Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Merge pull request #13645 from ↵ | Geir Storli | 2020-06-19 | 2 | -4/+7 |
|\ | | | | | | | | | vespa-engine/geirst/multi-threaded-hnsw-indexing-flag Multi-threaded hnsw indexing flag | ||||
| * | Wire flag to enable multi-threaded indexing for a hnsw index. | Geir Storli | 2020-06-19 | 2 | -4/+7 |
| | | |||||
* | | Typo fix. | Tor Egge | 2020-06-19 | 1 | -1/+1 |
| | | |||||
* | | Add description for nearest neighbor brute force limit index property. | Tor Egge | 2020-06-19 | 1 | -0/+7 |
| | | |||||
* | | Wire in nearest neighbor brute force limit. | Tor Egge | 2020-06-19 | 12 | -6/+78 |
| | | |||||
* | | ignore | Håvard Pettersen | 2020-06-19 | 1 | -0/+1 |
|/ | |||||
* | Merge pull request #13635 from vespa-engine/lesters/additional-onnx-operators | Jon Bratseth | 2020-06-19 | 3 | -1/+8 |
|\ | | | | | Lesters/additional onnx operators | ||||
| * | Add erf (the error function) | Lester Solbakken | 2020-06-18 | 3 | -1/+8 |
| | | |||||
* | | Merge pull request #13629 from ↵ | Geir Storli | 2020-06-18 | 3 | -17/+109 |
|\ \ | |/ |/| | | | | | vespa-engine/geirst/two-phase-nearest-neighbor-index-update Add support for updating a nearest neighbor index in two phases. | ||||
| * | Add support for updating a nearest neighbor index in two phases. | Geir Storli | 2020-06-18 | 3 | -17/+109 |
| | | |||||
* | | Merge pull request #13626 from ↵ | Håvard Pettersen | 2020-06-18 | 1 | -6/+6 |
|\ \ | | | | | | | | | | | | | vespa-engine/havardpe/test-fused-multiply-add-effect observe the results of multiply-add across implementations | ||||
| * | | observe the results of multiply-add across implementations | Håvard Pettersen | 2020-06-18 | 1 | -6/+6 |
| | | | | | | | | | | | | | | | | | | - add specific test showing diverging results when using FMA - re-write eval test to not fail due to FMA - re-write hitcollector test to not fail due to FMA | ||||
* | | | Merge pull request #13572 from vespa-engine/arnej/keep-and-follow-refs | Arne H Juul | 2020-06-18 | 10 | -57/+521 |
|\ \ \ | |_|/ |/| | | Arnej/keep and follow refs | ||||
| * | | update test now that TODO is done | Arne Juul | 2020-06-18 | 1 | -2/+2 |
| | | | |||||
| * | | print more statistics | Arne Juul | 2020-06-18 | 1 | -11/+33 |
| | | | |||||
| * | | rewrite with futures | Arne Juul | 2020-06-17 | 1 | -69/+50 |
| | | | |||||
| * | | add mutex protecting in_progress bitvector | Arne Juul | 2020-06-17 | 1 | -16/+20 |
| | | | |||||
| * | | rename me -> self_docid | Arne Juul | 2020-06-17 | 1 | -2/+3 |
| | | | | | | | | | | | | * other tests should now ensure that we don't get self-references. | ||||
| * | | minor stress-test tweaks: | Arne Juul | 2020-06-17 | 1 | -8/+4 |
| | | | | | | | | | | | | | | | | | | * update copyright line * cleanup unused includes etc * increase the numbers to test a lot more | ||||
| * | | refactor task data into common base class | Arne Juul | 2020-06-17 | 1 | -50/+26 |
| | | | |||||
| * | | also check graph symmetry | Arne Juul | 2020-06-17 | 2 | -1/+7 |
| | | | |||||
| * | | save node_ref in prepare_add_doc and check its validity in complete | Arne Juul | 2020-06-17 | 2 | -25/+37 |
| | | | |||||
| * | | untabify | Arne Juul | 2020-06-17 | 1 | -6/+6 |
| | | | |||||
| * | | get rid of test-internal race condition in MyDocVectorStore | Arne Juul | 2020-06-16 | 1 | -9/+7 |
| | | | |||||
| * | | fix bugs found by stress test | Arne Juul | 2020-06-16 | 3 | -5/+13 |
| | | | |||||
| * | | add stress test | Arne Juul | 2020-06-16 | 2 | -0/+375 |
| | | | |||||
| * | | add possible improvement points | Arne Juul | 2020-06-12 | 1 | -0/+2 |
| | | | |||||
| * | | no explicit clear of each level when removing document | Arne Juul | 2020-06-12 | 2 | -3/+8 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | * keep the levels as they were, so any concurrent search operation getting a reference to this docid can follow old links and avoid getting trapped if possible * ensure later GC can pick up any referenced data in HnswGraph::remove_node_for_document instead, it is more robust to have it in the graph layer anyway. | ||||
