TopicMailing ListsActivityParticipants
CI historytests failed
Patchsetv10
Stateapplies
Tests930 / 931
Failedtest_misc/016_subxact_toast_prune
Base0d behind master
toast table corrupted by vacuum - missing chunk number 0 for toast valuehackers
hackers
10 replies
about 24 hours ago
Pavel StehuleTomas VondraBruce MomjianAndrey BorodinPeter Geoghegan
1 commit
Fix GIN posting tree page deletion with incomplete splits.
masterunreleasedREL_19_STABLEunreleasedREL_18_STABLEunreleasedREL_17_STABLEunreleasedREL_16_STABLEunreleasedREL_15_STABLEunreleasedREL_14_STABLEunreleased
CI historyCI passed - final, thread committed
Patchsetv3
Statewon't retry
Resultsuccess
Tests1024 / 1024
Base1d behind master
GIN page deletion and page recycling bugshackers
hackers
2 replies
2 days ago
Peter GeogheganAndrey Borodin
CI historyCI passed
Patchsetv7
Stateapplies
Tests1023 / 1023
Base0d behind master
GiST multirange index scans can fail to return rowshackers
hackers
7 replies
4 days ago
Andrey BorodinPeter GeogheganAlexander KorotkovKirill Reshke
19 commits
aio: io_uring: Trigger async processing for large IOs
master19beta1REL_19_STABLE19beta1
read stream: Split decision about look ahead for AIO and combining
master19beta1REL_19_STABLE19beta1
read_stream: Only increase read-ahead distance when waiting for IO
master19beta1REL_19_STABLE19beta1
heapam: Keep buffer pins across index scan resets.
master19beta1REL_19_STABLE19beta1
heapam: Track heap block in IndexFetchHeapData.
master19beta1REL_19_STABLE19beta1
+14 more
PG20-2Needs review
Patch#4351
TagsPerformance
CICI score: 10/10
ReviewersJerry Zhang (xingtanzjr)
CI historyCI passed
Patchsetv497
Stateapplies
Tests1024 / 1024
Base0d behind master
index prefetchinghackers
hackers
496 replies
4 days ago
Peter GeogheganTomas VondraAndres FreundKonstantin KnizhnikAlexandre Felipe+13
2 commits
Fix GIN multiple-VACUUM-scans pending list bug.
masterunreleasedREL_19_STABLEunreleasedREL_18_STABLEunreleasedREL_17_STABLEunreleasedREL_16_STABLEunreleasedREL_15_STABLEunreleasedREL_14_STABLEunreleased
Fix GIN VACUUM posting tree root split bug.
masterunreleasedREL_19_STABLEunreleasedREL_18_STABLEunreleasedREL_17_STABLEunreleasedREL_16_STABLEunreleasedREL_15_STABLEunreleasedREL_14_STABLEunreleased
CI historyCI passed - final, thread committed
Patchsetv6
Statewon't retry
Resultsuccess
Tests1023 / 1023
Base5d behind master
GIN VACUUM can corrupt internal posting tree pageshackers
hackers
5 replies
5 days ago
Peter GeogheganAndrey Borodin
1 commit
GiST: Invalidate killed items consistently.
masterunreleasedREL_19_STABLEunreleasedREL_18_STABLEunreleasedREL_17_STABLEunreleasedREL_16_STABLEunreleasedREL_15_STABLEunreleasedREL_14_STABLEunreleased
CI historyCI passed - final, thread committed
Patchsetv7
Statewon't retry
Resultsuccess
Tests1021 / 1021
Base5d behind master
gistkillitems can LP_DEAD-mark the wrong tuples on a rescanhackers
hackers
6 replies
5 days ago
Peter GeogheganMichael PaquierAndrey Borodin
3 commits
Track RI fast-path FK-check batches per firing cycle
masterunreleasedREL_19_STABLEunreleased
Fix RI fast-path race with REINDEX CONCURRENTLY
masterunreleasedREL_19_STABLEunreleased
Fix cross-type foreign keys in the batched fast-path FK check
masterunreleasedREL_19_STABLEunreleased
CI historytests failed - final, thread committed
Patchsetv10
Statewon't retry
Resulttests failed
Tests876 / 934
Failedregress/foreign_key, regress/json, regress/jsonb, regress/json_encoding, regress/jsonpath
Base6d behind master
RI fast path gets cross-type foreign keys wronghackers
hackers
11 replies
6 days ago
Amit LangotePeter GeogheganMihail Nikalayeu
1 commit
GiST: Deprecate F_TUPLES_DELETED opaque area flag.
