dweiss commented on issue #14050:
URL: https://github.com/apache/lucene/issues/14050#issuecomment-2530596008
Thanks, Chris. The only Windows tests running at the moment are on Uwe's
vbox. Github's Windows testing is so slow as to be almost unusable. The results
from Uwe's box are sent to th
ChrisHegarty commented on issue #14050:
URL: https://github.com/apache/lucene/issues/14050#issuecomment-2529358488
Oops! Sorry, clearly I added this test and didn't noticed it fail on
Windows. For Lucene, I'm perfectly fine with returning "don't know" for all
Windows. Lemme look into why
dweiss commented on issue #14050:
URL: https://github.com/apache/lucene/issues/14050#issuecomment-2529232048
I don't have access to openjdk's Jira, maybe @ChrisHegarty or Uwe (or you?)
can file an issue there?
--
This is an automated message from the Apache Git Service.
To respond to the
dweiss commented on issue #14050:
URL: https://github.com/apache/lucene/issues/14050#issuecomment-2529159503
I've peeked at the flags returned by QueryWorkingSetEx - just wondering if
it'd be possible to query it directly using ffi. :) But I can't tell which of
these flags would correspond
rmuir commented on issue #14050:
URL: https://github.com/apache/lucene/issues/14050#issuecomment-2529176829
I think it won't return flags at all in that case.
--
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
rmuir commented on issue #14050:
URL: https://github.com/apache/lucene/issues/14050#issuecomment-2529138823
At least not everywhere: `Minimum supported server: Windows Server 2008`. So
maybe this one can be revisited in 2024 and fixed...
--
This is an automated message from the Apache Git
rmuir commented on issue #14050:
URL: https://github.com/apache/lucene/issues/14050#issuecomment-2529106991
It seems the commented-out code with hardcoded `false` is also in the JDK
master? I think we should at least report the bug since it is impacting us.
I am not sure it needs a ma
dweiss commented on issue #14050:
URL: https://github.com/apache/lucene/issues/14050#issuecomment-2529079912
This was my initial guess too. But if we hardcode it and then they fix the
JDK with some magic Windows API that returns the right value? Should we try to
detect if isLoaded returns t
rmuir commented on issue #14050:
URL: https://github.com/apache/lucene/issues/14050#issuecomment-2529068203
yeah, i'm suggesting adding this to the code:
```
if (Constants.WINDOWS) {
return DONT_KNOW;
}
```
--
This is an automated message from the Apache Git Service.
To re
dweiss commented on issue #14050:
URL: https://github.com/apache/lucene/issues/14050#issuecomment-2529066466
Yep, I think I understand the reason it returns Optional - a missing value
would indicate "don't know", so it's a tri-state knob: yes, no, don't know. The
problem is that on Windows
msokolov commented on issue #14047:
URL: https://github.com/apache/lucene/issues/14047#issuecomment-2528860574
also pushed 4f8035c01393735e347f42c0bdc5e7b932db7447 to address a build
failure by disabling randomization in more tests
--
This is an automated message from the Apache Git Serv
dweiss commented on issue #14050:
URL: https://github.com/apache/lucene/issues/14050#issuecomment-2528617386
While the test can be fixed easily (bypass the check on Windows), I wonder
if it's a good fix. In essence, this method lies on Windows... What are the
side-effects of this always ret
dweiss commented on issue #14050:
URL: https://github.com/apache/lucene/issues/14050#issuecomment-2528605123
Right. Seems like it's always false on Windows.
https://github.com/openjdk/jdk21/blob/890adb6410dab4606a4f26a942aed02fb2f55387/src/java.base/windows/native/libnio/MappedMemoryU
msokolov merged PR #14049:
URL: https://github.com/apache/lucene/pull/14049
--
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.ap
msokolov closed issue #14047: Randomize KNN vector format parameters in tests
URL: https://github.com/apache/lucene/issues/14047
--
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.
dweiss opened a new issue, #14050:
URL: https://github.com/apache/lucene/issues/14050
### Description
There is a check added to BaseDirectoryTest case in #13998 which has been
failing since that PR was merged (nearly all Windows builds failing on jenkins):
```
var loaded
jpountz commented on code in PR #13948:
URL: https://github.com/apache/lucene/pull/13948#discussion_r1876152549
##
lucene/core/src/java/org/apache/lucene/codecs/lucene90/Lucene90DocValuesProducer.java:
##
@@ -791,25 +792,26 @@ public BinaryDocValues getBinary(FieldInfo field) th
msokolov commented on PR #14049:
URL: https://github.com/apache/lucene/pull/14049#issuecomment-2528438728
makes sense, I'll add a CHANGES entry and commit
--
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 t
jpountz merged PR #14046:
URL: https://github.com/apache/lucene/pull/14046
--
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
19 matches
Mail list logo
