Home - Waterfall Grid T-Grid Console Builders Recent Builds Buildslaves Changesources - JSON API - About

Console View


Categories: connectors experimental galera main
Legend:   Passed Failed Warnings Failed Again Running Exception Offline No data

connectors experimental galera main
fb47ec835c4e...
fb47ec835c4e48c181b5e4402dc3846b65496240
Jan Lindström
MDEV-41171 I_S.TABLES fast path does not count examined rows

The fast path of MDEV-31342 sent rows but did not update the
examined row count. The slow query log showed rows_examined=0 for a
query that read three rows. Count each row that the fast path reads,
as the regular path does for the rows it reads from the temporary
table.

The test also covers Rows_sent, Empty_queries and the slow log.
It adds the ANALYZE SELECT case of MDEV-41165.

Assisted-by: Claude:claude-4.6-sonnet
f9ccc775c90b...
f9ccc775c90b056f747ef01386f84e7050add201
Jan Lindström
MDEV-41171 I_S.TABLES fast path does not count examined rows

The fast path of MDEV-31342 sent rows but did not update the
examined row count. The slow query log showed rows_examined=0 for a
query that read three rows. Count each row that the fast path reads,
as the regular path does for the rows it reads from the temporary
table.

The test also covers Rows_sent, Empty_queries and the slow log.
It adds the ANALYZE SELECT case of MDEV-41165.

Assisted-by: Claude:claude-4.6-sonnet
f98ab907e884...
f98ab907e884e064eb117964c2d8e6b535ccd32f
mariadb-PranavTiwari
MDEV-41137 Add CHARACTER SET ANY_CS and CHARSET OF / %CHARSET

Stored routines:

- CHARACTER SET ANY_CS: a routine parameter takes the character set and
  collation of the actual argument at call time, without conversion.
  Allowed on procedure and function parameters only.
- CHARSET OF var (var%CHARSET in sql_mode=ORACLE): a parameter or variable
  takes the character set and collation of a previous parameter or
  variable. Resolved at CREATE time if the source has a fixed character
  set, at call time if it is ANY_CS or anchored to one.
- information_schema.parameters reports NULL character set and collation
  for ANY_CS and call-time anchors.

New tests: main.sp-charset-any_cs, compat/oracle.sp-charset-any_cs.
f7efdec24fcf...
f7efdec24fcf796e9c0e4120ae39de66a2f3fc9e
Georg Richter
MDEV-42145: Enable crl/crl_path for wolfssl/gnutls

- Fixed abort if a crl or crl_path was passed to a server which was built
  with WolfSSL
- Enable tests for all configurations
e1b4f63b099f...
e1b4f63b099fad09649878e168f110a18253e68a
Vladislav Vaintroub
MDEV-42164 mariadb-import --use-threads can kill its own connection

When a worker fails, kill_tp_connections() sent mysql_kill() for every
worker connection over the failing worker's own connection. It could kill
itself first, and the remaining kills then failed on the dead connection,
so the other workers were not killed and the server kept running their
LOAD DATA after the client exited.

Skip the own connection when killing, and close it afterwards.

Assisted-by: Claude:claude-sonnet-5-5
da3fa5f96dd2...
da3fa5f96dd229c594031c379203e5776f517df8
Yuchen Pei
MDEV-25848 one range slot per ARRAY index

Before, every MEMBER OF / JSON_OVERLAPS / JSON_CONTAINS point got its
own cloned KEY_PART and key slot, and points were combined only through
index_merge (OR) or by picking one slot (AND).

Now all points on an ARRAY key part share the original slot:

* SEL_ARG_ARRAY holds a list of make_key() points ("the array contains
  any of them") and the summed records_in_range() estimate and cost,
  taken once when the point is built.
* key_or() of two ARRAY ranges takes the union of the points;
  key_and() keeps the operand with the lower estimate, the WHERE
  condition checks the rest.
* KEY_PART cloning (get_claimed_key_part()) is now RTREE-only, so
  ARRAY-only queries size SEL_TREE::keys by param->keys again.
* QUICK_RANGE_SELECT_ARRAY gets one QUICK_RANGE per point and passes
  all points to hlindex_read_first(); json_index reads them in turn
  and skips trefs it has already returned.
* check_quick_select_array() uses the stored estimate, and rejects
  non-ARRAY SEL_ARGs (e.g. IS NULL) on the ARRAY key part.