masterunreleasedREL_19_STABLEunreleasedREL_18_STABLEunreleasedREL_17_STABLEunreleasedREL_16_STABLEunreleasedREL_15_STABLEunreleasedREL_14_STABLEunreleased
CI historyCI passed - final, thread committed
Patchsetv3
Statewon't retry
Resultsuccess
Tests1019 / 1019
Base7d behind master
GiST wal_consistency_checking issuehackers
hackers
3 replies
Aug 16, 06:31
Peter GeogheganMichael Paquier
CI historyCI passed
Patchsetv4
Stateapplies
Tests1025 / 1025
Base1d behind master
Hash index bucket split bughackers
hackers
5 replies
Aug 10, 12:02
Peter GeogheganAmit Kapilavignesh C
1 commit
Add ssup_datum_uint32_cmp for comparing OIDs
masterunreleased
CI historytests failed - final, thread committed
Patchsetv1
Statewon't retry
Resulttests failed
Tests932 / 936
Failedregress/oid, test_plan_advice/001_replan_regress, pg_upgrade/002_pg_upgrade, recovery/027_stream_regress
Base16d behind master
btoidsortsupport issuehackers
hackers
7 replies
Aug 10, 10:44
David RowleyPeter GeogheganTom LaneJohn Naylor
1 commit
Fix unlikely incremental tuple deform bug with missing attrs
masterunreleasedREL_19_STABLE19beta3
CI historytests failed - final, thread committed
Patchsetv3
Statewon't retry
Resulttests failed
Tests675 / 676
Failedisolation/alter-table-set-not-null
Base18d behind master
Whole row var issuehackers
hackers
4 replies
Aug 06, 06:03
Peter GeogheganDavid Rowley
2 commits
Fix autovacuum-induced flakiness in backwards scan test.
masterunreleased
Add test coverage for nbtree backwards scans.
masterunreleased
nbtree backwards scan test coveragehackers
hackers
5 replies
Jul 29, 14:47
Peter GeogheganAndrey BorodinAlexander Lakhin
CI historydoes not apply to master
Patchsetv3
Stateneeds rebase
Resultsuccess
Tests1016 / 1016
Base25d behind master
Snapshot export on a standby corrupts hint bits on subxact overflowhackers
hackers
2 replies
Jul 29, 09:36
Bertrand DrouvotPeter Geoghegan
CI historyCI passed
Patchsetv1
Stateapplies
Tests1023 / 1023
Base1d behind master
Simplifying how nbtree represents array scan keyshackers
hackers
0 replies
Jul 26, 19:14
Peter Geoghegan
1 commit
Run nbtree test module tests under autoconf builds
masterunreleasedREL_19_STABLE19beta3
nbtree test module Makefile oversighthackers
hackers
2 replies
Jul 20, 02:16
Peter GeogheganMichael Paquier
1 commit
Add _bt_set_startikey row compare test coverage.
masterunreleasedREL_19_STABLE19beta3
nbtree _bt_set_startikey row compare test coveragehackers
hackers
0 replies
Jul 20, 00:33
Peter Geoghegan
5 commits
Add tests for nbtree empty index predicate locking.
masterunreleased
Fix another empty nbtree index SSI race.
masterunreleasedREL_19_STABLE19beta3REL_18_STABLE18.6REL_17_STABLE17.11REL_16_STABLE16.15REL_15_STABLE15.19REL_14_STABLE14.24
Fix race in SSI interaction with empty btrees.
master17beta1REL_19_STABLE17beta1REL_18_STABLE17beta1REL_17_STABLE17beta1REL_16_STABLE16beta3REL_15_STABLE15.4REL_14_STABLE14.9REL_13_STABLE13.12+2 more
Fix race in SSI interaction with gin fast path.
