aboutsummaryrefslogtreecommitdiffstats
path: root/searchcore/src/tests/proton/documentdb/document_subdbs/document_subdbs_test.cpp
Commit message (Collapse)AuthorAgeFilesLines
...
* Test that up -> retired -> up is reflectd correctly in document subdbs.Henning Baldersheim2021-11-121-1/+24
|
* Update 2017 copyright notices.gjoranv2021-10-071-1/+1
|
* Reduce visibility and do not provide syncable when not necessary.Henning Baldersheim2021-03-021-0/+4
|
* use size literals in searchcoreArne Juul2021-02-151-2/+3
|
* When making callbacks like notifying about new bucket show the gurad that is ↵Henning Baldersheim2021-02-141-13/+13
| | | | | | held. Then this guard can used instead of possibly making a deadlock if trying to take it yourself.
* Improve handling of changed alloc config.Tor Egge2021-02-021-2/+1
|
* Wire in config for compaction strategy.Tor Egge2021-02-011-1/+1
|
* Remove the syncFeedView upcall as it has no purpose other than forcing it ↵Henning Baldersheim2021-01-261-29/+0
| | | | | | more than necessary. Instead do it in the main documentdb after all subds are reconfigured.
* Move idestructorcallback.h to vespalib module and ↵Henning Baldersheim2021-01-141-2/+2
| | | | search::IDestructorCallback to vespalib::IdestructorCallback.
* Remove the ICommitable interface as it is now void.Henning Baldersheim2020-10-301-14/+1
| | | | Wait happens directly og the LidCommitState interface.
* GC unused commit()Henning Baldersheim2020-10-291-3/+1
|
* Drop immediate commit, as it is gone.Henning Baldersheim2020-10-231-5/+14
|
* - GC unused code.Henning Baldersheim2020-10-071-1/+0
| | | | - vespalib::Lock -> std::mutex
* Avoid taking a lock in the single master thread.Henning Baldersheim2020-09-151-2/+1
| | | | Reduce code visibility.
* Avoid commitAndWait. It is a blocking call that might deadlock.Henning Baldersheim2020-09-011-18/+10
| | | | This is already safely taken care of in syncViews.
* Revert "Revert "Only wait for th elids that you are interested in.""Henning Baldersheim2020-08-271-5/+5
|
* Revert "Only wait for th elids that you are interested in."Arnstein Ressem2020-08-271-5/+5
|
* Use two stage lid tracking.Henning Baldersheim2020-08-251-3/+3
|
* Only wait for th elids that you are interested in.Henning Baldersheim2020-08-231-5/+5
|
* Track pending uncommitted lids.Henning Baldersheim2020-08-231-2/+2
|
* Wire in interfaces for lid based wait.Henning Baldersheim2020-08-221-0/+6
|
* Ensure that we call commit and wait before changing feedview.Henning Baldersheim2020-08-211-15/+33
| | | | Use a token to ensure the we do not loose any tracked lids.
* Make RemoveOperation an interface and move implmentation to ↵Henning Baldersheim2020-04-081-3/+3
| | | | RemoveOperationWithDocId
* Use xxhash for most out hashing. It is significantly better quality hash.Henning Baldersheim2020-03-271-4/+4
| | | | From 0.7.3 is is also significantly faster.
* Revert "Revert "Since the most important aspect our use of hash maps/sets is ↵Henning Baldersheim2020-03-201-7/+7
| | | | for spee…""
* Revert "Since the most important aspect our use of hash maps/sets is for ↵Harald Musum2020-03-201-7/+7
| | | | spee…"
* Ordering in tests chnages when hashing changes.Henning Baldersheim2020-03-201-7/+7
|
* fastos::TimeStamp -> vespalib::durationHenning Baldersheim2019-12-161-1/+1
|
* Reduce code visibility and minor cleanup.Henning Baldersheim2019-11-271-2/+1
|
* FastOS_Time -> std::chrono.Henning Baldersheim2019-11-201-1/+2
|
* Do not require or provide Syncable.Henning Baldersheim2019-08-251-1/+1
|
* Move RcuVector and relevant support classes to vespalibTor Brede Vekterli2019-05-231-2/+2
| | | | | | | | | | | | | Having RCU support available across all our C++ modules open up new opportunities for optimizations. This changes the following: - `RcuVector` moved from `searchlib` to `vespalib` - `MemoryUsage` moved from `searchlib` to `vespalib` - Introduce a simplified, more generic `GrowStrategy` in `vespalib` used by the moved `RcuVector` which does not have any notion of documents. Existing searchlib `GrowStrategy` gets a simple function to convert to this generic strategy.
* Use the shared executor from the threading service, and remove the loose wire.Henning Baldersheim2019-05-141-54/+47
|
* Wire in the shared executor in the threading service.Henning Baldersheim2019-05-141-3/+3
|
* Remove AttributeMetricsCollection and DocumentDBMetricsCollection as they ↵Geir Storli2019-01-211-7/+4
| | | | are no longer needed.
* Remove all legacy proton metrics (that were deprecated on Vespa 6).Geir Storli2019-01-211-6/+4
|
* - Decouple index config from ProtonConfig.Henning Baldersheim2018-12-031-7/+5
| | | | | - Decouple grow config from Protonconfig. No semantic changes, just refactoring.
* Reduce code visibility and C++11ification.Henning Baldersheim2018-06-141-0/+1
|
* Copy matching metrics from LegacyDocumentDBMetrics to ↵Geir Storli2018-05-041-4/+4
| | | | | | DocumentDBTaggedMetrics using new naming scheme. Rank profile and docid partition are added as dimensions.
* Revert "Revert "Balder/compute summary cache automatically in backend""Henning Baldersheim2018-01-311-1/+1
|
* Revert "Balder/compute summary cache automatically in backend"Harald Musum2018-01-311-1/+1
|
* Add the HwInfo to the BootstrapConfig.Henning Baldersheim2018-01-301-1/+1
|
* Let proton subscribe to bucketspaces config.Tor Egge2018-01-091-0/+3
|
* Consolidate makeBucketSpace() test functions to document module.Tor Egge2017-11-141-2/+2
|
* Remove explicit ack and use feedtoken as a smartptrHenning Baldersheim2017-10-131-27/+12
|
* Use propagated bucket space instead of place holder.Tor Egge2017-10-061-0/+3
|
* Resolve the LogDocumentStoreConfig in DocumentDBConfigManager.Henning Baldersheim2017-10-061-8/+3
|
* There is a less intrusive way ...Henning Baldersheim2017-10-051-1/+1
|
* Wire in for live reconfiguration of DocumentStore.Henning Baldersheim2017-10-051-13/+10
|
* Use multiple threads for mbus invokeHenning Baldersheim2017-09-261-1/+1
|