ORed MEMBER OF and JSON_OVERLAPS now give a single range scan instead
of index_merge sort_union, with no MAX_KEY limit on the number of
points.
d55d11e46a96...
d55d11e46a961b6b0f451ef9aa96845b97fab560
Jan Lindström
MDEV-41165: use I_S fast path with max_statement_time

Remove the max_statement_time fallback from is_simple_is_query().
The fast path now runs with a statement timer set.

In get_schema_tables_result(), add the debug sync point
i_s_fast_path_before_eof before EOF is sent. If the query was
killed or timed out after the WHERE condition was evaluated,
send the kill message instead of EOF.

Test changes:
- mdev_31342: rename the max_statement_time case, expect
  is_fast_path in EXPLAIN FORMAT=JSON, add the End of 13.2 tests
  marker.
- mdev_31342_kill (new): KILL QUERY and max_statement_time expiry
  after the scan, before EOF, plus a Handler_tmp_write check that
  the fast path is used.

Co-authored-by: Oleksandr Byelkin <[email protected]>
cdc144c215e7...
cdc144c215e75e7c2222085432b669b70a8baaa2
Yuchen Pei
MDEV-25848 bound range optimizer work for many ARRAY index points

With one range slot per ARRAY index, the number of points in a
SEL_ARG_ARRAY is no longer bounded by MAX_KEY, and a JSON_OVERLAPS or
JSON_CONTAINS constant with N elements cost (see the TODOs):

* N records_in_range() index dives, one per leaf, even when the
  estimate is never used;
* O(N^2) work in key_or_array(), which copied both point lists into a
  new SEL_ARG_ARRAY on every OR and deduplicated with a linear scan.

Now:

* key_or_array() appends the smaller point list to the larger
  SEL_ARG_ARRAY in place, or to a copy of it if it is shared
  (use_count > 0), without sorting or deduplicating.
* SEL_ARG_ARRAY::normalize() sorts and deduplicates the points once,
  when they are needed: estimates, quick select, optimizer trace.
* SEL_ARG_ARRAY::estimate() computes rows and cost lazily, on first
  use (key_and() or check_quick_select_array()). With
  eq_range_index_dive_limit or more points it dives on that many
  points spread over the sorted list and extrapolates.
* JSON_CONTAINS skips duplicate elements of the constant and uses at
  most eq_range_index_dive_limit distinct ones for the range, since
  each AND costs a dive and fewer conjuncts give a superset.

Points are now read in sorted order, so the order of rows from an
ARRAY index range scan may differ.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
c857a7a57b2a...
c857a7a57b2a359c7a63a28598787998fb36fad2
Jan Lindström
MDEV-41171 I_S.TABLES fast path ignores ROWNUM and ROWS EXAMINED

The fast path of MDEV-31342 streams rows of a simple SELECT on
INFORMATION_SCHEMA.TABLES. ROWNUM() in the WHERE clause and LIMIT
ROWS EXAMINED are handled by the join. The fast path bypassed both
and returned every row without a warning.

Do not use the fast path for a query with ROWNUM() or with LIMIT
ROWS EXAMINED.

The test also covers @@sql_select_limit and FOUND_ROWS().

Assisted-by: Claude:claude-4.6-sonnet
c5272f132d38...
c5272f132d3897f596fd67d985aceba2aa00d1ee
Dave Gosselin
MDEV-38643: ST_GeomFromText marks its SELECT with UNCACHEABLE_RAND

This patch removes the flag from ST_GeomFromText, ST_GeomFromWKB,
ST_GeomFromGeoJSON, ST_AsGeoJSON.

