report from Jaime Casanova Discussion: https://postgr.es/m/CAJKUy5g2uZRrUDZJ8p-=giwcSHVUn0c9nmdxPSY0jF0Ov8VoEA@mail.gmail.com
* Fix order of operations in ExecEvalFieldStoreDeForm().Tom Lane2023-06-29
| | | | | | | | | | | | | | | | | | | | | | | | | If the given composite datum is toasted out-of-line, DatumGetHeapTupleHeader will perform database accesses to detoast it. That can invalidate the result of get_cached_rowtype, as documented (perhaps not plainly enough) in that function's API spec; which leads to strange errors or crashes when we try to use the TupleDesc to read the tuple. In short then, trying to update a field of a composite column could fail intermittently if the overall column value is wide enough to require toasting. We can fix the bug at no cost by just changing the order of operations, since we don't need the TupleDesc until after detoasting. (Other callers of get_cached_rowtype appear to get this right already, so there's only one bug.) Note that the added regression test case reveals this bug reliably only with debug_discard_caches/CLOBBER_CACHE_ALWAYS. Per bug #17994 from Alexander Lakhin. Sadly, this patch does not fix the missing-values issue revealed in the bug discussion; we'll need some more work to cover that. Discussion: https://postgr.es/m/17994-5c7100b51b4790e9@postgresql.org
* meson: Make some Meson style more consistent with surrounding codePeter Eisentraut2023-06-29
| | | | | Author: Tristan Partin <tristan@neon.tech> Discussion: https://www.postgresql.org/message-id/flat/CSPIJVUDZFKX.3KHMOAVGF94RV%40c3po
* Remove inappropriate raw_expression_tree_walker() codePeter Eisentraut2023-06-29
| | | | | | | | It was walking into the ColumnDef->compression field, which is not a node but a string. This code is currently not reachable (because the compression field is only set in situations that don't go through raw_expression_tree_walker()), but if it had been, this could have behaved erratically.
* Error message wording improvementsPeter Eisentraut2023-06-29
|
* Fix psql \?'s entries for \dp and \z.Nathan Bossart2023-06-28
| | | | | | | | | d913928c9c added support for the "S" modifier to psql's \dp and \z meta-commands, but it missed updating the corresponding entries in \?'s output. Author: Noriyoshi Shinoda Discussion: https://postgr.es/m/DM4PR84MB17342A51B3A1556CFBC7A4B2EE25A%40DM4PR84MB1734.NAMPRD84.PROD.OUTLOOK.COM
* Reword error messages for consistencyPeter Eisentraut2023-06-28
|
* Add timeline ID to file names generated with pg_waldump --save-fullpageMichael Paquier2023-06-28
| | | | | | | | | | | | | | | Not including the timeline IDs to the file names generated by pg_waldump for the individual blocks saved could cause some of these files to be overwritten when scanning segments across multiple timelines. Having this information is also as much useful as the LSNs, to be able to know from exactly which WAL segment a block is comes from. While on it, this fixes a few comments in the tests, where the format of the file was not described as matching with the reality. Reported-by: Fujii Masao Reviewed-by: Kyotaro Horiguchi, David Christensen Discussion: https://postgr.es/m/ZJp921+nITFnvBVS@paquier.xyz
* Ignore invalid indexes when enforcing index rules in ALTER TABLE ATTACH ↵Michael Paquier2023-06-28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | PARTITION A portion of ALTER TABLE .. ATTACH PARTITION is to ensure that the partition being attached to the partitioned table has a correct set of indexes, so as there is a consistent index mapping between the partitioned table and its new-to-be partition. However, as introduced in 8b08f7d, the current logic could choose an invalid index as a match, which is something that can exist when dealing with more than two levels of partitioning, like attaching a partitioned table (that has partitions, with an index created by CREATE INDEX ON ONLY) to another partitioned table. A partitioned index with indisvalid set to false is equivalent to an incomplete partition tree, meaning that an invalid partitioned index does not have indexes defined in all its partitions. Hence, choosing an invalid partitioned index can create inconsistent partition index trees, where the parent attaching to is valid, but its partition may be invalid. In the report from Alexander Lakhin, this showed up as an assertion failure when validating an index. Without assertions enabled, the partition index tree would be actually broken, as indisvalid should be switched to true for a partitioned index once all its partitions are themselves valid. With two levels of partitioning, the top partitioned table used a valid index and was able to link to an invalid index stored on its partition, itself a partitioned table. I have studied a few options here (like the possibility to switch indisvalid to false for the parent), but came down to the conclusion that we'd better rely on a simple rule: invalid indexes had better never be chosen, so as the partition attached uses and creates indexes that the parent expects. Some regression tests are added to provide some coverage. Note that the existing coverage is not impacted. This is a problem since partitioned indexes exist, so backpatch all the way down to v11. Reported-by: Alexander Lakhin Discussion: https://postgr.es/14987634-43c0-0cb3-e075-94d423607e08@gmail.com Backpatch-through: 11
* Remove dependency to query text in JumbleQuery()Michael Paquier2023-06-28
| | | | | | | | | | | Since 3db72eb, the query ID of utilities is generated using the Query structure, making the use of the query string in JumbleQuery() unnecessary. This commit removes the argument "querytext" from JumbleQuery(). Reported-by: Joe Conway Reviewed-by: Nathan Bossart Discussion: https://postgr.es/m/ZJlQAWE4COFqHuAV@paquier.xyz
* Translation updatesPeter Eisentraut2023-06-26
| | | | | Source-Git-URL: https://git.postgresql.org/git/pgtranslation/messages.git Source-Git-Hash: ab77975e9d2cde44da796c18af3ec1a66f0df7ae
* Change "..." to cstring in old input/output function comments.Heikki Linnakangas2023-06-26
| | | | | | | It was not clear what the "..." meant. Author: Steve Chavez Discussion: https://www.postgresql.org/message-id/CAGRrpzZzeh7zC3yaVG9di%3DydJ%2BiHwdXnFPB3evGFKvC1zf6ajA@mail.gmail.com
* Check for interrupts and stack overflow in TParserGet().Tom Lane2023-06-24
| | | | | | | | | | | | | | | TParserGet() recurses for some token types, meaning it's possible to drive it to stack overflow. Since this is a minority behavior, I chose to add the check_stack_depth() call to the two places that recurse rather than doing it during every single call. While at it, add CHECK_FOR_INTERRUPTS(), because this can run unpleasantly long for long inputs. Per bug #17995 from Zuming Jiang. This is old, so back-patch to all supported branches. Discussion: https://postgr.es/m/17995-9f20ff3e6389db4c@postgresql.org
* Error message refactoringPeter Eisentraut2023-06-23
| | | | | | Take some untranslatable things out of the message and replace by format placeholders, to reduce translatable strings and reduce translation mistakes.
* Fix incorrect error message in libpq_pipelineMichael Paquier2023-06-23
| | | | | | | | | | One of the tests for the pipeline mode with portal description expects a non-NULL PQgetResult, but used an incorrect error message on failure, telling that PQgetResult being NULL was the expected result. Author: Jelte Fennema Discussion: https://postgr.es/m/CAGECzQTkShHecFF+EZrm94Lbsu2ej569T=bz+PjMbw9Aiioxuw@mail.gmail.com Backpatch-through: 14
* Fix cache lookup hazards introduced by ff9618e82a.Nathan Bossart2023-06-22
| | | | | | | | | | | | | | | | | | | | | | | | | ff9618e82a introduced has_partition_ancestor_privs(), which is used to check whether a user has MAINTAIN on any partition ancestors. This involves syscache lookups, and presently this function does not take any relation locks, so it is likely subject to the same kind of cache lookup failures that were fixed by 19de0ab23c. To fix this problem, this commit partially reverts ff9618e82a. Specifically, it removes the partition-related changes, including the has_partition_ancestor_privs() function mentioned above. This means that MAINTAIN on a partitioned table is no longer sufficient to perform maintenance commands on its partitions. This is more like how privileges for maintenance commands work on supported versions. Privileges are checked for each partition, so a command that flows down to all partitions might refuse to process them (e.g., if the current user doesn't have MAINTAIN on the partition). In passing, adjust a few related comments and error messages, and add a test for the privilege checks for CLUSTER on a partitioned table. Reviewed-by: Michael Paquier, Jeff Davis Discussion: https://postgr.es/m/20230613211246.GA219055%40nathanxps13
generated by cgit v1.2.3 (git 2.25.1) at 2025-07-15 14:29:58 +0000