jimczi opened a new pull request, #16797: URL: https://github.com/apache/lucene/pull/16797
The stored fields data file is read with `NoReuseHint`: at random by searches, and front to back by merges (#16684). Its writes said `NoReuseHint` but not how the file is written. They now also say `DataAccessHint.SEQUENTIAL`, so a directory sees on flush and merge writes the same hints it sees on reads, and can choose how to write a file that is streamed once and not read back, for example with direct I/O for merges. Directories that ignore write hints are unaffected. `TestStoredFieldsMergeReadAdvice#testDataIsWrittenSequentiallyAndNotReused` checks the data file written by two flushes and a merge. Relates #16684 -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
