JDK-8356537 : TestFloat16VectorOperations wastes CPU time
  • Type: Bug
  • Component: hotspot
  • Sub-Component: compiler
  • Affected Version: 25
  • Priority: P4
  • Status: Closed
  • Resolution: Duplicate
  • Submitted: 2025-05-08
  • Updated: 2025-06-02
  • Resolved: 2025-06-02
The Version table provides details related to the release that this issue/RFE will be addressed.

Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.

To download the current JDK release, click here.
JDK 25
25Resolved
Related Reports
Duplicate :  
JDK-8355585 - Aarch64: Add aarch64 backend for Float16 vector operations
Relates :  
JDK-8346236 - Auto vectorization support for various Float16 operations
Description
This test needs too long for what it tests. 

The warmup set to 10000 seems excessive and causes the test to run for several minutes.

It should be possible to:
- trigger C2 compilation with fewer warmup iterations or
- find another way to force the auto-vectorization in the test.
Comments
The PR for JDK-8355585 also reduced the warmup to 50. Closing as fixed.
28-05-2025
Setting fixVersion to JDK 26 for now. Feel free to fix in JDK 25 if there's time.
09-05-2025
[~jbhateja] FYI
09-05-2025
ILW = Inefficient test, single test, no workaround = MLH = P4
09-05-2025