I've just seen 6 different tests each fail twice on a macos 12.6.1 x64 system where they fail because they get an unexpected screen capture colour
because the screen capture actually samples the pixels affected by the mouse pointer which should not be happening.
We've seen this before and I need to track down the earlier bug report but
the tests failing because of this are these

java/awt/Robot/HiDPIScreenCapture/ScreenCaptureTest.java
java/awt/Robot/CheckCommonColors/CheckCommonColors.java
java/awt/Paint/PaintNativeOnUpdate.java
javax/swing/text/DefaultCaret/HidingSelection/HidingSelectionTest.java
javax/swing/plaf/synth/SynthButtonUI/6276188/bug6276188.java
javax/swing/JTable/8236907/LastVisibleRow.java

This might be a macOS bug, or a system config issue, but I'd like to capture it here in JBS anyway.

Aha - the previous bug was this https://bugs.openjdk.org/browse/JDK-8252813

And it was the exact same lab test system as is causing the problem today!
But mostly different tests than before - I think we updated the previously failing tests.

I'll fill in more details later.