summaryrefslogtreecommitdiff
path: root/src/backend/access
AgeCommit message (Expand)Author
2006-05-29Som improve page split in multicolumn GiST index.Teodor Sigaev
2006-05-29Correct cheking in findParents(). iTeodor Sigaev
2006-05-28Remove traces of otherwise unused RELKIND_SPECIAL symbol. Leave the psql bitsAlvaro Herrera
2006-05-26Fix findParents() in case of multiple levels to find.Teodor Sigaev
2006-05-24* Add support NULL to GiST.Teodor Sigaev
2006-05-19Call MarkBufferDirty() before XLogInsert() during completion of insertTeodor Sigaev
2006-05-19Simplify gistSplit() and some refactoring related code.Teodor Sigaev
2006-05-19Rework completion of incomplete inserts. Now it writesTeodor Sigaev
2006-05-17Reduce size of critial section during vacuum full, criticalTeodor Sigaev
2006-05-10Clean up code associated with updating pg_class statistics columnsTom Lane
2006-05-10Reduce size of critical section and remove call of user-defined functions inTeodor Sigaev
2006-05-08Rewrite btree vacuuming to fold the former bulkdelete and cleanup operationsTom Lane
2006-05-07Rewrite btree index scans to work a page at a time in all cases (bothTom Lane
2006-05-03Fix typo noticed by Alvaro HerreraTeodor Sigaev
2006-05-02Clean up API for ambulkdelete/amvacuumcleanup as per today's discussion.Tom Lane
2006-05-02Suppress some gcc warnings.Tom Lane
2006-05-02GIN: Generalized Inverted iNdex.Teodor Sigaev
2006-04-25Arrange to cache btree metapage data in the relcache entry for the index,Tom Lane
2006-04-25Add statement_timestamp(), clock_timestamp(), andBruce Momjian
2006-04-20Ensure that we validate the page header of the first page of a WAL fileTom Lane
2006-04-17Fix the torn-page hazard for PITR base backups by forcing full page writesTom Lane
2006-04-14Make the world safe for full_page_writes. Allow XLOG records that try toTom Lane
2006-04-13Fix an ancient oversight in btree xlog replay. When trying to determine if anTom Lane
2006-04-05Fix a bunch of problems with domains by making them use special input functionsTom Lane
2006-04-05Add a field to the first page of each WAL file to indicate theTom Lane
2006-04-04Don't use BLCKSZ for the physical length of the pg_control file, butTom Lane
2006-04-04Modify all callers of datatype input and receive functions so that if theseTom Lane
2006-04-03Define a separately configurable XLOG_BLCKSZ symbol for the page sizeTom Lane
2006-04-03Fix thinko in gistRedoPageUpdateRecord: if XLR_BKP_BLOCK_1 is set, weTom Lane
2006-04-03Eliminate ajust scan code. Since concurrent GiST it doesn'tTeodor Sigaev
2006-04-01Remove the 'slow' path for btree index build, which built the btreeTom Lane
2006-03-31Clean up WAL/buffer interactions as per my recent proposal. Get rid of theTom Lane
2006-03-30Improve gist XLOG code to follow the coding rules needed to preventTom Lane
2006-03-29Clean up and document the API for XLogOpenRelation and XLogReadBuffer.Tom Lane
2006-03-28Disable full_page_writes, because turning it off risks causing crash-recoveryTom Lane
2006-03-28Repair longstanding error in btree xlog replay: XLogReadBuffer should beTom Lane
2006-03-24Arrange to emit a description of the current XLOG record as error contextTom Lane
2006-03-16Clean up representation of function RTEs for functions returning RECORD.Tom Lane
2006-03-14Improve parser so that we can show an error cursor position for errorsTom Lane
2006-03-10Add a CHECK_FOR_INTERRUPTS() in _bt_buildadd(). This fixes problemTom Lane
2006-03-05Update copyright for 2006. Update scripts.Bruce Momjian
2006-03-01This patch makes the error message strings throughout the backendNeil Conway
2006-02-21Cleanup the usage of ScanDirection: use the symbolic names for theNeil Conway
2006-02-14Move btbulkdelete's vacuum_delay_point() call to a place in the loop whereTom Lane
2006-02-14Add some missing vacuum_delay_point calls in GIST vacuuming.Tom Lane
2006-02-12Actually there's a better way to do this, which is to count tuplesTom Lane
2006-02-11Skip ambulkdelete scan if there's nothing to delete and the index is notTom Lane
2006-02-11Revert based on Tom's recommendation:Bruce Momjian
2006-02-11Allow VACUUM to complete faster by avoiding scanning the indexes when noBruce Momjian
2006-01-25Remove the no-longer-useful HashItem/HashItemData level of structure.Tom Lane