WangGuangxin opened a new pull request, #6646:
URL: https://github.com/apache/incubator-gluten/pull/6646

   ## What changes were proposed in this pull request?
   
   Since we are using Spark 3.2, and when we using parquet writer, a deadlock 
encountered. I think the root cause is the same with 
https://github.com/apache/incubator-gluten/issues/6622
   
![image](https://github.com/user-attachments/assets/d2fe299f-e681-426d-b3db-a90b54bd7793)
   
![image](https://github.com/user-attachments/assets/f90f800d-d0e1-46a7-aa0b-e9e9958fcfa6)
   
   
   Current code path makes write thread different with main task thread.
   In fact, the `VeloxWriteQueue` is uselss here I think, and since it will 
buffer input batch (max up to 64 column batch), it will cause offheap oom in 
some case.
   
   (Fixes: \#6645)
   
   ## How was this patch tested?
   Mannually
   
   
   (If this patch involves UI changes, please attach a screenshot; otherwise, 
remove this)
   
   


-- 
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]

Reply via email to