On Wed, 20 Apr 2022 04:13:09 GMT, Phil Race <p...@openjdk.org> wrote:
> As discussed in https://bugs.openjdk.java.net/browse/JDK-8285094 it seems > that the test > java/awt/Frame/GetGraphicsStressTest/GetGraphicsStressTest.java destabilises > the Xserver. > This causes java/awt/Frame/InvisibleOwner/InvisibleOwner.java to fail and > even before that > other tests which do pass are not visibly behaving as expected. > > I didn't find any Xserver logs of "bad stuff" happening so it just seems like > the Xserver was > having trouble keeping up and JDK was behaving correctly as it could despite > the Xserver sending > lots of requests to repaint. > > Just the "sleep" at the end of GetGraphicsStressTest.java seems to be enough > but I'd already > reworked InvisibleOwner.java and I'd like to think it is a bit better than > before. test/jdk/java/awt/Frame/InvisibleOwner/InvisibleOwner.java line 35: > 33: import java.awt.EventQueue; > 34: import java.awt.Frame; > 35: import java.awt.Graphics; Unused import test/jdk/java/awt/Frame/InvisibleOwner/InvisibleOwner.java line 135: > 133: ownedWindowBounds.y + > ownedWindowBounds.height / 2); > 134: robot.mousePress(InputEvent.BUTTON1_MASK); > 135: robot.mouseRelease(InputEvent.BUTTON1_MASK); Nit: Shouldn't we use BUTTON1_DOWN_MASK instead of deprecated BUTTON1_MASK? ------------- PR: https://git.openjdk.java.net/jdk/pull/8312
