summaryrefslogtreecommitdiffstats
path: root/searchcore/src/tests/proton
Commit message (Collapse)AuthorAgeFilesLines
* Move files in C++ modules back to original locationHarald Musum2020-09-111-0/+0
|
* Use full name in config definition file namesHarald Musum2020-09-101-0/+0
|
* Revert "Revert "Ignore summary override if attribute vector is missing." ↵Henning Baldersheim2020-09-101-1/+4
| | | | MERGEOK"
* Revert "Ignore summary override if attribute vector is missing."Jon Marius Venstad2020-09-091-4/+1
|
* Ignore summary override if attribute vector is missing.Tor Egge2020-09-091-1/+4
|
* adapt and convert between vespa and onnx typesHåvard Pettersen2020-09-051-1/+1
|
* Rename GenericTensorAttribute -> SerializedTensorAttribute.Geir Storli2020-09-041-5/+5
|
* Revert "Revert "Configure compression and chunk size""Henning Baldersheim2020-09-031-1/+2
|
* Account for extra invocation on summary executor too.Henning Baldersheim2020-09-031-1/+1
|
* Revert "Configure compression and chunk size"Arnstein Ressem2020-09-021-2/+1
|
* Merge pull request #14255 from ↵Lester Solbakken2020-09-021-61/+97
|\ | | | | | | | | vespa-engine/havardpe/unit-test-rank-setup-verification-with-onnx-model test onnx models (plus some cleanup)
| * test onnx models (plus some cleanup)Håvard Pettersen2020-09-021-61/+97
| |
* | Merge pull request #14245 from ↵Harald Musum2020-09-021-0/+1
|\ \ | |/ |/| | | | | vespa-engine/aressem/add-cost-to-expensive-cpp-tests Annotate some expensive tests to allow them to start first when runni…
| * Annotate some expensive tests to allow them to start first when running in ↵Arnstein Ressem2020-09-021-0/+1
| | | | | | | | parallel.
* | Configure compression and chunk sizeHenning Baldersheim2020-09-021-1/+2
| |
* | Merge pull request #14227 from vespa-engine/havardpe/handle-onnx-models-configHåvard Pettersen2020-09-027-3/+45
|\ \ | | | | | | handle onnx models config
| * | handle onnx models configHåvard Pettersen2020-09-017-3/+45
| | |
* | | Split interface and implementation in separate compilation units.Henning Baldersheim2020-09-011-0/+1
|/ /
* / 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.
* Use an allocator that automagically allocates large buffer by using mmap.Henning Baldersheim2020-08-312-0/+2
|
* Move streaming operator to namespace searched by ADL.Tor Egge2020-08-311-0/+4
|
* Style chnagesHenning Baldersheim2020-08-272-4/+1
|
* Refactor and testHenning Baldersheim2020-08-261-0/+96
|
* Use two stage lid tracking.Henning Baldersheim2020-08-256-35/+64
|
* Only wait for th elids that you are interested in.Henning Baldersheim2020-08-233-15/+15
|
* Track pending uncommitted lids.Henning Baldersheim2020-08-233-6/+6
|
* Wire in interfaces for lid based wait.Henning Baldersheim2020-08-224-6/+30
|
* Ensure that we call commit and wait before changing feedview.Henning Baldersheim2020-08-215-19/+43
| | | | Use a token to ensure the we do not loose any tracked lids.
* - Let LidReuseDelayer be owned by FeedView and configured only at ↵Henning Baldersheim2020-08-214-57/+50
| | | | | | | construction time. - No need for ILidReuseDelayer interface. - Update test to reflect that no extra task is dispatched in master thread for reconfiguraion of lid reuse delayer.
* Revert "Revert "Balder/use an actual fieldset repo""Henning Baldersheim2020-08-202-38/+36
|
* Use a fixed visibility-delay of 100ms during replayHenning Baldersheim2020-08-202-2/+1
|
* Transfer shared_ptr by value.Henning Baldersheim2020-08-201-1/+1
|
* Avoid time based commit handling inside storeonly feed view.Henning Baldersheim2020-08-193-65/+22
|
* Do the time tracking for immediate commit in the replay state instead of ↵Henning Baldersheim2020-08-192-44/+40
| | | | having it spread all over.
* Revert "Balder/use an actual fieldset repo"Arnstein Ressem2020-08-192-36/+38
|
* Some cleanup and refactoring.Henning Baldersheim2020-08-192-23/+18
|
* Merge pull request #14051 from vespa-engine/balder/use-an-actual-fieldset-repoHenning Baldersheim2020-08-182-38/+36
|\ | | | | Balder/use an actual fieldset repo
| * Change api so that we can drop clone from the interface and prepare for ↵Henning Baldersheim2020-08-142-38/+36
| | | | | | | | using a true repo.
* | wire in using new JSON-lite location formatArne Juul2020-08-181-7/+11
| |
* | allow extracting locations as termsArne Juul2020-08-162-5/+5
|/
* Make sure the entries in the set are unique.Henning Baldersheim2020-08-142-5/+6
| | | | | Make both a less and equal operator so std::unique can be used to. Add and interface so that whole sets can be tested effectively if they are subsets.
* Use a vector backed set for speed.Henning Baldersheim2020-08-132-12/+8
|
* GC unused code.Henning Baldersheim2020-08-136-36/+36
|
* Test that documents are empty.Henning Baldersheim2020-08-131-0/+2
|
* Handle all cases androp default: label, also add test for NONE and DOCID setsHenning Baldersheim2020-08-131-1/+10
|
* Revert "Revert "Revert "Revert "- Change APi for fetching documents to ↵Henning Baldersheim2020-08-136-84/+132
| | | | support optimal document retri…""""
* Revert "Revert "Revert "- Change APi for fetching documents to support ↵Henning Baldersheim2020-08-136-132/+84
| | | | optimal document retri…"""
* Insert into correct field setHenning Baldersheim2020-08-121-3/+3
|
* Control your negations... And test that you do.Henning Baldersheim2020-08-121-0/+43
|
* Revert "Revert "- Change APi for fetching documents to support optimal ↵Henning Baldersheim2020-08-126-84/+89
| | | | document retri…""