On Fri, 7 Apr 2023 17:57:41 GMT, Harshitha Onkar <[email protected]> wrote:

>> This test issue was due to race condition that occured when using 
>> `-Dsun.java2d.uiScale >= 2`.
>> As a result it led to incorrect mouse position being compared as opposed to 
>> the updated mouse position after the final mouse move event.
>> 
>> Following is the event log before and after fix.
>> 
>> **Before fix:**
>> 
>> 1st mouse move
>> 2nd mouse move
>> Frame-1 Mouse Event
>> Frame-2 Mouse Event
>> java.awt.Point[x=29,y=29]
>> 
>> 
>> **After fix:**
>> 
>> 1st mouse move
>> 2nd mouse move
>> Frame-1 Mouse Event
>> 3rd mouse move
>> Frame-2 Mouse Event
>> java.awt.Point[x=50,y=50]
>> 
>> 
>> Earlier Frame-2's mouseMoved() was being triggered on Robot's 2nd mouse move 
>> instead of 3rd mouse move. To fix it, the mouseMotionListener for Frame-2 is 
>> now added after Robot's 2nd mouse move is processed to avoid the race 
>> condition.
>> 
>> The updated test fix is checked on multiple uiScales including fractional 
>> scales for windows platform.
>> CI testing works as expected on all platforms.
>
> Harshitha Onkar has updated the pull request incrementally with one 
> additional commit since the last revision:
> 
>   added syncLocationToWindowManager(), changed literals to vars

@prrace Please review

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

PR Comment: https://git.openjdk.org/jdk/pull/13380#issuecomment-1508926025

Reply via email to