uschindler commented on PR #13339: URL: https://github.com/apache/lucene/pull/13339#issuecomment-2122358609
> > We may add a method like getByteBufferSlice(....). > > I experimented locally with similar before, and the performance impact when converting to/from MemorySegment was horrible. I don't disagree that for NIO/BB dir implementations this could be useful, but it would not intersect with the memory segment implementation, since it performs horribly (when converting from BB to MS, and vice versa). OK, this is not good. From my code review it looked like the DirectByteBuffer also only has a addres and therefor the whole thing is identical by performance (except that we have to get a MemorySegment slice first and then call `toByteBuffer()`). So this is interesting, but then I we can't do anything. Mabe theres some optimizations miissing in ByteBuffer support of vector API. So looks fine then. Working on a final review. -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
