On Mon, 1 Nov 2021 19:22:02 GMT, Sergey Bylokhov <s...@openjdk.org> wrote:
> From the stack traces in the bug report, it seems that the problem was in > some kind of deadlock in the SunToolkit.waitForIdle I was not able to reproduce the deadlock scenario - and i ran this test about 3 hours non-stop which resulted in more than 2000 invocations. I had couple of false positives where dump was expected but did not happened due to the mismatch between the windows that receiving key pressed and key released events and that is what i tried to fix to make test more stable. But neither locally nor on remote test machines i was not able to reproduce it exactly the way it reported. Giving the extensive use of Robot.waitForIdle() in other tests without any issue my only approach is to make this test more stable and enable it for execution and then see if it will timeout again. I doubt so. ------------- PR: https://git.openjdk.java.net/jdk/pull/6194