When called with one argument, these functions set UNCACHEABLE_RAND on
the enclosing SELECT.  In contrast, the forms with more arguments
never set that flag.

The flag was added in 2006 by commit 384407aaeb3, which moved the
function parsing out of sql_yacc.yy.  The prior grammar rules did not
set the flag for these, either.  The new code has the same form as the
code for ENCRYPT, which requires the UNCACHEABLE_RAND flag as its one
argument variant takes its salt from the clock.
ba76f3c07463...
ba76f3c07463408c8e2e2c70318ce675361c3be9
Vladislav Vaintroub
MDEV-38741 mtr: do not suppress the reason of a client disconnect

The global suppressions of "Aborted connection", in mariadb-test-run.pl
and mtr_warnings.sql, hid the line that the client disconnect watcher
writes when it aborts a command: "Aborted connection N to db: 'x' user:
'u' host: 'h' (Client disconnected)". Nothing showed when the watcher
aborted a connection that it should have left alone. main.locale found
one, because its server log is in German and the English entry did not
match.

Make the two entries skip a line that ends in "(Client disconnected)".
Other aborted connections are suppressed as before.

The tests that drop a client on purpose accept the reason:
have_disconnect_watcher.inc, which the tests of the watcher include,
and dirty_close, loaddata, sp_notembedded, ssl_timeout, mysqltest,
innodb_bug38231, innodb.xa_recovery, mysql_client_test,
mysql_client_test_nonblock, mysql_client_test_comp,
rpl_semi_sync_shutdown_await_ack, spider/bugfix.mdev_35959 (a Spider
link that times out), main.mysqldump (mariadb-import can leave a LOAD DATA
running when it fails, MDEV-42164) and
mariabackup.nolock_ddl_during_backup_end (mariabackup ends in the middle
of a backup). The Connector/C tests of the unit suite close connections
on purpose, but cannot call mtr.add_suppression(), so suite.pm adds it
before each of them.

A replica closes its connection on purpose when it is stopped, in any
phase of its setup, also while the master runs one of its first
queries. rpl_init.inc accepts the reason for every server of the
replication tests, without writing it to the binlog. A binlog dump
thread logs its disconnect with another message, and only with
log_warnings=4.

Assisted-by: Claude:claude-sonnet-5-5
b0e522e369e9...
b0e522e369e97889b9f3c8b02f7475b333dc83e9
Yuchen Pei
MDEV-25848 Allow "js IS NOT NULL AND <JSON_SARGABLE_PREDICATE on js>" to use range access
ad46266ac259...
ad46266ac259471c68816704ca6a6430f0d63723
Jan Lindström
MDEV-41171 I_S.TABLES fast path ignores ROWNUM and ROWS EXAMINED

The fast path of MDEV-31342 streams rows of a simple SELECT on
INFORMATION_SCHEMA.TABLES. ROWNUM() in the WHERE clause and LIMIT
ROWS EXAMINED are handled by the join. The fast path bypassed both
and returned every row without a warning.

Do not use the fast path for a query with ROWNUM() or with LIMIT
ROWS EXAMINED.

The test also covers @@sql_select_limit and FOUND_ROWS().

Assisted-by: Claude:claude-4.6-sonnet
ab10bb57a9ff...
ab10bb57a9ff63b01b994e790e6d9ff78a4e685a
Jan Lindström
MDEV-42139 fix spurious wakeup in gu_debug_sync_wait()

In galera_3nodes.galera_join_with_cc_X this let a waiter on the
"process_primary_configuration" sync point disappear from
wsrep_debug_sync_waiters between a wait_condition.inc poll and the
test's own follow-up SELECT, producing an empty result instead of the
point name.

Test now requires Galera library 26.4.29 where fix for
spurious wakeup in gu_debug_sync_wait() is.
aaa41aaabfe6...
aaa41aaabfe6c9e4fb0cfd474cf3d6343ab6e1ca
Hemant Dangi
MDEV-42122: check_port() must verify the listener is our own

Issue:
wait_for_listen() asks check_port() whether a process named
socat/nc/netcat is listening on the expected port, but passes no PID
to match against, so check_port() accepts any process with that name
anywhere on the system. On a busy host running many concurrent SSTs,
an unrelated process can transiently hold the same port number, and
wait_for_listen() then reports "ready" even though this invocation's
own listener never came up.

Solution:
When check_port() isn't given a specific PID to match, verify that a
candidate process found by name and port is this script's own
process or a descendant of it, rather than accepting any same-named
process. Existing callers that already pass a known PID are
unaffected.
a218d24ac2c5...
a218d24ac2c5546b29d909055d6cedfdf187a21d
Yuchen Pei
MDEV-25848 one range slot per ARRAY index

Before, every MEMBER OF / JSON_OVERLAPS / JSON_CONTAINS point got its
own cloned KEY_PART and key slot, and points were combined only through
index_merge (OR) or by picking one slot (AND).

Now all points on an ARRAY key part share the original slot:

* SEL_ARG_ARRAY holds a list of make_key() points ("the array contains
  any of them") and the summed records_in_range() estimate and cost,
  taken once when the point is built.
* key_or() of two ARRAY ranges takes the union of the points;
  key_and() keeps the operand with the lower estimate, the WHERE
  condition checks the rest.
* KEY_PART cloning (get_claimed_key_part()) is now RTREE-only, so
  ARRAY-only queries size SEL_TREE::keys by param->keys again.
* QUICK_RANGE_SELECT_ARRAY gets one QUICK_RANGE per point and passes
  all points to hlindex_read_first(); json_index reads them in turn
  and skips trefs it has already returned.
* check_quick_select_array() uses the stored estimate, and rejects
  non-ARRAY SEL_ARGs (e.g. IS NULL) on the ARRAY key part.

ORed MEMBER OF and JSON_OVERLAPS now give a single range scan instead
of index_merge sort_union, with no MAX_KEY limit on the number of
points.
9712fc8e22e8...
9712fc8e22e8f768f47502a27d277a8b0a8e11e6
Yuchen Pei
MDEV-25848 bound range optimizer work for many ARRAY index points

With one range slot per ARRAY index, the number of points in a
SEL_ARG_ARRAY is no longer bounded by MAX_KEY, and a JSON_OVERLAPS or
JSON_CONTAINS constant with N elements cost (see the TODOs):

* N records_in_range() index dives, one per leaf, even when the
  estimate is never used;
* O(N^2) work in key_or_array(), which copied both point lists into a
  new SEL_ARG_ARRAY on every OR and deduplicated with a linear scan.

Now:

* key_or_array() appends the smaller point list to the larger
  SEL_ARG_ARRAY in place, or to a copy of it if it is shared
  (use_count > 0), without sorting or deduplicating.
* SEL_ARG_ARRAY::normalize() sorts and deduplicates the points once,
  when they are needed: estimates, quick select, optimizer trace.
* SEL_ARG_ARRAY::estimate() computes rows and cost lazily, on first
  use (key_and() or check_quick_select_array()). With
  eq_range_index_dive_limit or more points it dives on that many
  points spread over the sorted list and extrapolates.
* JSON_CONTAINS skips duplicate elements of the constant and uses at
  most eq_range_index_dive_limit distinct ones for the range, since
  each AND costs a dive and fewer conjuncts give a superset.

Points are now read in sorted order, so the order of rows from an
ARRAY index range scan may differ.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

MDEV-25848 deduplicate JSON_OVERLAPS/JSON_CONTAINS elements when parsing

Skip duplicate elements of the constant in json_get_range_elements(),
using a hash on the element's JSON text, instead of in
json_get_func_mm_tree() for JSON_CONTAINS only. Duplicates now never
become Items, SEL_ARG_ARRAY leaves or make_key() calls, for
JSON_OVERLAPS as well, and json_get_func_mm_tree() only counts the
distinct elements against eq_range_index_dive_limit.

The text includes the quotes of strings, so "1" and 1 are different.
Equal values with different text, such as 1 and 1.0, are still
deduplicated later by SEL_ARG_ARRAY::normalize().

Co-Authored-By: Claude Opus 5.5 <[email protected]>
931c1e80568b...
931c1e80568bd6c5ca4f5af6a22caaed64c39110
Mohammad Tafzeel Shams
MDEV-39647: Deleted Virtual secondary records might never get purged
            and cause table bloating

Virtual column values are not stored in the clustered index. To check
whether a secondary index record that includes a virtual column can be
purged, row_purge_vc_matches_cluster() and row_vers_build_cur_vrow_low()
read those values from the undo log by invoking
trx_undo_prev_version_build() with TRX_UNDO_PREV_IN_PURGE.

Since MDEV-24402, this read was skipped if purge_sys.end_view saw the
transaction. That happens as soon as a purge batch has ended inside the
undo log of the transaction, even though the remaining undo log records
of the transaction have not been freed. With no values to compare,
row_purge_vc_matches_cluster() assumed a match, row_purge_is_unsafe()
returned true, and the delete-marked secondary index record was never
removed. CHECK TABLE...EXTENDED reported "Clustered index record not
found".

Changes:

- trx_undo_prev_version_build(): For TRX_UNDO_PREV_IN_PURGE, always read
the undo log record, using purge_sys_t::view_guard::PURGE. Only purge
tasks set this flag. The callers never read past the undo log record
being purged, and purge only frees undo log records before purge_sys.head.

- innodb.purge_secondary updates an indexed virtual column several
times in one transaction, with innodb_purge_batch_size=1 so that its
undo log is purged across several batches. CHECK TABLE...EXTENDED
must not report any leftover secondary index records.
80080e1c3580...
80080e1c3580f9dfe379b31c20e169d7508ca350
Pekka Lampio
MDEV-38999: Galera - stale reads in explicit transactions

wsrep_must_sync_wait() skipped a statement's causal wait unconditionally
once a multi-statement transaction was open, regardless of which
wsrep_sync_wait bit was configured - so later statements could run
against a view no fresher than whatever BEGIN itself happened to check.
A timed-out wait's error also stayed sticky until commit/rollback,
breaking ROLLBACK itself.

Fix: only skip the wait once the transaction has made its first write or
locking read (SELECT ... FOR UPDATE/LOCK IN SHARE MODE, including via a
stored procedure CALL, autocommit statements, and UNION) - after that,
re-waiting risks a self-deadlock against the local applier. Clear a
timed-out wait's error immediately instead of leaving it sticky.

Tests: galera_MDEV-38999*.test cover the bitmask gap, the self-deadlock
risk, stored-procedure/UNION/autocommit locking reads, and the query
cache's independent wait path.

Assisted-by Claude AI
78082434df48...
78082434df4847afe7a41761342ce26e8b05ef9a
Jan Lindström
MDEV-41165 Assertion in set_eof_status in I_S.TABLES fast path

The fast path of MDEV-31342 streams rows for a simple SELECT on
INFORMATION_SCHEMA.TABLES. Evaluating the WHERE clause can fail
after the statement is killed. SLEEP() sets ER_QUERY_INTERRUPTED
or ER_STATEMENT_TIMEOUT. get_schema_tables_result() then called
my_eof() with the error already set. This failed the assertion in
Diagnostics_area::set_eof_status().

Do not call my_eof() when the statement has an error. The existing
error handling after the loop reports the error once.

Do not use the fast path when max_statement_time is set.

Tests cover SET STATEMENT max_statement_time, KILL QUERY and KILL
CONNECTION.

Assisted-by: Claude:claude-4.6-sonnet
73dee65939d7...
73dee65939d7549f797368f0fb198b8d7f1a7b6a
rusher
Merge branch '3.4' into 3.4-perf
7153e1304e1e...
7153e1304e1e2d873e6b5f41baaf6dc6255f02bc
Georgi (Joro) Kodinov
MDEV-42175: Do not compile 32-bit instructions for wolfSSL on 64bit platforms.

