That bug report contains reproducers and thoughts. Some of the issues encountered:
* The OOM is thrown from an unexpected place
* The finally block is not run
In the future with value types this situation might become more common. We need to decide how we want to handle this.
- duplicates
-
JDK-8270010 Handling of failed reallocation of scalar-replaced objects skips some exception handlers
- Closed
- relates to
-
JDK-8226536 Catch OOM from deopt that fails rematerializing objects
- Resolved
-
JDK-8320205 some GC stress tests fail due to OOM in EA materialization
- Closed
-
JDK-8273095 vmTestbase/vm/mlvm/anonloader/stress/oome/heap/Test.java fails with "wrong OOME"
- Resolved
-
JDK-8269849 vmTestbase/gc/gctests/PhantomReference/phantom002/TestDescription.java failed with "OutOfMemoryError: Java heap space: failed reallocation of scalar replaced objects"
- Resolved
-
JDK-8289764 gc/lock tests failed with "OutOfMemoryError: Java heap space: failed reallocation of scalar replaced objects"
- Resolved
-
JDK-8270010 Handling of failed reallocation of scalar-replaced objects skips some exception handlers
- Closed
-
JDK-8066400 Reserve memory for OOME exception processing during deoptimization
- Open