github-actions[bot] commented on PR #13133:
URL: https://github.com/apache/lucene/pull/13133#issuecomment-1989681309
This PR has not had activity in the past 2 weeks, labeling it as stale. If
the PR is waiting for review, notify the d...@lucene.apache.org list. Thank you
for your contributi
rmuir commented on PR #13133:
URL: https://github.com/apache/lucene/pull/13133#issuecomment-1965433852
> I'm surprised by how slow this is with AVX off given that this can be
implemented with SSE2 :(.
Yes, it is surprising: we found the same situation with VectorUtil byte[]
methods.
uschindler commented on PR #13133:
URL: https://github.com/apache/lucene/pull/13133#issuecomment-1964672015
> I'm surprised by how slow this is with AVX off given that this can be
implemented with SSE2 :(.
This is why we try to avoid the incubating vector API as much as possible.
The
uschindler commented on PR #13133:
URL: https://github.com/apache/lucene/pull/13133#issuecomment-1964649237
Hi, you can try it out. Two important comments:
- In Java 21, MemorySegment is still a "preview" class, so it can't be used
in any public method signatures. So actually to pass the
mccullocht commented on PR #13133:
URL: https://github.com/apache/lucene/pull/13133#issuecomment-1964602685
@rmuir @uschindler Would it be sufficient to add a factory function to
`VectorizationProvider` for this like `GroupVIntUtil.Decoder
createGroupVIntDecoder(MemorySegment segment)`? I a
uschindler commented on PR #13133:
URL: https://github.com/apache/lucene/pull/13133#issuecomment-1964318304
I have to agree with Robert. We can't glue that code inside
MemorySegmentIndexInput.
The only way to do this is to move the vector optimized decoding to a
special Vectorization
rmuir commented on PR #13133:
URL: https://github.com/apache/lucene/pull/13133#issuecomment-1964303622
Hi, a couple suggestions:
1. Somehow, we need to avoid Vector API code inside the MemorySegment code.
Just because MemorySegment is available, does not mean Vector API is usable,
on
