iverase commented on pull request #2094: URL: https://github.com/apache/lucene-solr/pull/2094#issuecomment-735732072
> SegmentInfos certainly cannot know the endianness of the file up-front. But for other file formats, we could know this on a per-file-format basis? E.g. Lucene86PointsFormat always uses big endian but Lucene90PointsFormat will always use little endian? @jpountz That is true, but what I understand is the files read by for example `CompressingStoredFieldsReader`, they hold versioning in the header? > So I'd like to see benchmark results before anything is committed. @rmuir I have created [JMH benchmarks](https://github.com/iverase/endianness_benchmark) that read longs using BytesBuffer and LongBuffer with different endianness. Results are here: ``` Benchmark (byteOrder) Mode Cnt Score Error Units ReadLongBenchmark.readBytesBuffer LE thrpt 25 9.015 ± 0.012 ops/us ReadLongBenchmark.readBytesBuffer BE thrpt 25 8.333 ± 0.040 ops/us ReadLongBenchmark.readLongsBuffer LE thrpt 25 24.510 ± 0.191 ops/us ReadLongBenchmark.readLongsBuffer BE thrpt 25 9.981 ± 0.034 ops/us ``` ---------------------------------------------------------------- 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. 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
