baibaichen opened a new pull request, #13069:
URL: https://github.com/apache/gluten/pull/13069
## What changes are proposed in this pull request?
Fixes #13068.
Spark can invoke GlutenDriverEndpoint.onStart() before setupEndpoint
returns, leaving the constructor-assigned driverEndpoint field null.
Use the registered self reference in onStart() and remove the redundant
field. Add a Velox regression test under gluten-ut/spark41 that invokes
onStart() before setupEndpoint returns.
## How was this patch tested?
- Compiled the changed endpoint and regression suite with Scala 2.13.17
and ran ScalaTest against Spark 4.1.1 on JDK 17: 1 test passed.
- Verified that the same test against the original implementation fails
with NullPointerException at driverEndpoint.address.
- Ran dev/format-scala-code.sh, including formatting and license-header
validation of the regression suite.
- dev/check.py header main --fix could not complete because the checkout
lacks dev/license-header.py; Scala headers were validated with Spotless.
- No full native rebuild or full Maven test suite was run.
## Was this patch authored or co-authored using generative AI tooling?
Generated-by: OpenAI Codex
--
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]
