RFR: 8271855: [TESTBUG] Wrong weakCompareAndSet assumption in UnsafeIntrinsicsTest

[Martin Doerr]

On Wed, 4 Aug 2021 12:38:04 GMT, Martin Doerr <mdoerr at openjdk.org> wrote:

> We need a fix for test failures due to wrong assumption. (See JBS issue for details.) I suggest to give weakCompareAndSet several chances.

The test should usually pass within 2 iterations. It could happen that 2 different reasons make it fail (e.g. GC touches the cache line, OS takes the process from the CPU, interrupt occurs, ...). So, I think 2 iterations should usually work, 3 ones should be enough to prevent very sporadic failures in the future.

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

PR: https://git.openjdk.java.net/jdk/pull/4992