[lldb-dev] [PATCH] Avoid leaking file descriptors.

Piotr Rak piotr.rak at gmail.com
Tue Mar 25 18:25:18 PDT 2014


My feeling was when I looked at it bit, that it is related to some thing
I've seen about symbols earlier. In some cases we got some
strange symbol (address pointing to location where it fails set
breakpoint), but have not nailed it down to anything specific, and did not
have time to dig deeper yet. It looks like edge case, rather like something
totally broken though.

And indeed, this one doesn't look very nice.


2014-03-26 1:51 GMT+01:00 <jingham at apple.com>:

> Huh, that's failing doing "set a breakpoint and run to it".  The set a
> breakpoint part works, I've already checked that, but we haven't done
> anything but "LaunchSimple" and we fail at "did I hit this breakpoint".
>
> That's pretty bad, but I'm surprised you aren't seeing a similar failure
> from lots of tests...
>
> Jim
>
> On Mar 25, 2014, at 3:21 PM, Piotr Rak <piotr.rak at gmail.com> wrote:
>
> > +lldb-dev I am sorry
> >
> >
> > 2014-03-25 23:12 GMT+01:00 Piotr Rak <piotr.rak at gmail.com>:
> > UNSUPPORTED: LLDB (clang-i386) :: test_step_in_with_dsym_python
> (TestStepNoDebug.ReturnValueTestCase) (requires Darwin)
> > FAIL: LLDB (clang-i386) :: test_step_in_with_dwarf_python
> (TestStepNoDebug.ReturnValueTestCase)
> > Length: 5.
> > I got: 11.
> > UNSUPPORTED: LLDB (clang-i386) :: test_step_out_with_dsym_python
> (TestStepNoDebug.ReturnValueTestCase) (requires Darwin)
> > FAIL: LLDB (clang-i386) :: test_step_out_with_dwarf_python
> (TestStepNoDebug.ReturnValueTestCase)
> > Length: 5.
> > I got: 11.
> > UNSUPPORTED: LLDB (clang-i386) :: test_step_over_with_dsym_python
> (TestStepNoDebug.ReturnValueTestCase) (requires Darwin)
> > FAIL: LLDB (clang-i386) :: test_step_over_with_dwarf_python
> (TestStepNoDebug.ReturnValueTestCase)
> > Length: 5.
> > I got: 11.
> > ======================================================================
> > FAIL: test_step_in_with_dwarf_python
> (TestStepNoDebug.ReturnValueTestCase)
> >    Test stepping in using avoid-no-debug with dwarf.
> > ----------------------------------------------------------------------
> > Traceback (most recent call last):
> >   File "/home/prak/soft/llvm/tools/lldb/test/lldbtest.py", line 319, in
> wrapper
> >     return func(self, *args, **kwargs)
> >   File "/home/prak/soft/llvm/tools/lldb/test/lldbtest.py", line 370, in
> wrapper
> >     return func(self, *args, **kwargs)
> >   File
> "/home/prak/soft/llvm/tools/lldb/test/functionalities/step-avoids-no-debug/TestStepNoDebug.py",
> line 63, in test_step_in_with_dwarf_python
> >     self.get_to_starting_point()
> >   File
> "/home/prak/soft/llvm/tools/lldb/test/functionalities/step-avoids-no-debug/TestStepNoDebug.py",
> line 102, in get_to_starting_point
> >     self.assertTrue(len(threads) == 1, "Stopped at inner breakpoint.")
> > AssertionError: False is not True : Stopped at inner breakpoint.
> > Config=i386-clang
> > ======================================================================
> > FAIL: test_step_out_with_dwarf_python
> (TestStepNoDebug.ReturnValueTestCase)
> >    Test stepping out using avoid-no-debug with dsyms.
> > ----------------------------------------------------------------------
> > Traceback (most recent call last):
> >   File "/home/prak/soft/llvm/tools/lldb/test/lldbtest.py", line 319, in
> wrapper
> >     return func(self, *args, **kwargs)
> >   File "/home/prak/soft/llvm/tools/lldb/test/lldbtest.py", line 370, in
> wrapper
> >     return func(self, *args, **kwargs)
> >   File
> "/home/prak/soft/llvm/tools/lldb/test/functionalities/step-avoids-no-debug/TestStepNoDebug.py",
> line 29, in test_step_out_with_dwarf_python
> >     self.get_to_starting_point()
> >   File
> "/home/prak/soft/llvm/tools/lldb/test/functionalities/step-avoids-no-debug/TestStepNoDebug.py",
> line 102, in get_to_starting_point
> >     self.assertTrue(len(threads) == 1, "Stopped at inner breakpoint.")
> > AssertionError: False is not True : Stopped at inner breakpoint.
> > Config=i386-clang
> > ======================================================================
> > FAIL: test_step_over_with_dwarf_python
> (TestStepNoDebug.ReturnValueTestCase)
> >    Test stepping over using avoid-no-debug with dwarf.
> > ----------------------------------------------------------------------
> > Traceback (most recent call last):
> >   File "/home/prak/soft/llvm/tools/lldb/test/lldbtest.py", line 319, in
> wrapper
> >     return func(self, *args, **kwargs)
> >   File "/home/prak/soft/llvm/tools/lldb/test/lldbtest.py", line 370, in
> wrapper
> >     return func(self, *args, **kwargs)
> >   File
> "/home/prak/soft/llvm/tools/lldb/test/functionalities/step-avoids-no-debug/TestStepNoDebug.py",
> line 46, in test_step_over_with_dwarf_python
> >     self.get_to_starting_point()
> >   File
> "/home/prak/soft/llvm/tools/lldb/test/functionalities/step-avoids-no-debug/TestStepNoDebug.py",
> line 102, in get_to_starting_point
> >     self.assertTrue(len(threads) == 1, "Stopped at inner breakpoint.")
> > AssertionError: False is not True : Stopped at inner breakpoint.
> > Config=i386-clang
> > ----------------------------------------------------------------------
> > Ran 6 tests in 0.675s
> >
> >
> >
> > 2014-03-25 22:12 GMT+01:00 <jingham at apple.com>:
> >
> > It seems weird to me that TestStepNoDebug.py should be failing
> intermittently on Linux.  It is a pretty straight-forward single-threaded
> test, so if it is going to fail, I would expect it to fail altogether.
> >
> > What does it look like when it fails?
> >
> > Jim
> >
> > On Mar 25, 2014, at 2:01 PM, Todd Fiala <tfiala at google.com> wrote:
> >
> > > btw on this one:
> > > > TestStepNoDebug.py
> > >
> > > I disabled the tests in that file that were failing intermittently for
> me, early today.
> > >
> > >
> >
> > _______________________________________________
> > lldb-dev mailing list
> > lldb-dev at cs.uiuc.edu
> > http://lists.cs.uiuc.edu/mailman/listinfo/lldb-dev
> >
> >
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/lldb-dev/attachments/20140326/494f16f8/attachment.html>


More information about the lldb-dev mailing list