Re: [PR] [GLUTEN-12356][VL] Preserve precision for long decimal to double cast [gluten]

["Smallfu666 (via GitHub)" <gi...@apache.org>]

Smallfu666 commented on PR #12605:
URL: https://github.com/apache/gluten/pull/12605#issuecomment-5216227618

   Thanks @rui-mo — agreed that fixing this in Velox is preferable to carrying the string-based workaround in Gluten, especially given the performance cost here.
   
   Please feel free to pick up the Velox work. The regression matrix and benchmark results in this PR should be reusable for validating the upstream fix, including the reported `DECIMAL(38,19)` case and the values around the double precision boundary.
   
   I'll keep this PR open for now as a reference rather than pushing the workaround further. Once the Velox fix is available, we can either reduce this PR to the Gluten regression coverage / Velox update, or close it if the upstream coverage is sufficient.
   


-- 
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: commits-unsubscribe@gluten.apache.org

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


---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@gluten.apache.org
For additional commands, e-mail: commits-help@gluten.apache.org