Dear P.O.:
On 10.08.2017 23:14, oor...@jonases.se wrote:
> Dear Rony et all,
>
> I continue to run & crash/stall the BSF demo program saving the crash log as
> I go along. I will
> save and upload new ones as I go along unless told otherwise. I guess it will
> be more of the same
> most of the time, hopefully you can quickly ignore those.
>
> It seems that most of the time it is the main thread, the dispatcher that
> gets crashed by java
> (running, as it seems ooRexx code)
It *always* crashes in ooRexx after an AttachThread() to the Java-GUI-thread (a
non-ooRexx-thread).
> Is the information below helpful at all or should I stop logging? Is there
> anything more/else you
> want me to do? I have more information available locally but it is to big for
> uploading, I can
> search for specific names of routines if you want me to.
For me it is already sufficient. Your MacOSX debug info is much more detailed
than the
platform-independent hs_error*log (which is great in its own right).
When trying to find the cause of these crashes, unfortunately, my current
knowledge of the ooRexx
kernel and applying the debugging environments is much too scarce. (Trying to
peek around the call
stacks and inspecting the different records for literally months has not
surfaced anything
meaningful, short of knowing the exact inner workings of the ooRexx kernel and
the assumptions that
play a role there.)
After months of trying to simplify the test cases that are able to cause the
crashes (without Java
and BSF4ooRexx in the picture) I came up with a few ooRexx-only crashes that I
have reported and
which code got uploaded to the ooRexx bug database. These programs are
artificial, but their
complexity was augmented according to the code-paths the BSF4ooRexx
applications employ. You can
find them in the ooRexx bug-database with this link:
<https://sourceforge.net/p/oorexx/bugs/search/?q=crash>.
So these are the reasons why I have been hoping that those ooRexx developers
who are acquainted and
experienced with the ooRexx kernel may become able to debug this, if supplying
test cases that
exhibit the crashes. This has worked out e.g. for
<https://sourceforge.net/p/oorexx/bugs/1445/>. So
therefore this call for help for debugging this crash.
---
The actual test application
(<https://www.dropbox.com/sh/xjnzwm6r0qnugzc/AACN67fJc4CZ2NOYxNaSiewFa?dl=0>)
that uses BSF4ooRexx
has a few benefits:
* it crashes the same on all operating systems and with all versions of
ooRexx 5.0beta (32 or 64 bit),
* the crashes occur far earlier than with the above mentione ooRexx-/C++ only
test applications,
so hopefully can be easier debugged in the ooRexx kernel with the
appropriate knowledge that
would allow one to set the break-points at the right locations in the
ooRexx kernel,
* it only crashes after ooRexx attached to the Java-GUI-thread, which is a
thread that is
guaranteed to not have been created by ooRexx.
Best regards,
---rony
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Oorexx-devel mailing list
Oorexx-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/oorexx-devel
