aboutsummaryrefslogtreecommitdiffstats
path: root/staging_vespalib
Commit message (Collapse)AuthorAgeFilesLines
* ChildProc -> ChildProcessArne Juul2020-07-171-2/+2
|
* rename SlaveProc -> ChildProcArne Juul2020-07-171-2/+2
|
* Use EXTERNAL_DEPENDS for external dependencies.Tor Egge2020-07-071-0/+1
|
* avoid naked exitArne Juul2020-07-031-1/+2
|
* use appropriate lockHåvard Pettersen2020-07-021-1/+1
|
* use event barrier for syncHåvard Pettersen2020-07-022-49/+68
|
* improve syncHåvard Pettersen2020-07-022-26/+42
| | | | | | - lock once to post all sync tasks - only post to non-idle strands - let sync tasks bypass blocking based on task limit
* Let getExecutorId be a pure interface.Henning Baldersheim2020-06-269-50/+60
|
* Run prepare step of two-phase puts in the shared executor instead of the ↵Geir Storli2020-06-251-0/+31
| | | | | | | | attribute field executor. Benchmarking has shown that we get an uneven balance between the underlying threads when using the attribute field executor. This is a sequenced task executor, where each underlying thread only is allowed to handle a subset of the prepare tasks (based on the local document id). Using the shared executor should give an even balance.
* Update the test to what is important for the adaptive executor.Henning Baldersheim2020-06-251-6/+2
|
* Let the executor create an executor id to its liking.Henning Baldersheim2020-06-257-16/+27
| | | | Taske full advantage over strands instead of being limited to threads.
* Reduce from warning to infoHenning Baldersheim2020-05-131-1/+1
|
* /proc/self/smaps is linux specific.Tor Egge2020-04-211-0/+4
|
* Disable some unit tests on Darwin.Tor Egge2020-04-141-2/+7
|
* Add control for kind_of_watermark and reaction time.Henning Baldersheim2020-04-065-8/+44
|
* Fix format string in SequencedTaskExecutor benchmark.Tor Egge2020-04-061-1/+2
|
* Fix merge conflict after revert.Henning Baldersheim2020-04-061-2/+2
|
* Merge branch 'master' into ↵Henning Baldersheim2020-04-052-13/+3
|\ | | | | | | balder/move-sequenced-task-executors-to-staging_vespalib
| * Revert "Balder/control naptime"Henning Baldersheim2020-04-052-13/+3
| |
* | Also allow for testing of the adaptive task executor.Henning Baldersheim2020-04-041-8/+13
| |
* | Move sequenced task executors to staging vespalibHenning Baldersheim2020-04-0418-0/+1627
|/
* Control nap timeHenning Baldersheim2020-04-042-3/+13
|
* pointre => pointerHenning Baldersheim2020-03-301-1/+1
|
* Avoid double lookup.Henning Baldersheim2020-03-302-12/+12
|
* Only start LRU mode once half fullHenning Baldersheim2020-03-291-1/+1
|
* Merge pull request #12727 from ↵Henning Baldersheim2020-03-262-13/+23
|\ | | | | | | | | vespa-engine/balder/let-optimize-flag-control-index-and-summary-executor-too Add shutdown to thread interface.
| * Add shutdown to thread interface.Henning Baldersheim2020-03-232-13/+23
| | | | | | | | Let the optimize config control index and summary executor too.
* | Unify the metrics for queuesize similar to what we have for the spi queues.Henning Baldersheim2020-03-252-8/+6
|/
* Use fundamental integer types in vespalib::visit.Tor Egge2020-03-222-16/+32
|
* Revert "Revert "Since the most important aspect our use of hash maps/sets is ↵Henning Baldersheim2020-03-202-20/+20
| | | | for spee…""
* Revert "Since the most important aspect our use of hash maps/sets is for ↵Harald Musum2020-03-202-20/+20
| | | | spee…"
* Merge pull request #12646 from ↵Henning Baldersheim2020-03-202-20/+20
|\ | | | | | | | | vespa-engine/balder/make-default-hashing-use-2N-and-instead-of-prime-mod Since the most important aspect our use of hash maps/sets is for spee…
| * Since the most important aspect our use of hash maps/sets is for speed we ↵Henning Baldersheim2020-03-202-20/+20
| | | | | | | | | | | | change from using prime numbers as table size and modulo to using 2^N and using simple AND.
* | Reserve space for vector.Henning Baldersheim2020-03-201-1/+1
|/ | | | Varoius code cleanup for c++11.
* - Use a single common lock.Henning Baldersheim2020-03-122-22/+39
| | | | | - Introduce 2 stage startSync/sync. - avoid loosing wakeup on sync.
* Use std::mutex/condition_variable.Henning Baldersheim2020-03-122-42/+28
| | | | Use shorter wait during sync, as that is urgent and synchronous.
* In order to drain Q faster on sync, and also detect that we are in sync faster,Henning Baldersheim2020-03-112-12/+39
| | | | we wake the consumer unconditionally on sync, and also unconditionally wake the producer when consumer is idle.
* Use aquire over relaxed.Henning Baldersheim2020-03-051-1/+1
|
* Use acquire/release semantics for _rpHenning Baldersheim2020-03-022-2/+2
|
* Use proper memory_orderingHenning Baldersheim2020-02-282-13/+16
|
* Thread safe on put.Henning Baldersheim2020-02-282-17/+21
|
* Add single threaded thoughput optimized executor with high and low watermark ↵Henning Baldersheim2020-02-276-0/+269
| | | | at 25% / 75%.
* Fix format string in ProcessMemoryStats::create method.Tor Egge2020-02-071-1/+1
|
* Increase memory sampling frequency to every 20s instead of every 60s in ↵Henning Baldersheim2020-02-051-0/+2
| | | | | | order to detect changes faster. Also log anomalies.
* Move the transaction implementation from StructuredFieldValue to DocumentHenning Baldersheim2020-01-231-4/+4
|
* c_str -> dataHenning Baldersheim2020-01-212-10/+10
|
* GC a load of unused code. ByteBuffer towards read only.Henning Baldersheim2020-01-202-2/+2
|
* Remove unused lambda capture in findMissing function.Tor Egge2020-01-091-2/+2
|
* Merge pull request #11705 from vespa-engine/arnej/use-visit-rangesArne H Juul2020-01-091-49/+27
|\ | | | | use vespalib::visitRanges
| * use vespalib::visitRangesArne Juul2020-01-081-49/+27
| |