On Wed, 21 Apr 2021 19:43:40 GMT, Sergey Bylokhov <s...@openjdk.org> wrote:
> The Robot.waitForIdle() is implemented via SunToolkit.realSync(), I am not > sure how it might solve the problem. It will not solve the problem as there is no problem as such. Test failed twice in two years on the same machine with the problematic OS configuration. Test has not failed before for 6 years and since configuration was fixed has not resurfaced. I'm just using this opportunity to get rid of the non-public API usage in a public test. ------------- PR: https://git.openjdk.java.net/jdk/pull/3606
