summaryrefslogtreecommitdiffstats
path: root/searchlib/src
Commit message (Collapse)AuthorAgeFilesLines
* Merge pull request #26474 from vespa-engine/balder/use-thread-safe-randomv8.141.29Henning Baldersheim2023-03-172-14/+13
|\ | | | | Use thread safe random to see if that stabilizes the logdatastore_test.
| * Use std::minstd_randHenning Baldersheim2023-03-171-2/+4
| |
| * Use thread safe random to see if that stabilizes the logdatastore_test.Henning Baldersheim2023-03-172-14/+11
| |
* | Extend attributes.def with raw data type.Tor Egge2023-03-172-0/+3
|/
* Merge pull request #26462 from vespa-engine/arnej/handle-expression-in-sliceJon Bratseth2023-03-171-0/+2
|\ | | | | * make InputRecorder handle slice with embedded expressions
| * * make InputRecorder handle slice with embedded expressionsArne Juul2023-03-161-0/+2
| | | | | | | | * special handling for Generate and Slice in BindingExtractor
* | Remove incorrect expectation.Tor Egge2023-03-161-1/+0
|/
* Handle SingleRawAttribute in AttributeUpdater.Tor Egge2023-03-161-2/+3
|
* - GC unused _activeBuffers member.Henning Baldersheim2023-03-153-24/+23
| | | | - Just keep a very small ChangeVector, it might never be used.
* Merge pull request #26449 from ↵Henning Baldersheim2023-03-151-1/+1
|\ | | | | | | | | vespa-engine/toregge/factor-out-implementation-dependent-sizeof-std-set Factor out implementation dependent sizeof(std::set<NameId>).
| * Factor out implementation dependent sizeof(std::set<NameId>).Tor Egge2023-03-151-1/+1
| |
* | Factor out implementation dependent sizeof(std::string).Tor Egge2023-03-151-8/+10
|/
* Prefer std::vectorHenning Baldersheim2023-03-144-25/+19
|
* Update test after rebaseHenning Baldersheim2023-03-141-2/+2
|
* Update expected sizesHenning Baldersheim2023-03-141-2/+2
|
* summary.cache.initialentries and summary.log.chunk.skipcrconread does not ↵Henning Baldersheim2023-03-1420-87/+32
| | | | have any purpose.
* Merge pull request #26433 from ↵Henning Baldersheim2023-03-144-14/+46
|\ | | | | | | | | vespa-engine/balder/also-consider-static-cost-of-caches Balder/also consider static cost of caches
| * Consider platform dependent sizeof(std::mutex) in test.Henning Baldersheim2023-03-141-2/+3
| |
| * Increment the correct metricHenning Baldersheim2023-03-141-1/+1
| |
| * Track static cache usageHenning Baldersheim2023-03-142-2/+14
| |
| * Wire in and test static memory usage for caches.Henning Baldersheim2023-03-144-14/+33
| |
* | Handle that std:.mutex is platform dependentHenning Baldersheim2023-03-141-1/+1
|/
* Update typos in comments.Henning Baldersheim2023-03-141-2/+2
|
* Use std::vector instead of vespalib::ArrayHenning Baldersheim2023-03-146-33/+24
|
* GC unused includesHenning Baldersheim2023-03-144-7/+4
|
* Pre sizing is not necessary.Henning Baldersheim2023-03-146-25/+25
|
* Use a hash map to avoid a presize vector.Henning Baldersheim2023-03-143-12/+18
|
* Ensure locks are held when accessing _holdFileChunks.Henning Baldersheim2023-03-142-12/+14
|
* Track memory usage of LogDataStoreHenning Baldersheim2023-03-142-0/+16
|
* Merge pull request #26417 from vespa-engine/balder/add-state-ptr-in-buffer-metaHenning Baldersheim2023-03-135-41/+34
|\ | | | | Allocate BufferState in stash and add pointer to BufferAndMeta
| * Allocate BufferState in stash and add pointer to BufferAndMetaHenning Baldersheim2023-03-135-41/+34
| |
* | Reduce creation of Document instances without DocumentTypeRepo.Geir Storli2023-03-134-6/+4
| |
* | Adjust default build settings for Ubuntu 22.04.Tor Egge2023-03-131-0/+3
|/
* Merge pull request #26403 from vespa-engine/bratseth/autoscaling-completionValerij Fredriksen2023-03-102-6/+3
|\ | | | | Bratseth/autoscaling completion
| * Merge branch 'master' into bratseth/autoscaling-completionJon Bratseth2023-03-1013-37/+19
| |\
| * | Cooldown from completion rather than start of scalingJon Bratseth2023-03-102-6/+3
| | |
* | | Handle single value raw attribute in attribute dfw.Tor Egge2023-03-102-0/+14
| |/ |/|
* | Use a non const Store when you are in write thread, and a const when not.Henning Baldersheim2023-03-1011-27/+15
| |
* | Compacting is only checked during move in write threadHenning Baldersheim2023-03-102-4/+2
| |
* | Just use maximum buffersizeHenning Baldersheim2023-03-101-5/+2
| |
* | Use BufferMeta instead of BufferState to get array sizeHenning Baldersheim2023-03-101-3/+2
|/
* Report more of the static memory usage for datastores.Henning Baldersheim2023-03-096-32/+37
|
* Test both used and allocated memory.Henning Baldersheim2023-03-092-6/+22
|
* Add saver and loader for SingleRawAttribute.Tor Egge2023-03-0913-4/+350
|
* Add test showing accounted memory usage for empty attributes.Henning Baldersheim2023-03-081-0/+24
|
* no .output allowed for rankingExpression(foo)Arne Juul2023-03-081-1/+1
|
* Add function comment.Tor Egge2023-03-082-2/+10
|
* Factor out common code to make a EntryRef vector snapshot.Tor Egge2023-03-0823-112/+87
|
* Move BlobSequenceReader from search::tensor to search::attribute namespace.Tor Egge2023-03-086-7/+9
|
* Merge pull request #26348 from vespa-engine/toregge/add-raw-buffer-storeGeir Storli2023-03-075-58/+117
|\ | | | | Add RawBufferStore.