Bug 3240 - TestBreakpoints failiure on x86 FC5
Summary: TestBreakpoints failiure on x86 FC5
Status: RESOLVED FIXED
Alias: None
Product: frysk
Classification: Unclassified
Component: general (show other bugs)
Version: unspecified
: P2 normal
Target Milestone: ---
Assignee: Mark Wielaard
URL:
Keywords:
Depends on:
Blocks: 1522 2104
  Show dependency tree / graph
 
Reported: 2006-09-21 14:46 UTC by Mike Cvet
Modified: 2007-01-15 18:01 UTC (History)
0 users

See Also:
Host:
Target:
Build:
Last reconfirmed:

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Mike Cvet 2006-09-21 14:46:45 UTC 
I must have run 'make check' 50 times, testing something else before this popped up:

Running testInsertRemove(frysk.proc.TestBreakpoints) ...Exception in thread
"Thread-25" java.lang.NullPointerException
   at frysk.proc.Breakpoint.set(TestRunner)
   at frysk.proc.Breakpoint.install(TestRunner)
   at frysk.proc.TaskState$BlockedSignal.handleAddCodeObserver(TestRunner)
   at frysk.proc.Task.handleAddCodeObserver(TestRunner)
   at frysk.proc.TaskCodeObservation.handleAdd(TestRunner)
   at frysk.proc.ProcState$3.handleAddObservation(TestRunner)
   at frysk.proc.Proc.handleAddObservation(TestRunner)
   at frysk.proc.Proc$10.execute(TestRunner)
   at frysk.event.EventLoop.runEventLoop(TestRunner)
   at frysk.event.EventLoop.run(TestRunner)
   at frysk.proc.TestBreakpoints$EventLoopRunner.run(TestRunner)
Comment 1 Mark Wielaard 2006-09-21 15:28:19 UTC 
Confirmed on x86_64 also.
After lots of runs (./TestRunner -r 1000 frysk.proc.TestBreakpoints)
Comment 2 Mark Wielaard 2006-10-11 14:03:19 UTC 
The behaviour has changed. Now testInsertRemove hangs, but doesn't show the
exception. This seems to be some bad interaction with an execedEvent. But it is
hard to trigger with -c FINE logs.
Comment 3 Mark Wielaard 2007-01-15 18:01:52 UTC 
We weren't properly waiting till the code observer was installed and so an
ExecedEvent could happen before the observer was properly installed. Fixed by:

2007-01-15  Mark Wielaard  <mark@klomp.org>

        Fixes bug #3240
        * TestBreakPoints.java (testInsertRemove): Make sure the observer
        is properly installed.

Now ./TestRunner -r 1000 frysk.proc.TestBreakpoints still fails as follows:

Running testHitAndRun(frysk.proc.TestBreakpoints) ...PASS
Running testInsertRemove(frysk.proc.TestBreakpoints) ...PASS
Running testAddLots(frysk.proc.TestBreakpoints) ...ERROR
  pipe (24)

Time: 0.428
There was 1 error:
1) testAddLots(frysk.proc.TestBreakpoints)pipe (24)
   at frysk.testbed.ForkTestLib.throwErrno(TestRunner)
   at frysk.testbed.ForkTestLib.fork(TestRunner)
   at frysk.proc.TestBreakpoints.setUp(TestRunner)
   at frysk.junit.Runner.runCases(TestRunner)
   at frysk.junit.Runner.runArchCases(TestRunner)
   at frysk.junit.Runner.runTestCases(TestRunner)
   at TestRunner.main(TestRunner)

FAILURES!!!
Tests run: 3,  Failures: 0,  Errors: 1

Failed after run #168

But this is always after #168 runs (on either a single processor x86 and a
multi-core x86_64. And I believe this is actually bug #3863 (libunwind leaking).