| * | | keep and follow node_refs when traversing HNSW graph | Arne Juul | 2020-06-12 | 6 | -43/+99 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * cannot assert in HnswGraph read operations, because nodes may be removed or replaced by another thread * read operations on the HNSW graph need to load the node_ref (and check that it's valid) when following links; keep and use that node_ref to ensure succeeding operations actually access information that belongs to the same version of the node. * to ensure the entry point is internally consistent, check that the atomic value is the same after reading the corresponding node_ref, and retry if it was updated concurrently. | ||||
| * | | add getters checking if held references are still valid | Arne Juul | 2020-06-12 | 1 | -2/+24 |
| | | | |||||
* | | | Remove default tensor spec in doctype builder. | Geir Storli | 2020-06-17 | 1 | -10/+2 |
| | | | |||||
* | | | Test that attribute writer can handle put in two phases. | Geir Storli | 2020-06-17 | 1 | -21/+27 |
| | | | |||||
* | | | Implement initial support for two-phase puts in attribute writer. | Geir Storli | 2020-06-17 | 2 | -4/+3 |
| |/ |/| | | | | | This is only turned on for tensor attributes with a hnsw index that allows multi-threaded indexing. | ||||
* | | Revert "Two phase put in attribute writer" | Harald Musum | 2020-06-15 | 3 | -23/+26 |
| | | |||||
* | | Merge pull request #13592 from ↵ | Geir Storli | 2020-06-15 | 3 | -26/+23 |
|\ \ | | | | | | | | | | | | | vespa-engine/geirst/two-phase-put-in-attribute-writer Two phase put in attribute writer | ||||
| * | | Remove default tensor spec in doctype builder. | Geir Storli | 2020-06-15 | 1 | -10/+2 |
| | | | |||||
| * | | Test that attribute writer can handle put in two phases. | Geir Storli | 2020-06-15 | 1 | -21/+27 |
| | | | |||||
| * | | Implement initial support for two-phase puts in attribute writer. | Geir Storli | 2020-06-15 | 2 | -4/+3 |
| | | | | | | | | | | | | This is only turned on for tensor attributes with a hnsw index that allows multi-threaded indexing. | ||||
* | | | Merge pull request #13586 from ↵ | Geir Storli | 2020-06-15 | 1 | -10/+77 |
|\ \ \ | | | | | | | | | | | | | | | | | vespa-engine/toregge/use-foreach-key-range-for-attribute-iterator-termwise-evaluation Use foreach_key_range for attribute iterator termwise evaluation | ||||
| * | | | Factor out common code. | Tor Egge | 2020-06-15 | 1 | -36/+31 |
| | | | | |||||
| * | | | Use foreach_key_range for attribute iterator termwise evaluation | Tor Egge | 2020-06-15 | 1 | -10/+82 |
| |/ / | | | | | | | | | | when inner iterator is a btree iterator. | ||||
* / / | use typify_invoke instead of dispatch_0/select_2 | Arne Juul | 2020-06-15 | 2 | -5/+6 |
|/ / | |||||
* | | Merge pull request #13563 from ↵ | Geir Storli | 2020-06-15 | 4 | -0/+191 |
|\ \ | | | | | | | | | | | | | vespa-engine/toregge/add-foreach-key-range-method-to-btree-iterator Add foreach_key_range method to btree iterator, to scan a range of | ||||
| * | | Add foreach_key_range method to btree iterator, to scan a range of | Tor Egge | 2020-06-12 | 4 | -0/+191 |
| | | | | | | | | | | | | the tree and call function for each key. | ||||
* | | | Merge pull request #13560 from vespa-engine/arnej/follow-std-rules | Arne H Juul | 2020-06-12 | 2 | -11/+27 |
|\ \ \ | |_|/ |/| | | revert back to getting random number in the [0,1) range | ||||
| * | | revert back to getting random number in the [0,1) range | Arne Juul | 2020-06-12 | 2 | -11/+27 |
| |/ | | | | | | | | | | | | | | | | | * for whatever reason, std::uniform_real_distribution doesn't allow a>b when specifying the [a,b) half-open range. So we need to get a number from [0,1) and then do (1.0-unif) to convert to the (0,1] range that we want. * extend the unit test with more initial numbers, and write it in a more compact way. | ||||
* | | unit test two-phase add | Arne Juul | 2020-06-11 | 1 | -0/+56 |
| | | |||||
* | | rename internal prepare result | Arne Juul | 2020-06-11 | 2 | -13/+13 |
| | | |||||
* | | remove stub implementations in NearestNeighborIndex | Arne Juul | 2020-06-11 | 2 | -14/+11 |
| | | | | | | | | | | * make API pure virtual * implement something that will work in mock | ||||
* | | wire in new prepare/complete API | Arne Juul | 2020-06-11 | 2 | -2/+32 |
| | |