-
Bug
-
Resolution: Duplicate
-
P4
-
24
java.lang.OutOfMemoryError: Java heap space
at java.base/java.lang.Thread.beforeSleep(Thread.java:461)
at java.base/java.lang.Thread.sleepNanos(Thread.java:491)
at java.base/java.lang.Thread.sleep(Thread.java:527)
at TestDisposerRace.giveGCAChance(TestDisposerRace.java:104)
at TestDisposerRace.generateOOME(TestDisposerRace.java:115)
at TestDisposerRace.run(TestDisposerRace.java:57)
at TestDisposerRace.main(TestDisposerRace.java:44)
giveGCAChance is called after the previous loop has been broken by OOME being thrown. All it does is sleep in a loop and so potentially allow GC a chance to catch up and clear memory before the test moves on. However, since the Loom integration the implementation of Thread.sleep has changed significantly and it can now allocate and so encounter a secondary OOME - hence the current failure.
The test needs more resilience against OOME in unexpected places.
at java.base/java.lang.Thread.beforeSleep(Thread.java:461)
at java.base/java.lang.Thread.sleepNanos(Thread.java:491)
at java.base/java.lang.Thread.sleep(Thread.java:527)
at TestDisposerRace.giveGCAChance(TestDisposerRace.java:104)
at TestDisposerRace.generateOOME(TestDisposerRace.java:115)
at TestDisposerRace.run(TestDisposerRace.java:57)
at TestDisposerRace.main(TestDisposerRace.java:44)
giveGCAChance is called after the previous loop has been broken by OOME being thrown. All it does is sleep in a loop and so potentially allow GC a chance to catch up and clear memory before the test moves on. However, since the Loom integration the implementation of Thread.sleep has changed significantly and it can now allocate and so encounter a secondary OOME - hence the current failure.
The test needs more resilience against OOME in unexpected places.
- duplicates
-
JDK-8299419 Thread.sleep(millis) may throw OOME
- Resolved
- relates to
-
JDK-8299419 Thread.sleep(millis) may throw OOME
- Resolved