WolfSSL doesn't need some 32 bit code if it detects proper 64 bit mode.
Added a check for __uint128_t and made the 32 bit file conditional on this.
See https://github.com/wolfSSL/wolfssl/issues/11874 for details.
69e7c616f872...
69e7c616f87288bbc9a9d35701792e8b32e9240c
Yuchen Pei
MDEV-25848 bound range optimizer work for many ARRAY index points

With one range slot per ARRAY index, the number of points in a
SEL_ARG_ARRAY is no longer bounded by MAX_KEY, and a JSON_OVERLAPS or
JSON_CONTAINS constant with N elements cost (see the TODOs):

* N records_in_range() index dives, one per leaf, even when the
  estimate is never used;
* O(N^2) work in key_or_array(), which copied both point lists into a
  new SEL_ARG_ARRAY on every OR and deduplicated with a linear scan.

Now:

* key_or_array() appends the smaller point list to the larger
  SEL_ARG_ARRAY in place, or to a copy of it if it is shared
  (use_count > 0), without sorting or deduplicating.
* SEL_ARG_ARRAY::normalize() sorts and deduplicates the points once,
  when they are needed: estimates, quick select, optimizer trace.
* SEL_ARG_ARRAY::estimate() computes rows and cost lazily, on first
  use (key_and() or check_quick_select_array()). With
  eq_range_index_dive_limit or more points it dives on that many
  points spread over the sorted list and extrapolates.
* JSON_CONTAINS skips duplicate elements of the constant and uses at
  most eq_range_index_dive_limit distinct ones for the range, since
  each AND costs a dive and fewer conjuncts give a superset.

Points are now read in sorted order, so the order of rows from an
ARRAY index range scan may differ.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
68263c2df543...
68263c2df543b02e5893e2855defbd96bc7a1fd3
Alexey (Holyfoot) Botchkov
MDEV-41821 XMLISVALID returns unexepected result, various corner cases.

Control of empty builtin types added.
complexType with only optional particles allows empty element.
maxOccurs now handled in <choice>.
No type in <element> now is allowed.
6388bebe7cf2...
6388bebe7cf20e2f7749e0daacfc395a6fbeb068
Jan Lindström
MDEV-41165: use I_S fast path with max_statement_time

Remove the max_statement_time fallback from is_simple_is_query().
The fast path now runs with a statement timer set.

In get_schema_tables_result(), add the debug sync point
i_s_fast_path_before_eof before EOF is sent. If the query was
killed or timed out after the WHERE condition was evaluated,
send the kill message instead of EOF.

Test changes:
- mdev_31342: rename the max_statement_time case, expect
  is_fast_path in EXPLAIN FORMAT=JSON, add the End of 13.2 tests
  marker.
- mdev_31342_kill (new): KILL QUERY and max_statement_time expiry
  after the scan, before EOF, plus a Handler_tmp_write check that
  the fast path is used.

Co-authored-by: Oleksandr Byelkin <[email protected]>
61d46089cacf...
61d46089cacfdb1213acb7d7973644f001bee01b
rusher
[misc] Move the error-state test into CLEAR_CLIENT_ERROR and CLEAR_CLIENT_STMT_ERROR
618ffe2e01a9...
618ffe2e01a9bd5eeefa69515c392ab3f2b4f622
Hemant Dangi
Merge branch '10.11' into 10.11-MDEV-42122
5ae71bae6ce9...
5ae71bae6ce9423c9c2757b581b85528c8fec117
Vladislav Vaintroub
MDEV-42164 mariadb-import --use-threads can kill its own connection

When a worker fails, kill_tp_connections() sent mysql_kill() for every
worker connection over the failing worker's own connection. It could kill
itself first, and the remaining kills then failed on the dead connection,
so the other workers were not killed and the server kept running their
LOAD DATA after the client exited.

Skip the own connection when killing, and close it afterwards.
The test counts the KILL commands the server receives.

