[code-reflection] Integrated: Juan Fumero reported this issue on Linux/OpenCL

Gary Frost gfrost at openjdk.org
Wed Jun 11 10:15:58 UTC 2025


Juan Fumero reported issue on one of his linux/OpenCL tests.

Code was reinterpreting a cast from address of int on local stack.  Even though we were only 'requesting' a dim of one.  The runtime was clearly accessing 'past' the value addressed.  This was a bad use of reinterpret cast. Code now creates a suitable size_t[3] array.

-------------

Commit messages:
 - Juan Fumero reported this issue on Linux/OpenCL

Changes: https://git.openjdk.org/babylon/pull/439/files
  Webrev: https://webrevs.openjdk.org/?repo=babylon&pr=439&range=00
  Stats: 7 lines in 1 file changed: 5 ins; 0 del; 2 mod
  Patch: https://git.openjdk.org/babylon/pull/439.diff
  Fetch: git fetch https://git.openjdk.org/babylon.git pull/439/head:pull/439

PR: https://git.openjdk.org/babylon/pull/439


More information about the babylon-dev mailing list