master17beta1REL_19_STABLE17beta1REL_18_STABLE17beta1REL_17_STABLE17beta1REL_16_STABLE16beta3REL_15_STABLE15.4REL_14_STABLE14.9REL_13_STABLE13.12+2 more
Fix race in SSI interaction with bitmap heap scan.
master17beta1REL_19_STABLE17beta1REL_18_STABLE17beta1REL_17_STABLE17beta1REL_16_STABLE16beta3REL_15_STABLE15.4REL_14_STABLE14.9REL_13_STABLE13.12+2 more
BUG #17949: Adding an index introduces serialisation anomalies.bugs
bugs
27 replies
Jul 19, 19:53
Thomas MunroDmitry DolgovArtem AnisimovPeter GeogheganPG Bug reporting form+1
1 commit
Fix GiST index-only scan column alignment issue.
masterunreleasedREL_19_STABLE19beta3REL_18_STABLE18.6REL_17_STABLE17.11REL_16_STABLE16.15REL_15_STABLE15.19REL_14_STABLE14.24
Bug: GiST index-only scans can use tuple descriptor with incorrect alignment when deforminghackers
hackers
3 replies
Jul 17, 19:56
Peter GeogheganTomas Vondra
1 commit
Use fake LSNs consistently in hash index AM.
masterunreleasedREL_19_STABLE19beta3
Hash index AM fake LSN oversighthackers
hackers
2 replies
Jul 16, 22:11
Peter GeogheganEwan Young
CI historyretired, base too old
Patchsetv2
Statewon't retry
Resultsuccess
Tests1005 / 1005
Base3mo behind master
TID recycling race during nbtree index-only scans that run on a standbyhackers
hackers
2 replies
Jul 07, 18:32
Peter GeogheganMatthias van de Meent
1 commit
Do not lock in BufferGetLSNAtomic() on archs with 8 byte atomic reads
master19beta1REL_19_STABLE19beta1
PG20-1Committed
Patch#6334
TagsPerformance
CINeeds rebase
CommitterTomas Vondra (fuzzycz)
Remove header lock BufferGetLSNAtomic() on architectures with 64 bit atomic operationshackers
hackers
11 replies
Mar 11, 13:37
Andres FreundAndreas KarlssonTomas VondraPeter Geoghegan
1 commit
Standardize cleanup lock terminology.
master15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
PG20-1Withdrawn
Patch#5542
TagsBugfix
CINeeds rebase
CI historyretired, base too old
Patchsetv28
Statewon't retry
Resultsuccess
Tests937 / 937
Base8mo behind master
Why doesn't GiST VACUUM require a super-exclusive lock, like nbtree VACUUM?hackers
hackers
27 replies
Dec 24, 2025
Matthias van de MeentPeter GeogheganMihail NikalayeuAndrey BorodinHeikki Linnakangas+1
2 commits
Clarify why _bt_killitems sorts its items array.
master19beta1REL_19_STABLE19beta1
Return TIDs in desc order during backwards scans.
master19beta1REL_19_STABLE19beta1
PG19-4Committed
Patch#5824
TagsPerformance
CICI score: 10/10
ReviewersMircea Cadariu (mcadariu)
CommitterPeter Geoghegan (pgeoghegan)
Returning nbtree posting list TIDs in DESC order during backwards scanshackers
hackers
16 replies
Dec 11, 2025
Peter GeogheganMircea CadariuChao LiVictor YegorovAndy Fan
2 commits
Avoid pointer chasing in _bt_readpage inner loop.
master19beta1REL_19_STABLE19beta1
Relocate _bt_readpage and related functions.
master19beta1REL_19_STABLE19beta1
PG19-4Committed
Patch#6286
TagsRefactoring Only
CICI score: 10/10
CommitterPeter Geoghegan (pgeoghegan)
Moving _bt_readpage and _bt_checkkeys into a new .c filehackers
hackers
7 replies
Dec 08, 2025
Peter GeogheganVictor YegorovHeikki Linnakangas
CI historyretired, base too old
Patchsetv1
Statewon't retry
Resultsuccess
Tests896 / 896
Base1y behind master
Removing BTScanPosUnpinIfPinned idiom from nbtree, simplifying mark/restore supporthackers
hackers
1 reply
Dec 03, 2025
Peter Geoghegan
Loading more results...