Assisted-by: Claude:claude-sonnet-5-5
5370405937a6...
5370405937a659351b9203337e76e4597cfb6eb0
Alexey (Holyfoot) Botchkov
MDEV-36261 XMLTYPE: methods - step 1

Method functions added to the XMLTYPE.
  functions added here are:

  FUNCTION XMLTYPE.createNonSchemaBasedXML() RETURN XMLType
  FUNCTION XMLTYPE.getClobVal() RETURN LONGTEXT
  FUNCTION XMLTYPE.getNamespace() RETURN LONGTEXT
  FUNCTION XMLTYPE.isFragment() RETURN INTEGER
  FUNCTION XMLTYPE.getNumberVal() RETURN DOUBLE
  FUNCTION XMLTYPE.getStringVal() RETURN LONGTEXT
  FUNCTION XMLTYPE.getRootElement() RETURN LONGTEXT
51c022c0063b...
51c022c0063b1d412d6d6a1b9c3f6595ff27d15b
mariadb-PranavTiwari
d
4abb8707d46c...
4abb8707d46ca81f2780a982cad74297a3b65f32
Yuchen Pei
MDEV-25848 bound range optimizer work for many ARRAY index points

With one range slot per ARRAY index, the number of points in a
SEL_ARG_ARRAY is no longer bounded by MAX_KEY, and a JSON_OVERLAPS or
JSON_CONTAINS constant with N elements cost (see the TODOs):

* N records_in_range() index dives, one per leaf, even when the
  estimate is never used;
* O(N^2) work in key_or_array(), which copied both point lists into a
  new SEL_ARG_ARRAY on every OR and deduplicated with a linear scan.

Now:

* key_or_array() appends the smaller point list to the larger
  SEL_ARG_ARRAY in place, or to a copy of it if it is shared
  (use_count > 0), without sorting or deduplicating.
* SEL_ARG_ARRAY::normalize() sorts and deduplicates the points once,
  when they are needed: estimates, quick select, optimizer trace.
* SEL_ARG_ARRAY::estimate() computes rows and cost lazily, on first
  use (key_and() or check_quick_select_array()). With
  eq_range_index_dive_limit or more points it dives on that many
  points spread over the sorted list and extrapolates.
* JSON_CONTAINS skips duplicate elements of the constant and uses at
  most eq_range_index_dive_limit distinct ones for the range, since
  each AND costs a dive and fewer conjuncts give a superset.

Points are now read in sorted order, so the order of rows from an
ARRAY index range scan may differ.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
441efb75f9f4...
441efb75f9f4d688a2731d9059683ff165b1a798
mariadb-PranavTiwari
Added
40e829509c78...
40e829509c789d7ae404000c81303ffa735f59f1
Hemant Dangi
MDEV-42122: check_port() must verify the listener is our own

Issue:
wait_for_listen() asks check_port() whether a process named
socat/nc/netcat is listening on the expected port, but passes no PID
to match against, so check_port() accepts any process with that name
anywhere on the system. On a busy host running many concurrent SSTs,
an unrelated process can transiently hold the same port number, and
wait_for_listen() then reports "ready" even though this invocation's
own listener never came up.

Solution:
When check_port() isn't given a specific PID to match, verify that a
candidate process found by name and port is this script's own
process or a descendant of it, rather than accepting any same-named
process. Existing callers that already pass a known PID are
unaffected.
2664ea18b8eb...
2664ea18b8eb7023b4334e904b9bb71a80230d0f
Yuchen Pei
MDEV-25848 one range slot per ARRAY index

Before, every MEMBER OF / JSON_OVERLAPS / JSON_CONTAINS point got its
own cloned KEY_PART and key slot, and points were combined only through
index_merge (OR) or by picking one slot (AND).

