Uploaded image for project: 'JDK'
  1. JDK
  2. JDK-4882815

Regression: w/JVMTI-libjdwp backend: nsk/jdi/BScenarios/singlethrd/tc01x00[12]

    Details

    • Subcomponent:
    • Resolved In Build:
      tiger
    • CPU:
      generic
    • OS:
      generic
    • Verification:
      Verified

      Description

      This JVMTI-based backend has not been putback yet, this bug is for tracking this
      regression of jdi behavior.

       nsk/jdi/BScenarios/singlethrd/tc01x002 and
       nsk/jdi/BScenarios/singlethrd/tc01x001
                    Step into a line that contains a 'new' object creation is expected
                     to land in the new class constructor, but it seems to
                     come back to the same line, requiring an extra step.
                     It appears the class isn't "prepared" in my case, and a
                     step complete might be happening in the class
                     loader before it gets to the constructor.
                     If an earlier 'new' is done just to prime the class
                     into being fully 'prepared/initialized', the step into
                     the next 'new' works fine and the test passes for me.


      Test failure will look something like this:

      # event ===>>> ###@###.###01x001a:51 in thread main
      #event info:
      # thread - main
      # source - tc01x001a.java
      # method - performTest
      # line - 51
      ## ERROR: debugger FAILURE> StepEvent steps to line 51, expected line number is 56
      #

      Where line 51 is at the 'new', and line 56 is the constructor code that hasn't
      been reached yet.

          51 new tc01x001aClass1();
          52 }
          53 }
          54
          55 class tc01x001aClass1 {
          56 tc01x001aClass1() {
          57 tc01x001a.log.display("tc01x001aClass1::constructor is called");
          58 }
          59 }

      ###@###.### 2003-06-23

        Attachments

          Issue Links

            Activity

              People

              • Assignee:
                ohair Kelly Ohair (Inactive)
                Reporter:
                ohair Kelly Ohair (Inactive)
              • Votes:
                0 Vote for this issue
                Watchers:
                0 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved:
                  Imported:
                  Indexed: