[GitHub] [orc] xbthink commented on a diff in pull request #1574: ORC-1473: Fix zero copy read bugs

["xbthink (via GitHub)" <gi...@apache.org>]

xbthink commented on code in PR #1574:
URL: https://github.com/apache/orc/pull/1574#discussion_r1323864572


##########
java/core/src/java/org/apache/orc/impl/RecordReaderUtils.java:
##########
@@ -131,11 +131,21 @@ public boolean isTrackingDiskRanges() {
       return zcr != null;
     }
 
+    /**
+     * @deprecated Use {@link #releaseAllBuffers()} instead. This method was
+     * incorrectly added and shouldn't be used anymore.
+     */
+    @Deprecated
     @Override
     public void releaseBuffer(ByteBuffer buffer) {

Review Comment:
   I'm sorry for later response, I missed this comment...
   I checked hive llap codes, it directly uses the original buffer of zrc `readBuffer` returning to create `BufferChunk`, so there seems to be no problem.
   I think if external systems use this interface directly, they should trace the bytebuff of zrc `readBuffer` returning by themselves, then realease it.
   In previous ORC code, we don't trace the original zrc returning bytebuffer, so it is buggy.
   WDYT, thanks for your review and look forward to your reply.



-- 
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-unsubscribe@orc.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org