Now all points on an ARRAY key part share the original slot:

* SEL_ARG_ARRAY holds a list of make_key() points ("the array contains
  any of them") and the summed records_in_range() estimate and cost,
  taken once when the point is built.
* key_or() of two ARRAY ranges takes the union of the points;
  key_and() keeps the operand with the lower estimate, the WHERE
  condition checks the rest.
* KEY_PART cloning (get_claimed_key_part()) is now RTREE-only, so
  ARRAY-only queries size SEL_TREE::keys by param->keys again.
* QUICK_RANGE_SELECT_ARRAY gets one QUICK_RANGE per point and passes
  all points to hlindex_read_first(); json_index reads them in turn
  and skips trefs it has already returned.
* check_quick_select_array() uses the stored estimate, and rejects
  non-ARRAY SEL_ARGs (e.g. IS NULL) on the ARRAY key part.

ORed MEMBER OF and JSON_OVERLAPS now give a single range scan instead
of index_merge sort_union, with no MAX_KEY limit on the number of
points.
250e254bb8d7...
250e254bb8d768733b7387b91e01c29489890876
mariadb-PranavTiwari
MDEV-41137 Add CHARACTER SET ANY_CS and CHARSET OF / %CHARSET
Stored routines:

- CHARACTER SET ANY_CS: a routine parameter takes the character set and
  collation of the actual argument at call time, without conversion.
  Allowed on procedure and function parameters only.
- CHARSET OF var (var%CHARSET in sql_mode=ORACLE): a parameter or variable
  takes the character set and collation of a previous parameter or
  variable. Resolved at CREATE time if the source has a fixed character
  set, at call time if it is ANY_CS or anchored to one.
1c16867dd082...
1c16867dd082214a05cb80476ecd0f7301647d28
Jan Lindström
MDEV-41165 Assertion in set_eof_status in I_S.TABLES fast path

The fast path of MDEV-31342 streams rows for a simple SELECT on
INFORMATION_SCHEMA.TABLES. Evaluating the WHERE clause can fail
after the statement is killed. SLEEP() sets ER_QUERY_INTERRUPTED
or ER_STATEMENT_TIMEOUT. get_schema_tables_result() then called
my_eof() with the error already set. This failed the assertion in
Diagnostics_area::set_eof_status().

Do not call my_eof() when the statement has an error. The existing
error handling after the loop reports the error once.

Do not use the fast path when max_statement_time is set.

Tests cover SET STATEMENT max_statement_time, KILL QUERY and KILL
CONNECTION.

Assisted-by: Claude:claude-4.6-sonnet
10c72dfba883...
10c72dfba8835e4b6f41026644554f43d147e48b
Vladislav Vaintroub
MDEV-42164 mariadb-import --use-threads can kill its own connection

When a worker fails, kill_tp_connections() sent mysql_kill() for every
worker connection over the failing worker's own connection. It could kill
itself first, and the remaining kills then failed on the dead connection,
so the other workers were not killed and the server kept running their
LOAD DATA after the client exited.

Skip the own connection when killing, and close it afterwards.

Assisted-by: Claude:claude-sonnet-5-5
08b63a793c46...
08b63a793c46cac5f18f08193e7a5057259a0f6c
Pekka Lampio
MDEV-38999: Galera - stale reads in explicit transactions

wsrep_must_sync_wait() skipped a statement's causal wait unconditionally
once a multi-statement transaction was open, regardless of which
wsrep_sync_wait bit was configured - so later statements could run
against a view no fresher than whatever BEGIN itself happened to check.
A timed-out wait's error also stayed sticky until commit/rollback,
breaking ROLLBACK itself.

Fix: only skip the wait once the transaction has made its first write or
locking read (SELECT ... FOR UPDATE/LOCK IN SHARE MODE, including via a
stored procedure CALL, autocommit statements, and UNION) - after that,
re-waiting risks a self-deadlock against the local applier. Clear a
timed-out wait's error immediately instead of leaving it sticky.

Tests: galera_MDEV-38999*.test cover the bitmask gap, the self-deadlock
risk, stored-procedure/UNION/autocommit locking reads, and the query
cache's independent wait path.

Assisted-by Claude AI
BuildBot (0.8.9) working for the MariaDB project.
Page built: Tue 10 Nov 2026 11:40:02 (UTC)