singh264 commented on issue #12916:
URL: https://github.com/apache/lucene/issues/12916#issuecomment-1865016211
> If any of you manages to reproduce, I'm interested in the command that
you used
I was able to reproduce the failure in
`TestIndexWriterThreadsToSegments.testSegmentCountO
uschindler commented on issue #12916:
URL: https://github.com/apache/lucene/issues/12916#issuecomment-1865001473
I made test fail on my AMD Ryzen 3700 (the Policeman Jenkins Sever):
```sh
$ ./gradlew -p lucene/core -Dtests.seed=F7B4CD7A5624D5EC beast --tests
TestIndexWriterThreadsT
uschindler commented on issue #12916:
URL: https://github.com/apache/lucene/issues/12916#issuecomment-1864923856
Thanks. Maybe the OpenJ9 people can help how they reproduced. You can use
`gradlew beast` instead of `gradlew test` to run the forked copies inside
gradle.
More looing in
jpountz commented on issue #12916:
URL: https://github.com/apache/lucene/issues/12916#issuecomment-1864917501
Thanks. I'm trying to reproduce failures locally with the following command,
without luck so far with JDK 17 and JDK 21. I'll dig more tomorrow. If any of
you manages to reproduce,
uschindler commented on issue #12916:
URL: https://github.com/apache/lucene/issues/12916#issuecomment-1864883276
See this comment:
https://github.com/eclipse-openj9/openj9/issues/18400#issuecomment-1834577142
--
This is an automated message from the Apache Git Service.
To respond to the m
jpountz commented on issue #12916:
URL: https://github.com/apache/lucene/issues/12916#issuecomment-1864882629
Thanks I had missed that, I'll look more into it.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL ab
uschindler commented on issue #12916:
URL: https://github.com/apache/lucene/issues/12916#issuecomment-1864881002
They made statistics in the linked issue. Hotspot also fails. So they
rejected it as openj9 issue.
--
This is an automated message from the Apache Git Service.
To respond to th
uschindler commented on issue #12916:
URL: https://github.com/apache/lucene/issues/12916#issuecomment-1864879720
This is a real bug and not one of openj9. You can reproduce this bug with
enough tries on hotspot, too.
--
This is an automated message from the Apache Git Service.
To respond
jpountz commented on issue #12916:
URL: https://github.com/apache/lucene/issues/12916#issuecomment-1864843912
Does someone understand if adding synchronization is fixing a real bug of it
it just helps hide a J9 bug? This method is subject to contention and #12199
was about avoiding locking
jpountz closed pull request #12205: Remove remaining sources of contention on
indexing.
URL: https://github.com/apache/lucene/pull/12205
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific
jpountz commented on PR #12205:
URL: https://github.com/apache/lucene/pull/12205#issuecomment-1864768401
This test failure is sneaky, I extracted some bits from this PR into #12958.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to Git
jpountz commented on PR #12954:
URL: https://github.com/apache/lucene/pull/12954#issuecomment-1864765767
> so maybe we can consider an other approach: try to avoid the for-loop in
reset() if the instance can be reused
+1 this sounds like a good idea!
--
This is an automated message
easyice commented on PR #12954:
URL: https://github.com/apache/lucene/pull/12954#issuecomment-1864749851
I took several hours to confirm the results, the benchmark shows it became
faster, this exceeded my expectation, we think the speedup is due to remove the
loop that initializes the `
dungba88 commented on code in PR #12831:
URL: https://github.com/apache/lucene/pull/12831#discussion_r1421299753
##
lucene/analysis/common/src/java/org/apache/lucene/analysis/charfilter/NormalizeCharMap.java:
##
@@ -111,7 +111,7 @@ public NormalizeCharMap build() {
for
jpountz commented on PR #12205:
URL: https://github.com/apache/lucene/pull/12205#issuecomment-1864651149
The above failure is a bit scary, I'll try to split this PR.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
dungba88 commented on PR #12948:
URL: https://github.com/apache/lucene/pull/12948#issuecomment-1864650439
I think we can remove the `size()` method in 10.0?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above
jpountz commented on issue #12957:
URL: https://github.com/apache/lucene/issues/12957#issuecomment-1864480721
I just pushed the change, thanks @mikemccand for putting me on the right
track.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log
jpountz closed issue #12957: Reproducible test failure with Terms#intersect on
the default codec
URL: https://github.com/apache/lucene/issues/12957
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to th
mikemccand commented on issue #12957:
URL: https://github.com/apache/lucene/issues/12957#issuecomment-1864450920
OK the `DirectPostingsFormat` failure is also happy with this fix. +1 to
merge. Thanks @jpountz!
--
This is an automated message from the Apache Git Service.
To respond to th
mikemccand commented on issue #12957:
URL: https://github.com/apache/lucene/issues/12957#issuecomment-1864448812
> > Terms.intersect(Automaton a, BytesRef startTerm) requires that startTerm
is accepted by the incoming automaton, yet the way CheckIndex is calling it can
clearly violate that.
jpountz merged PR #12952:
URL: https://github.com/apache/lucene/pull/12952
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscr...@lucene.apa
jpountz commented on issue #12957:
URL: https://github.com/apache/lucene/issues/12957#issuecomment-1864407118
Oh I see, I created binary automata, but the API implicitly treats automata
as UTF32 automata, so you need to tell it explicitly that it's a binary
automaton. And something like tha
jpountz commented on issue #12957:
URL: https://github.com/apache/lucene/issues/12957#issuecomment-1864395031
> Terms.intersect(Automaton a, BytesRef startTerm) requires that startTerm
is accepted by the incoming automaton, yet the way CheckIndex is calling it can
clearly violate that.
mikemccand commented on issue #12957:
URL: https://github.com/apache/lucene/issues/12957#issuecomment-1864386831
I'll try to fix `CheckIndex` so that it only uses `startTerm` that is
accepted by the automaton.
--
This is an automated message from the Apache Git Service.
To respond to the
mikemccand commented on issue #12957:
URL: https://github.com/apache/lucene/issues/12957#issuecomment-1864385415
OK I think the issue here may be that `Terms.intersect(Automaton a, BytesRef
startTerm)` requires that `startTerm` is accepted by the incoming automaton,
yet the way `CheckIndex`
ChrisHegarty commented on PR #12953:
URL: https://github.com/apache/lucene/pull/12953#issuecomment-1864374811
I'm way too slow here, sorry. Belated LGTM. And thanks for following up on
this.
--
This is an automated message from the Apache Git Service.
To respond to the message, please lo
jpountz opened a new issue, #12957:
URL: https://github.com/apache/lucene/issues/12957
### Description
The new CheckIndex checks are causing some test failures with the default
codec, which are reproducible and look like real bugs? I started looking but
I'm not familiar enough with B
27 matches
Mail list logo
